Posters are listed alphabetically in order of the first presenter. Posters will be presented in two sessions on Saturday and Sunday afternoon immediately after the concurrent sessions at Humboldt-Universität zu Berlin. All presenters are invited to pin-up their posters on the Saturday morning/lunchtime of the meeting. They are then requested to stand by their posters at the time indicated (either Saturday or Sunday afternoon); though they are welcome to present their posters in both sessions if they so wish. Posters must be removed at the end of the poster session Sunday evening.

Perception of Sharpness is Illusory in Peripheral Vision

Felipe Aedo Jury, LNFP- CNRS
Delphine Pins, LNFP- CNRS UMR8160

Saturday, June 6th, 17:30–19:30: Poster No. 25

Visual performance decreases with eccentricity. However, our conscious perception is homogeneous across the visual field and we are unaware of the decrease in our visual sensitivity. An explanation for this phenomenon is that the brain fills in the missing information, probably through iconic memory from saccades, giving rise to the impression of homogeneity across the visual field. As a consequence of this fill-in process misperceptions of sharpness at large eccentricities might occur. Here we test whether perception is veridical or illusory at different eccentricities. Low-pass filtered, 10 cycles per degree (cpd), pictures of faces and buildings were presented to subjects in central vision for 100 ms. After a delay (1000–1500 ms.) a second stimuli was presented for 100 ms at variable eccentricities (00–480), whose sharpness was varied through different low-pass filters (3 and 30 cpd). The subject's task was to judge whether the second picture was ''sharper'' or ''more blurred'' than the first one. The results showed that for pictures of faces as well as for buildings, subjects rate them as sharper than they actually were. This misperception of sharpness increases as eccentricity increase. Our results then provide further evidence that sense of sharpness in peripheral vision stems from an illusion. To test whether the duration of the second stimuli affects the perception of sharpness we systematically varied its duration (13-300ms). Increasing the duration of the second stimuli leads to a more veridical perception, but this effect interacts with eccentricity. Thus, larger eccentricities require longer stimulus duration to lead to veridical perception. To test whether attention modulates this phenomenon we manipulate spatial attention. Validly cued trials led to veridical perception whereas invalidly cued trials led to the same misperception of sharpness previously found. Perception of sharpness in peripheral vision is illusory when time pressures are set on the visual system, but they can be overcome when stimulus duration is increased or attention is spatially directed.

How Do We Learn to See the World? On the Development of Conscious Perception

Thorsten Albrecht, Georg-August University Göttingen
Susan Klapötke, University of Göttingen
Uwe Mattler, University of Göttingen

Saturday, June 6th, 17:30–19:30: Poster No. 15

Backward masking is a technique that is often used in visual science to prevent participants from becoming aware of stimuli. Individual differences in masking effects are often deemed undesirable complications but we think that they might be used to study perceptual learning and the development of conscious perception. We report the spontaneous development of stable individual differences in a meta-contrast masking paradigm. Participants performed a discrimination task on targets that were followed by a mask after different stimulus onset asynchronies (SOA). After two days of training two groups of participants could be distinguished: In one group performance increased with increasing SOA, in the other group performance decreased with increasing SOA. Although the difference between groups was negligible at the beginning of the experiment it became successively clear over the course of training, and remained unchanged after several weeks without training. Findings suggest that participants differ in directing selective attention to one of two aspects of the stimulus display to extract information required to do the task. We think this top-down guidance determines the level in the physiological hierarchy of visual operations where perceptual learning takes place, which might lead to a development of stimulus awareness.

What Matters for Perceptual Consciousness? A (Non) Empirical Question

Saray Ayala, University of British Columbia

Sunday, June 7th, 17:30–19:30: Poster No. 46

Some philosophers endorse the traditional view that nothing outside the brain is part of the minimal substrate of perceptual experience (Block, 2005). Others philosophers claim that perceptual experience depends on outside-the-brain factors (Noë, 2004, 2007). There are two main ways in which this dependency between non-neural processes and experience can be understood. Only one of them is actually an (extended) alternative to the traditional view. According to the radical reading, the specific fine-grained details ofn organism«s body and the characteristic ways in which she interacts with objects and properties, are said to be a constraint on perceptual experiences. It follows from this that perceptual experiences of differently embodied organisms with different sensorimotor routines cannot be identical to our own. According to a moderate reading, however, the fine-grained details of an organism’s body are not a constraint on perceptual experience. Perceptual experience ultimately depends on representations and computational processes that are insensitive to details of implementation. Thus, differently embodied organisms could in principle have the same perceptual experience, as long as they have access to the same gross information and then can form the same internal representations. It is an open empirical question whether every difference in embodiment makes a difference to the content and character of any conscious perceptual experience that ensues. According to the moderate view, the defender of the radical (extended) alternative is not offering any evidence to decide the empirical question. The radical conclusion (that differently embodied organisms necessarily inhabit different perceptual worlds) is not proven. My goal here is to show that there is no possibility of deciding this debate empirically. As the debate is introduced in Clark (2008), it seems that what he calls an open empirical question is not a possible empirical goal for the radical defender. With the aid of an empirical illustration (the case of tactile-visual sensory substitution, Bach y Rita & Kercel, 2003), I will conclude that the allegedly empirical question is, in fact, not empirical at all: no empirical evidence could possibly refute the moderate reading because she already assumes that experience is multiply realizable.

Implicit Learning of Likes and Dislikes

Robert Balas, Warsaw School of Social Sciences and Humanities
Joanna Sweklej, Warsaw School of Social Sciences and Humanities

Saturday, June 6th, 17:30–19:30: Poster No. 113

Evaluative conditioning (EC) is a process of changing the evaluation of initially neutral stimulus (conditioned stimulus - CS) due to its repeated pairing with either positive or negative stimulus (unconditioned stimulus - US). As such it is regarded as a basic learning process that results in forming attitudes and preferences. Conscious awareness impact on EC is now a question of considerable debate. Specifically, there is no consensus whether EC requires subject’s awareness of CS-US contingencies as well as awareness of stimuli themleves. The presented research examines whether EC is possible without perceptal and contingency awareness. During conditioning phase CS-US pairings were presented to subjects with either CS (Experiment 1) or US (Experiment 2) presented subliminally to check whether perceptual awareness facilitates EC. EC effect was present in both direct and indirect evaluations of conditioned stimuli. In Experiment 3 both CS and US were presented subliminally. Again, the EC effect was detected both in direct and indirect evaluations. Those results indicated that perceptual awareness is not a necessary condition to acquire evaluative responses. In Experiment 4 attentional resources were manipulated by applying secondary task in conditioning phase. The data showed intact EC suggesting that the process is fairly independent of conscious attention. In conclusion, we argue that evaluative conditioning is a form of implicit learning that does not require neither conscious awareness of what is being conditioned nor attentional resources devoted to change human preferences and evalauations.

Representationalism and the Structure of Conscious Experience

Magdalena Balcerak Jackson, University of Cologne

Saturday, June 6th, 17:30–19:30: Poster No. 31

In this paper I present a novel argument in favour of representationalism. Representationalism is the thesis that the phenomenal character of all mental states is determined by, or supervenes on their representational properties. The debate about representationalism is important because once we show that representationalism is true, we have very good reasons to believe that the specific way we consciously experience our various mental states can be explained by an adequate theory of representational content. My argument starts from the observation that our conscious mental states are epistemically significant for us in virtue of having a specific phenomenal character. More specifically, any two phenomenally different mental states have different epistemic significance. To argue for this core idea I make use of two sorts of facts: 1. facts about the complex evaluative and comparative abilities that we, subjects of experience, have with respect to the phenomenal character of our own mental states, and 2. facts about the structure that our conscious experience exhibits. This enables me to make more precise what it means for a mental state to be epistemically significant in virtue of having a specific phenomenal character, and to show subsequently that differences in epistemic significance of this kind just amount to differences in representational content of a specific form. As a result, the paper not only establishes the representationalist thesis, but also shows which kind of theory of content can explain the specific phenomenal character of our various conscious mental states.

What Unconscious Pain Tells Us About Consciousness

Fiorella Battaglia, University of Pisa

Sunday, June 7th, 17:30–19:30: Poster No. 48

Nociceptive input and algic experience are relatively independent phenomena and together with unconscious pain belong to the phenomenology of pain. To demonstrate it, it is necessary to integrate different perspectives as phenomenological theories on subjective experience of pain, analytical remarks about epistemology of pain, and research findings on neuronal correlates of conscious and unconscious pain. It is important to clarify whether those approaches may shed further light on the nature and function of the consciousness. Research on pain stresses the value of subjective experience. On the other hand, objective damage to tissues is the starting point for an empiric approach to pain. Furthermore, an important element of pain is considered its verbalization. The subjective experience of pain and its verbalization are separable phenomena. If we identify conscious pain with ''what it is like to be'' (Nagel 1975) in that state, what answer is possible to give to the question: What is pain without its subjective experience? Clinical and neuropsychological studies have yielded to the surprising result of the existence of unconscious pain. ''An injured soldier and athletes can deny that they are in pain without qualifications or acknowledgment that they might be apprised of considerations that would lead them to change their mind'' (Hill 2006). Thus we need other parameters to define the algic experience. We must take into consideration the neuronal bases of pain. The somatosensory brain areas S1 and S2, the posterior parietal cortex, the insula and the anterior cingulate cortex constitute the ''pain matrix'' (Melzack 1999) and can influence phenomena of selection, integration and evaluation on which consciousness relies. In the first part of this paper, I wish to describe a model of phenomenal consciousness, by referring to previous models of phenomenal consciousness. In the second part, I wish to redefine this model, by showing that the concept of ''unconscious pain'', far from questioning the subjective character of the algic experience, is useful for deciphering the factors that influence the algic experience, which that are at the basis of our ability to fell it.

Neural Signature of the Conscious Processing of Auditory Regularities

Tristan Bekinschtein, INSERM, Cognitive Neuro-imaging Unit, France.
Stan Dehaene, Neurospin Center, CEA, Gif sur Yvette, France
Benjamin Rohaut, Cognitive Imaging Unit, Salpe Paris, France
Francois Tadel, Neurospin Center, CEA, Gif sur Yvette, France
Laurent Cohen, Cognitive Imaging Unit, Salpe Paris, France
Lionel Naccache, Cognitive Imaging Unit, Salpe Paris, France

Sunday, June 7th, 17:30–19:30: Poster No. 136

Can conscious processing be inferred from neurophysiological measurements? Some models stipulate that the active maintenance of perceptual representations across time requires consciousness. Capitalizing on this assumption, we designed an auditory paradigm that evaluates cerebral responses to violations of temporal regularities that are either local in time or global across several seconds. Local violations led to an early response in auditory cortex, independent of attention or the presence of a concurrent visual task, whereas global violations led to a late and spatially distributed response that was only presentwhen subjects were attentive and aware of the violations. We could detect the global effect in individual subjects using functional MRI and both scalp and intracerebral event-related potentials. Recordings from 8 noncommunicating patients with disorders of consciousness conÞrmed that only conscious individuals presented a global effect. Taken together these observations suggest that the presence of the global effect is a signature of conscious processing, although it can be absent in conscious subjects who are not aware of the global auditory regularities. This simple electrophysiological marker could thus serve as a useful clinical tool.

Neurocognition and Temperament in Depersonalization Disorder

Heather Berlin, Mount Sinai School of Medicine
Holly Hamilton, Mount Sinai School of Medicine
Daphne Simeon, Mount Sinai School of Medicine

Sunday, June 7th, 17:30–19:30: Poster No. 90

Background: Dissociation is a disruption in the usually integrated functions of consciousness, memory, identity, or perception. Depersonalization Disorder (DPD) is a dissociative disorder characterized by a persistent or recurrent feeling of being detached from one’s mental processes or body, accompanied by intact reality testing. Their alterations in consciousness center around a sense of unfamiliarity/unreality and hypo-emotionality. Objective: Explore the neurocognitive and temperament profile of DPD patients to better understand their underlying neurobiology and to gain insights into the neural basis of dissociation. Methods: 19 DPD patients and 22 matched healthy controls (HCs) were given a comprehensive neuropsychological battery (CANTAB), the Iowa Gambling Task, a time perception task, and questionnaires of impulsivity, temperament, emotion, and frontal behavior (measures orbitofrontal cortex (OFC) dysfunction). Results: Compared to HCs, DPD patients performed significantly better on the Intra-Extra Dimensional Set Shift task (IED) (an analogue of the Wisconsin Card Sorting test) and did no worse on any other cognitive task. However, they did have a significantly faster subjective sense of time (overestimated and underproduced 10–90 seconds time intervals), and their long-term time estimation positively correlated with their attentional impulsivity. DPD patients experienced more childhood trauma (physical and emotional abuse and neglect), negative emotions, and dissociation, and were more impulsive, neurotic, and harm avoidant and less extraverted, agreeable, conscientiousness, and self-directed. DPD patients also had more frontal behaviors which positively correlated with their emotionality, neuroticism, and childhood trauma. Conclusions: Dorsolateral prefrontal cortex (DLPFC) hyperactivation may explain DPD patients' enhanced IED performance, a task sensitive to DLPFC function. Memory suppression and emotional regulation has been associated with increased DLPFC and reduced limbic activation. DLFC inhibition of limbic structures may mediate DPD patients’ ability to dissociate and their hypo-emotionality. However, OFC dysfunction (measured by frontal behaviors) appears to be related to their other problems, such as their time perception deficits, negative emotions, neuroticism, impulsivity, and childhood trauma. Further investigation is needed to determine how this “splitting of consciousness” relates to the NCC. Integration of various cortical and subcortical areas may be necessary for cohesive conscious experience. Dissociation may involve disruption of cortico-cortical, thalamo-cortical, or limbic-cortical connectivity.

Function and Dysfunction of Phenomenal States: Exemplified in Affective Disorders

Felix Bermpohl, Berlin School of Mind and Brain

Saturday, June 6th, 17:30–19:30: Poster No. 127

Drawing on current theories of consciousness, a model will be proposed that focuses on phenomenal states. According to the model, phenomenal states result from integrating multimodal internal and external inputs. This integration process is characterized by selection and appraisal with regard to self-relevance and provides a coherent percept even in complex situations. This percept provides the basis for further cognitive processes and advantageous responses. The model will be confronted with clinical-phenomenological, neuropsychological and neuroimaging findings in patients with affective disorders. It is suggested that these patients show a fundamental alteration in the generation of phenomenal states (in the sense of the proposed model). This alteration is characterized by a mood-congruent bias in the integration process and may provide the basis for various emotional, cognitive and behavioral symptoms and deficits. - These considerations (1) have heuristic value for understanding affective disorders and (2) support models of consciousness that suggest a functional role of phenomenal states.

Can We Blink Without Masking? Evidence from the Spatial Attentional Blink

Vincent Berthet, Laboratoire de Sciences Cognitives et Psychology
Sid Kouider, Laboratoire de Sciences Cognitives et Psychol

Saturday, June 6th, 17:30–19:30: Poster No. 7

The Attentional Blink (AB) is a well-known RSVP paradigm in which two visual targets (T1 and T2) are embedded in a stream a distractors. In this paradigm, performance on T2 is largely impaired when it appears briefly after T1 (i.e. within 200–500ms). This paradigm is thought to reveal the time course of attention (Raymond, Shapiro, & Arnell, 1992). An important feature of the AB paradigm concerns the necessity of a light masking of the two targets (e.g. Brehaut, Enns, & Di Lollo, 1999). Thus, although the classical interpretation of the AB effect refers to the limited capacity of attentional resources, this interpretation is not straightforward since masking T2 also contributes to its impaired visibility at short lags. Here, by contrast, we demonstrate that the attentional blink can occur without masking. Using a spatial variant of the paradigm in which stimuli appeared without being followed by masks but instead at different locations on the screen, we report a standard attentional blink effect. This effect was more pronounced at the shortest lag and positively related to the spatial distance between T1 and T2. These results support a clear capacity limited account of the AB effect without any consideration on masking and call for more consideration of the role of temporal attention in theories of the AB.

Different Effects of Within- and Across-Experiment Variation of Auditory and Visual Stimulus Intensity

Lars Boenke, Leibniz Institute for Neurobiology
Felix Ball, Leibniz Institute for Neurobiology
Anna Fiedler, Leibniz Institute for Neurobiology
Matthias Deliano, Leibniz Institute for Neurobiology
Frank Ohl, Leibniz Institute for Neurobiology

Sunday, June 7th, 17:30–19:30: Poster No. 86

It is known that variation of physical stimulus parameters within experiments and across experiments can have different effects on both neuronal processing and conscious perception (Boenke et al. 2009). In multisensory temporal-order perception, it is well established that the perceived temporal order depends on the physical intensities of the stimuli involved. For example, Neumann et al. (1992) demonstrated effects of both variation of the auditory stimulus intensity and variation of the visual intensity on the point of subjective simultaneity (PSS) by combining 3 light intensities and 3 tone intensities. However, Roufs (1963), when performing a series of experiments in each of which the auditory intensity was held constant and only the visual intensity was varied, and then compared across this series of experiments, found only an effect of the light intensity but not of the auditory intensity on the PSS. To test whether the effect of intensity variation of a given sensory modality in a multisensory stimulus on the conscious perception of temporal order depends on whether it is held constant or is independently varied in an experiment, we have conducted 5 experiments. In experiment 1, using two sound intensities (Alow, Ahigh) and two light intensities (Vlow, Vhigh), we could replicate the findings of Neumann et al. (1992). In experiments 2 (Alow, Vlow, Vhigh) and 3 (Ahigh, Vlow, Vhigh) we only varied the light intensity and held the sound intensity constant. In experiments 4 and 5 we set the light intensity to a constant low or high value, respectively, and only varied auditory intensities. In experiments 2 and 3 we found a similar result as Roufs (1963), namely that varying the sound intensities across experiments did not affect the PSS. In experiments 4 and 5 we found that the variation of light intensities across experiments was similar to variation within an experiment (cf. experiment 1). In conclusion, these data imply a difference in the ability of the auditory and visual sensory system to compensate intensity variation of their adequate modality in an audiovisual compound stimulus.

Differential Encoding of Mechanisms for Human Decision Making

Carsten Bogler, Max Planck Institute
Stefan Bode, MPI Leipzig Attention and Awareness Group
Carsten Bogler, Max Planck Institute
Chun Siong Soon, MPI for Human Cognitive and Brain Sciences
John-Dylan Haynes, BCCN, Charitè - Universitätsmedizin Berlin

Sunday, June 7th, 17:30–19:30: Poster No. 70

Object identity is known to be encoded in distributed activity patterns in human lateral occipital cortex (LOC). Sufficient visual input to LOC is important for guiding perceptual decision making (Williams et al., 2007). Consistent with this, we found that decisions about object categories could be predicted from precuneus, but not LOC, if stimuli were rendered nearly invisible (Bode & Haynes, in prep.). Precuneus as well as anterior medial prefrontal cortex also encoded free decisions even in the abscence of visual tasks (Soon et al., 2008). In the present study, we aimed to disentangle the contribution of decision-related brain regions for different types of decisions.We directly compared (a) decisions in perceptual decision making using highly visible objects, (b) perceptual decisions made with insufficient visual input and (c) free decisions for objects. Multivariate pattern classification was used to predict the decision outcomes, thereby allowing us to compare the underlying patterns of the neural resonses (Haynes & Rees, 2006). We found perceptual decisions about highly visible objects encoded in bilateral visual cortex and LOC. Perceptual decisions with insufficient visual input, however, were encoded in the precuneus only. Free decisions could be decoded from medial prefrontal cortex. Interestingly, precuneus also showed a tendency to encode free decisions, which was confirmed by cross-condition pattern classification. Our results support the assumtion that different networks are involved in different kinds of object related decisions. By removing visual information, subjects were forced to make pure guesses, in which case the precuneus encoded the decision outcome rather than visual areas. The precuneus was also a key region for free, internal decisions, supporting the notion that guessing could be regarded as an internal decision here. Anterior medial prefrontal cortex, however, was uniquely involved when decisions were intended to be internal and might therefore be exclusively important for free decisions made without an external frame of reference.

Bistable Perception: A Twofold Dependence on History

Jochen Braun, Cognitive Biology, Otto-von-Guericke University
Alexander Pastukhov, Otto-von-Guericke Universität

Sunday, June 7th, 17:30–19:30: Poster No. 4

Bistable perception is one of the most striking examples for fluctuations of visual awareness. Theoretical analyses of its dynamics offer unexpected insights into the nature of the underlying neural representation. For instance, bistable perception seems poised at a particular point in state space, namely, the very boundary between oscillatory and bistable regimes (see Pastukhov et al., ASSC 09). Moreover, reversals of perceptual dominance seem to reflect stochastic integration over multiple populations that are independently multi-stable (see Gigante et al., ASSC09). These insights are founded on the empirical observation of a twofold dependence on history: in the short run, the dominance of a percept acts to destabilize this percept (and to stabilize the alternative percept) whereas, in the long run, the dominance of a percept acts to stabilize this percept (and to destabilize its alternative). Here we demonstrate both effects simultaneously in one experiment. Observers viewed a bistable display (kinetic-depth effect) and reported perceptual experience continuously. The display was interrupted either immediately after the first reported reversal or immediately after the second reported reversal. After a blank period of 8 seconds, the experiment was resumed. To evaluate the results, we computed a cumulative history for each percept, by convolving all preceding dominance intervals with an exponential decay. Carrying out this computation with different decay constants, we obtained either short histories or long histories for each point in time. We find that the duration of first percepts (after the blank) correlates positively with their long history. The maximal correlation is obtained for a build-up constant of 0.2-0.4 Tdom and a decay constant of 20–40 Tdom (where Tdom is the mean dominance period). Simultaneously, the duration of second percepts correlates negatively with their short history. In this case, the maximal correlation is obtained with a decay constant of 0.4-0.7 Tdom during stimulation intervals and a decay constant of 2-10 Tdom during blank intervals (hope I got this right, could not make sense of your text). Although both correlations are highly significant, their strength is moderate (correlation coefficients 0.15 and -0.25 for long and short histories respectively), implying largely noise-driven transitions.

Perception vs. Motor Matching of Slope: Unconscious Overestimates

Bruce Bridgeman, University of California, Santa Cruz
Merrit Hoover, University of California, Santa Cruz
Eric Chiu, University of California, Merced
Joshua Quan, University of California, Santa Cruz

Sunday, June 7th, 17:30–19:30: Poster No. 22

Slope perception can measure the interaction between objective reality and perception. Previous studies of slope estimation have used relatively long distances. Due to physiological evidence showing that some neurons in premotor and parietal cortex respond only to objects within arm«s reach, this study was designed to compare verbal and proprioceptive estimates of slopes in near and far space. Observers stood on a real hill on the U. of Ca. Santa Cruz campus and estimated slopes between themselves and a series of traffic cones at distances up to 16m. The proprioceptive measure was the posture of the right forearm held parallel to the slope, photographed with a digital camera. Verbal estimates in degrees greatly overestimated slopes, and the overestimates increased with distance by a log function. Proprioceptive estimates were more accurate at all ranges, but also increased with distance by a log function. When observers walk up and down the hill to be estimated before making their judgments, they continue to overestimate slopes just as much as non-walkers, despite experiencing more accurate slope perception at near distance on every part of the slope; appearance trumps experience. In a further experiment, a 1m segment at 16m from the observers was estimated to be steeper than a 1m segment at the observers' feet, showing that distance from the observer, not length of segment, is the origin of the length/slope effect. The results can be understood as an implicit slope previously measured in darkness, modulated by depth cues available at near distances.

Memory and Consciousness: Trace Distincitveness

Lionel Brunel, University Lyon 2
Ali Oker, University Lyon 2
Remy Versace, University Lyon 2

Saturday, June 6th, 17:30–19:30: Poster No. 109

In multiple trace theory of memory, the memory trace emerges by the activation and the integration of sensorimotor dimensions. In this view, it would be necessary for the trace to be sufficiently integrated and distinctive than others memory trace in order to be explicitly retrieved. The aim of this study is to show that manipulating the memory trace distinctiveness (through an isolation paradigm) can influence the feeling of recollection but not the familiarity, and also can influence the production of false recognition. Classically, works on isolation paradigm established that isolated materials are systematically associated to a better recall. Here, we have manipulated two kinds of isolation. First, items (words) were perceptually isolated (object word which typically referred to a noise, e.g. motor, or not, e.g. knife). Second, items (words) were conceptually isolated (two kinds of words which referred to ''living category'' or to ''artefact category''). In our study, the isolation paradigm was followed by a recognition associated to a confidence degree for each ''yes/no'' subject’s response. The results showed that the two kinds of isolation yield to the same effect on recognition times (old responses) whereas only the conceptually manipulation yielded an effect on confidence degree (old and new responses). In particular, we obtained a principal effect on the latencies for the two kinds of isolation (isolated items are recognized faster than other items) and an interaction on the confidence degree: for the conceptual isolation, isolated items were associated to a better confidence degree than other items whereas for perceptual isolation, they were not. Moreover, the results showed that the production of false recognition was due to the manipulation of isolation, in particular for non-isolated items. For us, the faster recognition times for isolated items were due to the familiarity process whereas the modulation of confidence degree by the conceptual isolation was due to the recollection process. The fact that conceptual isolation also produced an effect on recognition times supports the uni-process theory idea that what is at stake for the feeling of recollection is the memory trace.

How Conscious Are Minimally Conscious State Patients? A PET Study of Residual Brain Metabolism

Marie-Aurèlie Bruno, Coma Science Group, Cyclotron Research Center, University of liège
Caroline Schnakers, Cyclotron Research Centre, liège, Belgium
Melanie Boly, Cyclotron Research Centre, Belgium
Roland Hustincx, CHU liège, mèdecine nuclèaire
Audrey Vanhaudenhuyse, Coma Science Group, Cyclotron Research Center
Murielle Kirsch, Coma science group, CHU liège
Steve Majerus,
Claire Bernard, Mèdecine Nuclèaire, CHU liège
Gustave Gustave, University of liège
Steven Laureys, Cyclotron Research Centre, liège, Belgium

Saturday, June 6th, 17:30–19:30: Poster No. 129

Objectives: Patients in a minimally conscious state (MCS) show restricted signs of awareness but are unable to communicate consistently and reliably [1]. We here tested the hypothesis that this heterogeneous clinical entity can be subcategorized in low-MCS (i.e., patients showing movements to command) and high-MCS (i.e., patients only showing non-reflex behavior such as visual fixation or pursuit or localization of noxious stimuli), each characterized by its own specific residual cerebral brain function. Methods: Using FDG-PET, we assessed regional cerebral glucose metabolism (rCMRGlu) in 16 low-MCS (10 men; mean age 46 [SD 19] years; 5 traumatic) and 21 high- MCS (16 men; mean age 39 [SD 15] years; 11 traumatic). Data were preprocessed and analyzed by means of statistical parametric mapping (SPM5). Results were thresholded for significance at p<0.05 corrected for multiple comparisons. Results: Compared to low-MCS, high-MCS patients showed higher rCMRGlu in Broca’s and Wernicke's regions (areas 44 & 45, peak voxel x y z stereotaxic coordinates -42 12 4 mm; T value =2.50). Other identified areas were premotor postcentral and precentral cortices (areas 6, 3 and 4; coordinates -8 -6 66 mm; T = 3.62). Conclusion: The differences in brain metabolism between high- and low-MCS was not identified in widespread frontoparietal ''consciousness areas'' but in language, sensorimotor and premotor areas. These findings suggest that the main difference between these two subcategories of MCS, clinically separated by the presence of command-following, is their ability to express consciousness (verbally or non-verbally) rather than their level of consciousness per se. Reference: [1] Giacino et al, The minimally conscious state: definition and diagnostic criteria, Neurology. 2002.

Organizational Closure Through Neuronal Signal Regeneration as a Possible Basis for Conscious Awareness

Peter Cariani, Harvard Medical School

Saturday, June 6th, 17:30–19:30: Poster No. 99

We outline a new kind of theory for the neural requisites of conscious awareness and its contents. Our philosophy is a modernized Aristotelian hylomorphism in which mind is the informational organization of the brain, and awareness is a necessary (but epiphenomenal) concomitant of (regenerative) organizations of neuronal activity. The structure of awareness supervenes on that of neuronal informational processes both in terms of states of consciousness and specific contents of experience. Every experiential dimension has a corresponding neural coding dimension (psychoneural isomorphism). We hypothesize that the neuronal, organizational requisite of unified, continuous awareness is an “autopoiesis of neuronal signals. Analogous to autocatalytic networks in the theory of life that regenerate material components and relations (autopoiesis = self-production), we envision sets of mutually-amplifying and reinforcing neuronal signals that actively regenerate themselves in recurrent circuits. Those mutually-compatible signal sets that exceed a threshold signal-to-background level evoke concomitant changes in awareness. The maintained coherence of signals may be necessary for their successful regeneration in recurrent circuits, such that conscious awareness can be disrupted not only by suppression of neural activity, but also by ''scrambling'' of signals. Thus, even as neuronal activity persists under general anesthesia and seizure, the forms of the neural signals may be disrupted in a manner that renders them incapable of regeneration. The framework may explain why ignition and self-sustained activity in recurrent/re-entrant circuits is necessary for awareness. However, unlike other systems-based explanations, this theory posits that the necessary and sufficient basis for awareness is a ''circular causal'' cybernetic organizational closure of signals rather than an information-complexity threshold per se (e.g. Basar's effective dimensionality metric or Tononi's Phi measure). While these latter measures usefully quantify the complexity of the contents of awareness, by themselves they lack the dynamic organizational sustainability criterion that regenerative processes satisfy. In analogy to life, ability to regenerate organization rather than attain static complex states is what distinguishes a living (albeit simple) organism from a nonliving (albeit complex) material system. In similar fashion, regenerative organizational closure may distinguish conscious from non-conscious systems. The hypothesis is empirically testable once central neural codes are identified.

Deduction Without Awareness

Reverberi Carlo, BCCN, Charitè - Universitätsmedizin
Burigo Michele, IRCSS E.Medea, Italy
Cherubini Paolo, University Milano - Bicocca, Italy

Sunday, June 7th, 17:30–19:30: Poster No. 10

We investigated whether two basic deductive inferences, namely modus ponens and disjunctive syllogisms, are automatic processing steps carried out even in the absence of awareness. We used an exhaustive set of conditional and disjunctive problems concerning numbers, with a subliminal second premise, followed by a target number. Participants evaluated whether the target number was odd or even. The target number could match, or not match a valid conclusion endorsed in the previous problem. We found that evaluations of target numbers matching the conclusion of a modus ponens were faster than the evaluations of numbers following all other types of problems. This finding suggests that unlike disjunctive syllogisms, modus ponens is automatic, can be triggered by subliminal stimuli and can be performed without awareness. The finding extends the range of high-level cognitive activities that can be carried out unconsciously to include critical deductive inference schemata.

Is the Body Schema Sufficient for the Sense of Embodiment? an Alternative to de Vignemont's Model

Glenn Carruthers, Macquarie Centre for Cognitive Science

Saturday, June 6th, 17:30–19:30: Poster No. 37

De Vignemont argues that the sense of ownership comes from the localization of bodily sensation on a map of the body that is part of the body schema. This model should be taken as a model of the sense of embodiment. I argue that the body schema lacks the theoretical resources needed to explain this phenomenology. Furthermore, there is some reason to think that a deficient sense of embodiment is not associated with a deficient body schema. The data de Vignemont uses to argue that the body image does not underlie the sense of embodiment does not rule out the possibility that part of the body image I call 'offline representations' underlies the sense of embodiment. An alternative model of the sense of embodiment in terms of offline representations of the body is presented.

Suffering: A Complex Adaptive Systems Approach

Richard Chapman, The University of Utah

Saturday, June 6th, 17:30–19:30: Poster No. 125

People experiencing sustained pain, depression, loss, ostracism or a traumatic event experience negative affect, cognitive difficulties, fatigue, disturbed sleep, diminished appetite and libido, along with many problems of physical discomfort and dysfunction. The literature on stress addresses these disorders but does not provide a comprehensive account of how these problems arise, the physiological mechanisms of the disorders and their relationship to mental processes, or what would constitute a curative intervention versus symptom control. This presentation advances a complex adaptive systems view of stress in which physical, mental or social trauma generates a complex, normally adaptive stress response. This response extends beyond the nervous system to involve endocrine and immune systems, and the combined effect of responses across these systems contributes to the subjective experience of emotion and bodily awareness. Through a common chemical language comprising neurotransmitters, peptides, endocannabinoids, cytokines and hormones, an ensemble of interdependent nervous, endocrine, and immune processes operates in concert to cope with the stressor. These processes act as a single agent and thus comprise a supersystem. Regulation within the supersystem depends upon negative feedback loops, and the supersystem's resources for coping with threat depend upon positive feedback processes. Emotional distress, physical disorders, disturbed biorhythms and fatigue result from unresolved dysregulation within the supersystem. Social stressors can compound the stress resulting from a physical injury or depression, or they an act alone to dysregulate the supersystem. When components of the supersystem fall into dysregulation, health, function and sense of well-being deteriorate, and the person experiences suffering. Many of the multi-system disorders that resist conventional medical treatment are the product of supersystem dysregulation. Individuals vary in their vulnerability to dysregulation and to dysfunction in particular organ systems due to the unique interactions of genetic, epigenetic and environmental factors, as well as the unique past experiences that characterize each person. The presentation covers several ways to quantify supersystem dysregulation. Other key constructs are stress, allostasis, and hormesis. A key feature of this framework is the psychophysiological systems interdependence of brain activity, endocrine system, immune system and autonomic nervous system.

The Dilemma of Inverted Spectrum Thought Experiment

Emma Chien, National Yang Ming University
Allen Houng, National Yang Ming University

Sunday, June 7th, 17:30–19:30: Poster No. 60

Inverted spectrum' is a thought experiment used to argue against functionalism. In the case of inverted spectrum, a person's color vision is systematically inverted relative to our color vision. For example, this person's color experience of seeing a red apple is identical with our color experience of seeing green grass, while his experience of seeing green grass is identical with our experience of seeing a red apple. Despite such inversion, there is no functional or behavioral difference between this person and us. Thus, this case is a counterexample of functionalism, according to which functionally identical mental states are identical in all aspects, including their phenomenal characters. If the case of inverted spectrum is possible, then functionalism is proved to be wrong. In this paper, I will argue that, according to psychological and neurophysiological researches of color vision, the case of inverted spectrum is impossible. Because our phenomenal color space and biological color space are asymmetrical, it is impossible to invert experiential states without making any difference. Furthermore, there are specific associations of our emotional reactions to colors. Based on the asymmetry of our color space and the color-emotion association, it is implausible for color experiences to be inverted while maintaining our functions and behaviors unchanged. Furthermore, if the inverted spectrum is possible, then color experiences are epiphenomenal. Because the only way to keep the functions and behaviors unchanged while having inverted spectrum is to admit that color experiences have no causal powers and are not participating in the causal chain of our response to colors. The case of inverted spectrum then becomes no more threat to functionalism.

Qualia: Realism Without Cartesianism

Ron Chrisley, University of Sussex

Saturday, June 6th, 17:30–19:30: Poster No. 45

Some argue that in order to take consciousness seriously, one must acknowledge the existence of qualia, and this amounts to admitting the existence of states that are 1) ineffable, intrinsic, private, and immediate (Dennett); or that are 2) not implied by the non-phenomenal facts (Chalmers). The only other possible position is thought to be some kind of anti-realism or fictionalism: denial of the existence of such states. If there are no things that have the properties attributed in 1) or 2), then the term ''qualia'' should go the way of the alchemical term ''phlogiston'', and be eliminated from scientific discourse. The only options seem to be Cartesianism or eliminativism. This talk offers a middle way for those who find both of these extreme positions unpalatable. It is argued that all one need do to be a phenomenal realist, and thus be taking consciousness seriously, is to acknowledge that terms like “qualia” do, unlike ''phlogiston'', refer. But one can do this without committing oneself to the claim that there are things that are ineffable, intrinsic, private, and immediate, or which are not implied by the non-phenomenal facts. This is because it may be that the term ''qualia'', like ''gold'', has its reference fixed not by description (e.g., having the properties mentioned in 1) and 2)), but by causal relations between the term/concept ''qualia'' and whatever caused it to be introduced in the first place, and/or whatever conditions resulted in it being a useful term. Just as showing that the beliefs the ancients had about gold were false did not amount to showing that there is no such thing as gold, so also does showing that there is nothing that meets conditions 1) and 2) fall short of establishing that qualia do not exist. Of course, it *may* turn out that ''qualia'' *is* like ''phlogiston'': it so poorly maps to anything actual that there will be no place even for a successor concept of qualia in a future science of the mind. The point is that this should be decided empirically/experientially; it is not the place of philosophy to prejudge the issue.

Early EEG Signals Predict "Free" Decisions Several Seconds Before They Are Made

Thomas Christophel, BCCN, Charitè - Universitätsmedizin Berlin
John-Dylan Haynes, BCCN, Charitè - Universitätsmedizin Berlin

Saturday, June 6th, 17:30–19:30: Poster No. 79

To which degree unconscious brain processes influence or even cause our conscious decisions is a question of immense interest. For simple movements, the perceived onset of the ''will'' to move is preceded by a deflection in the human EEG (Libet 1985). This so called readiness potential begins up to 500 ms before the will to move reaches awareness and it can show lateralization for left and right hand movements (Haggard and Eimer 1999). Recently, we demonstrated that the outcome of a free decision can be predicted from fMRI signals even up to 7 s before it enters awareness (Soon et al. 2008). Here, we probe this early information content of medial prefrontal cortex using electrophysiological data. While observing a rotating clock healthy subjects decided freely at a time of their own choice to press the right or the left button. After each button press, they judged the time of their conscious decision using an adjustable clock. Standard 64-Channel EEG data was recorded, preprocessed and introduced into a infomax independent component decomposition. Single trial independent component time courses were then utilized in combination with a support vector classification algorithm to decode the subjects’ choices. In all subjects, the offset of the awareness of will to move relative to the button press did not exceed one second. Importantly, we successfully extracted information about the forthcoming button press from recordings obtained as early as five seconds before the button press. Our findings provide further evidence that brain activity carries predictive information about a person’s decisions several second before their decision is consciously made, thus confirming previous fMRI studies (Soon et al. 2008). This implies that this early predictive information is not restricted to the hemodynamic domain, since we can now identify similar information sources in EEG signals using powerful analysis methods. Thus, conscious decisions are presumably at least partially determined by unconscious decision processes.

The Red Inside Your Head: Decoding Seen Colour from Activity in Human Visual Cortex

Colin Clifford, School of Psychology, University of Sydney
Erin Goddard, School of Psychology, University of Sydney
Damien Mannion, School of Psychology, University of Sydney
Scott McDonald, School of Psychology, University of Sydney
Sam Solomon, Medical Sciences, University of Sydney
Colin Clifford, School of Psychology, University of Sydney

Saturday, June 6th, 17:30–19:30: Poster No. 137

Recent studies employing multivariate pattern classification (MVPC) have shown that the orientation of spatial image structure and the direction of image motion can be decoded from primary visual cortex of the human brain. The success of MVPC in these instances has been attributed to biased sampling by the voxel array of an orderly representation of the relevant visual attribute across the cortical surface. Here, we used MVPC to investigate the representation of colour in human visual cortex. The existence of a systematic spatial mapping of chromatic selectivity across the surface of primate visual cortex is less well established than for orientation and motion. Furthermore, the cortical mechanisms of colour vision in humans have not been as well characterised as those in subcortical areas. We used MVPC to ask where in human visual cortex we could find evidence of a representation of colour that transforms signals from subcortical inputs. We used fMRI at 3T (1.5mm cubic voxels; TR=3s) to test for higher order cortical representations of colour capable of classifying stimuli that cannot be distinguished by the postulated red-green (L-M) and blue-yellow (S-(L+M)) subcortical opponent channels. Subjects (n=5) viewed each of two plaid patterns modulating in colour between orange-cyan or lime-magenta. Multivariate pattern classifiers restricted to each of several visual areas (V1, V2, V3, V3A/B, hV4, VO, MT+) were trained to discriminate the two patterns. The classifiers were trained on signals from 9 trials and tested on a tenth; this procedure was repeated 10 times. Classifiers performed significantly better than chance as early as V1. The success of the classifiers implies that subcortical chromatic channels are recombined early in cortical processing to form novel representations of colour. It also suggests a possible systematicity in the spatial mapping of colour onto the cortical surface. Work is now ongoing to extend this approach to the decoding of purely subjective colour from the brains of synaesthetic observers.

Sleep in Disorders of Consciousness

Victor Cologan, Centre de Recherches du Cyclotron
Manuel Schabus, University of Salzburg, Austria & Cyclotron Research Centre, liège
Pierre Maquet, Cyclotron Research Centre, Belgium
Steven Laureys, Cyclotron Research Centre, liège, Belgium

Sunday, June 7th, 17:30–19:30: Poster No. 106

From a behavioural as well as neurobiological point of view, the sleep-wake system and consciousness are intimately connected. Therefore, a better characterization of sleep timing and architecture in patients suffering from clinical disorders of consciousness (DOC) might improve our understanding of the neural correlates of consciousness. From research on sleep in DOC (coma, vegetative state, locked-in syndrome), it is known that a wide spectrum of sleep disturbances - ranging from almost normal patterns to severe changes in sleep architecture and quantity - can be present under these conditions. Interestingly, some sleep features even appear to have diagnostic and prognostic value. However, defining sleep in severely brain injured patients is problematic. As their behavioural and electrophysiological signs of wake and sleep differ in many ways from healthy individuals, the applicability of standard scoring criteria is object of debate. Here we present first data from patients in a vegetative state (VS; n=9) and, for the first time, the minimally conscious state (MCS; n=9) using bedside 24H polysomnography after completing the behavioural Coma Recovery Scale-Revised (CRS-R, 2004). In MCS, epochs of higher muscular, eye and heart activity, probably indicating wake, alternates with epochs of predominant delta activity, muscular atonia and heart-rate decrease, probably indicating sleep. On the other hand, most VS showed more iso-electric recordings with few epochs of sleep-like activity. We aim to discuss these recordings from an exploratory point of view, compare the two clinical entities and critically assess the applicability of standard sleep criteria in these patients.

Signal Detection-Theoretic Model Explains Libet's Awareness of Intention and Attentional Prior Entry

Matthew Davidson, Columbia University
Stanislav Nikolov, Massachusetts Institute of Technology
Tashina Graves, Columbia University
Dobromir Rahnev, Columbia University
Hakwan Lau, Columbia University

Sunday, June 7th, 17:30–19:30: Poster No. 68

Two apparently paradoxical findings in the field of human time perception are considered within a framework of probabilistic signal detection. First, it has been found that spontaneous, self-paced motor actions are preceded by an early preparatory signal (the readiness potential, RP) recordable from scalp and intracranial electroencephalograms up to 1-2 seconds prior to action execution. However, according to Libet's research, subjects only report conscious awareness of the intent to act to be about 200 ms before action execution. Second, psychophysical findings support the principle of attentional prior entry, which states that attended stimuli are perceived faster than unattended stimuli even when occurring simultaneously. However, electrophysiological studies have failed to find corresponding temporal differences between neural signals for attended and unattended stimuli. We suggest that the key to understanding these phenomena is to think of time perception in terms of signal detection theory (SDT). We modeled signal onset perception as a continuous process of signal detection. Given a signal mixed with noise, the brain could determine the onset by performing signal detection at every time point. The probability of detection then depends on both the signal and noise distributions. In our framework, the brain uses a consistent criterion, and gives a positive response when the signal plus noise at the given time point crosses the criterion, and a negative response otherwise. We performed simulations using slow-rising signals with Gaussian noise added, which mimics natural neural signals such as the RP. According to our model, the statistically optimal criterion for detecting on onset of the RP would be a conservative one. At this conservative criterion, the early part of the signal will not be detected. This accounts for the Libet findings of delayed awareness for action. To explain attentional prior entry, we treated attention as boosting the signal-to-noise ratio relative to the unattended signal. Our simulations showed that this causes the attended signal to be consistently detected earlier than the unattended signal. Both of these results suggest that many aspects of conscious awareness could be explained in terms of signal detection of neural activity.

Biased Partial Awareness: How Partial Information and Contexts Create Complete Conscious Experience

Vincent de Gardelle, LSCP, DEC-ENS

Saturday, June 6th, 17:30–19:30: Poster No. 1

Current theories of consciousness often assume that the relevant content of consciousness corresponds to a stimulus taken as a whole. For instance, in several taxonomies of conscious states proposed recently (e.g., subliminal vs. P-consciousness vs. A-consciousness, or subliminal vs. preconscious vs. global access), the question is to determine the conditions for a whole stimulus to reach a certain stage. Similarly, in another debate, deciding whether consciousness is a graded or all-or-none phenomenon is also treated only by considering the stimulus as a whole. Here, I will defend the partial awareness framework, and I will present a rich set of empirical data from our lab supporting the hypothesis that conscious experience reflects an elaborated construct resulting from the interaction between partial information in the stimulus and contextual factors internal to the subject. I will present several experimental studies (e.g. Sperling-like partial-report experiments, masked face priming, low-level orientation tasks) in support of a ''biased partial awareness'' state of consciousness. Partial awareness corresponds to a situation in which some of the information in the stimulus is consciously detected, while information at other levels of processing remains non-conscious. In that case, I will provide evidence that a perceptual recovery process occurs to produce a complete conscious percept. This mechanism relies on the stimulus information (only some of which might then become conscious), as well as on contextual factors (e.g. current task and expectations, the set of active representations in the current situation, etc.). Crucially, the ''biased partial awareness'' paradigms I will present show that conscious perception may not faithfully correspond to the information presented in the stimulus. Finally, I will argue that the notion of partial information may enrich consciousness studies. Relying on the basic but powerful psychological notion of a hierarchy of levels of processing, which has been underestimated in recent debates on consciousness, is crucial for investigating the mechanisms supporting the complex elaboration of conscious experience.

The Feeling of Understanding Consciousness

Herman De Regt, Tilburg University
Hans Dooremalen, University of Amsterdam

Saturday, June 6th, 17:30–19:30: Poster No. 53

Why can we not solve the hard problem of consciousness? According to David Chalmers (1996) this is due to the fact that phenomenal consciousness cannot be functionally (i.e. causally) analyzed. As science uses functional explanations, he argues, we need to change science in order to be able to solve the hard problem. On the one hand, we agree with Chalmers that a functional analysis of phenomenal experiences would solve the hard problem. On the other, and contrary to Chalmers, we argue that this can in fact be done and that science needs no revolutionary change of fundamental theories. Using the empirical data provided by Derek Denton et alia (1999), we present a functional analysis of the experience of thirst. Additionally - still following Chalmers' model of explanation - it can be shown how thirst is realized, thereby providing a full explanation, and solving the hard problem of thirst (cf. Hohwy & Frith 2004). Still, many would reject this as an explanation of the experience of thirst. The main reason for not accepting this explanation as an explanation, we argue, is that it does not feel right (Gopnik 1998): the explanation does not generate the feeling of understanding thirst. We show that in general scientific studies of phenomenal consciousness simply lack the right philosophical backing to generate the feeling of understanding consciousness. We aim to offer such a philosophical framework. Our main thesis in this respect is that it is not necessary for science to change in order to generate the feeling of understanding of consciousness, but that it is necessary to change our view of science in order to generate such feeling. That view of science can be traced back to the work of John Dewey (1929) and more recently to the work of Bas van Fraassen (2002), the kernel of which is that the inference from the success of the natural sciences (including the neurosciences) to materialism is not coercive. Accepting this might finally generate in us the feeling of understanding phenomenal consciousness.

Workspace and Sensorimotor Theories: Complementary Approaches to Experience

Jan Degenaar, Faculty of Philosophy, University of Groningen
Fred Keijzer, University of Groningen,Faculty of Philosophy

Saturday, June 6th, 17:30–19:30: Poster No. 49

A serious difficulty for theories of consciousness is to go beyond mere correlation between physical processes and experience. Currently, neural workspace and sensorimotor contingency theories are among the most promising approaches to make any headway here. In this talk we explore the relation between these two sets of theories. Workspace theories, as proposed by Baars, Deheane, Naccache, Varela and others, build on large-scale activity within the brain. Sensorimotor theories on the other hand include external processes in their explanations, stressing the sensorimotor contingencies that arise from our interaction with the environment, as proposed for example by Hurley, Myin, No' and O'Regan. Despite the basic differences between the approaches, we argue that workspace- and sensorimotor theories are complementary rather than competitive. By combining these theories, a number of problems that hamper these individual theories may be overcome and their strengths combined. We argue that workspace theories have better prospects for explaining how there can be consciousness in the first place, while sensorimotor theories are most promising for making sense of the specific phenomenal character of experiences. References: Baars, B.J. (1988), A Cognitive Theory of Consciousness (Cambridge: Cambridge University Press). Dehaene, S.; Naccache, L. (2001), 'Towards a cognitive neuroscience of consciousness: basic evidence and a workspace framework', Cognition, 79, pp. 1-37. Hurley, S.; Noë, A. (2003), 'Neural plasticity and consciousness', Biology and Philosophy, 18, pp. 131-68. O'Regan, J.K.; Myin, E.; No', A. (2005), 'Sensory consciousness explained (better) in terms of 'corporality' and 'alerting capacity', Phenomenology and the Cognitive Sciences, 44, pp. 369-87. O'Regan, J.K.; No', A. (2001), 'A sensorimotor account of vision and visual consciousness', Behavioral and Brain Sciences, 24, pp. 939-1031. Varela, F.J.; Lachaux, J.-P.; Rodriguez, E.; Martinerie, J. (2001), 'The brainweb: phase synchronization and large-scale integration', Nature Reviews Neuroscience, 2, pp. 229-39.

The Behavioural Quantification of the Resting State

Athena Demertzi, University of liège
Audrey Vanhaudenhuyse, Coma Science Group, Cyclotron Research Center, University of liège
Quentin Noirhomme, Coma Science Group, CRC, University of liège
Melanie Boly, Cyclotron Research Centre, Belgium
Serge Brèdart, University of liège
Steven Laureys, Cyclotron Research Centre, liège, Belgium

Sunday, June 7th, 17:30–19:30: Poster No. 18

Background: Despite neuroimaging evidence that awareness of environment (external) and of self (internal) are anticorrelated, switching their activation at an average rate of 0.06 Hz (Boly et al., 2008, Soddu unpublished data), little is known about the behavioural quantification of the resting state. The present study provides behavioural data on the relationship between external and internal awareness. Methods: 31 healthy volunteers were in a resting condition (i.e. sitting with eyes closed), avoiding structural thinking (e.g. counting). 66 auditory prompts were presented at random intervals via headphones. The participants' task was to rate on a keyboard their external and internal awareness state as it was before the presentation of the prompt on a 4-point scale (0 = absent; 1 = mild; 2 = moderate 3 = maximal). The content of awareness was indentified via thought sampling. Results: At the individual level, 24/31 subjects showed significant anticorrelation between internal and external awareness (1/31 positive correlation, 6/31 no significant correlation, p< .05). At the group level, Spearman's r was calculated at -.44, (p< 0.02 two-tailed). On average, the switching from internal to external occurred at 0.05Hz (range: 0.01-0.1Hz). Self-reports for external awareness included auditory (100% subjects), somesthetic (90%), olfactory (20%) and visual (10%). Self-reports for internal awareness included experiment-related thoughts (80% subjects), autobiographical (65%) and inner speech (20%). Discussion: Our results confirm the predicted anticorrelation between internal and external awareness at the behavioural level. The temporal dynamics of external to internal switch is in line with previous neuroimaging data. Our study bridges the cognitive and physiological characteristics of the brain 'default' resting state activity.

Dissociating Automatic and Conscious Influences in Associative Learning

Arnaud Destrebecqz, Universitè Libre de Bruxelles
Pierre Perruchet, Universitè de Bourgogne - CNRS
Philippe Peigneux, Universitè libre de Bruxelles
Axel Cleeremans, Universitè Libre de Bruxelles

Sunday, June 7th, 17:30–19:30: Poster No. 138

A dissociation between subjective expectancy and motor behavior has been reported in a simple associative learning task (Perruchet, Cleeremans, & Destrebecqz, 2006). According to previous conditioning studies (Clark, Manns, & Squire, 2001), this dissociation is observed when the to-be-associated events co-terminate and thus overlap in time (a training regimen called delay conditioning), but not when they are separated by a temporal delay (trace conditioning). In this latter situation indeed, there tends to be a direct relationship between subjective expectancy and behavior. In this study, we further investigated this issue in a series of experiments where conscious and unconscious components of performance have been pit against each other. In Experiments 1-3, subjects performed a simple reaction time task in which a preparatory signal (a tone) either overlapped with or terminated earlier than the imperative stimulus (a visual target presented in 50% of the trials). After each response, subjects also had to state how much they expected the imperative stimulus to be displayed on the next trial. Results indicate that reaction times tend to decrease when the tone is consistently followed by the visual target across successive trials, whereas conscious expectancy for the target decreases at the same time. Importantly, we systematically found that the temporal relationship between the tone and the target failed to influence performance. In a fourth experiment, we examined whether these results extend to a two-choice reaction time task. To our surprise, we observed a direct relationship between subjective expectancies and reaction time in that situation. We nevertheless observed that the introduction of a delay between the tone and the target had, once again, no effect on performance.

Reductive Explanation and A Priori Entailment

Esa Diaz-Leon, University of Manitoba

Saturday, June 6th, 17:30–19:30: Poster No. 55

David Chalmers and Frank Jackson (2001) have argued that in order to provide a reductive explanation of consciousness in physical terms, we have to show that microphysical truths entail phenomenal truths a priori. In this paper, I examine their arguments for that conclusion, and I criticize a crucial premise in the argument, namely, the claim that microphysical truths entail ordinary non-phenomenal macroscopic truths a priori, such as 'water covers 60% of the Earth surface'. I focus on the question about the possession conditions of macroscopic concepts, such as WATER or PLANET, and I distinguish two views: a strong view and a weak view. These two views differ over the amount of information that they attribute to competent users of the concepts. According to the strong view, competent users of a concept have the ability to infer truths involving such concept from lower-level truths, whereas according to the weak view, all that is required in order to be a competent user of a concept is to be able to infer truths involving that concept from other truths, which need not be lower-level truths. I argue, firstly, that the strong view is problematic because it trivializes the notion of a priori knowledge; and secondly, that if we endorse the weak view, then it follows that sentences such as 'If the microphysical facts are so and so, then water covers the 60% of the Earth surface' are not a priori true. Therefore, I conclude that Chalmers and Jackson have not presented a convincing case for the claim that physical truths a priori entail ordinary macroscopic truths.

Subliminal Behavioral Priming: It is All in the Brain, but Whose Brain?

Stèphane Doyen, Universitè Libre de Bruxelles
Olivier Klein, UPsyS
Cora-Lise Pichon, ULB
Axel Cleeremans, Universitè Libre de Bruxelles

Saturday, June 6th, 17:30–19:30: Poster No. 123

Priming subjects may automatically and unconsciously activate a representation (e.g.: the stereotype of aged persons) that lead them to act according to it (e.g.: walking slower than usual when exiting the experiment room). In order to explain this effect, some authors suggested the existence of a direct link between perception and behavior, regulating some of the automatic part of our social life. Though this concept seems widespread in the field of social cognition on the one hand, the very idea of subliminal semantic activation remains highly controversial in the field of cognitive psychology on the other hand. In an effort to conciliate these two views, we conducted a series of experiments, all trying to replicate and improve Bargh, Chen, and Burrows' (1996) now classic experiment. To question their results, we introduced more accurate walking speed as well as more throughout awareness measurements. We also manipulated the expectations of the experimenter on primed subjects’ behavior in order to assess the effect on walking speed that could result from the experimenter own beliefs.

Imaging Transitions in Consciousness

Martin Dresler, Max Planck Institute of Psychiatry
Renate Wehrle, Max Planck Institute of Psychiatry, Munich
Victor Spoormaker, Max Planck Institute of Psychiatry, Munich
Stefan Koch, Charitè University Hospital, Berlin
Axel Steiger, Max Planck Institute of Psychiatry, Munich
Hellmuth Obrig, Charitè University Hospital, Berlin
Philipp Sämann, Max Planck Institute of Psychiatry, Munich
Michael Czisch, Max Planck Institute of Psychiatry, Munich

Saturday, June 6th, 17:30–19:30: Poster No. 107

Common categorizations of consciousness distinguish between basal and higher-order aspects of this multifaceted concept. Basal consciousness comprises perceptions and sensations, whereas higher-order consciousness constitutes reflections on these perceptions. In rapid eye movement (REM) sleep, ordinary dreaming comprises only basal aspects of consciousness. There is conscious perception of dream content but higher-order aspects of consciousness are usually absent: The person does not realize that he is dreaming and has no perception of the self as an agent endowed with intentionality and free will. Higher-order consciousness therefore seems to be bound to wakefulness. However, a special type of dreaming - referred to as 'lucid dreaming' - is characterized by full-blown consciousness, including all higher-order aspects: The dreamer is able to reflect on his state of consciousness and realizes that he is dreaming. Using a combined fMRI/EEG approach, we could reveal neural activity related to the genesis of higher-order consciousness by contrasting ordinary REM sleep with physiologically verified lucid REM sleep. We find increased activation in a range of neo-cortical regions, including bilateral precuneus, cuneus and parietal, prefrontal and occipito-temporal cortices, to be related to this categorical shift in consciousness. This activation shows remarkable overlap with neo-cortical regions that have highest expansion in humans relative to non-human primates.

Evaluating the Contribution of Discrete Perceptual Mechanisms to Psychometric Performance

Julien Dubois, CNRS, Centre de Recherche Cerveau et Cognition
Rufin VanRullen, CNRS,Centre de Recherche Cerveau et Cognition

Saturday, June 6th, 17:30–19:30: Poster No. 23

What are the brain mechanisms underlying perception? It has been suggested that attention and/or perception could operate discretely, or that attention could relay information to perception discretely, i.e. in successive cycles. The most telling manifestation of a periodic sampling process is temporal aliasing, by which a periodic stimulus evolving in one direction can be interpreted in the opposite direction. In a typical psychophysical task subjects can be asked to judge the direction of a periodic stimulus, and their performance can be recorded as a function of the stimulus temporal frequency: aliasing would predict some oscillations in the resulting performance function. However, the quest for aliasing in a variety of tasks is subject to some pitfalls. One pitfall is the presence of non-periodically sampled (i.e. continuous) processes which could also contribute to perception and thus cover the tracks of any underlying periodic process. Another one is that the rate at which samples are taken may slightly vary across subjects, as well as across trials for a single subject. In addition, the appearance of oscillations in a given performance function could sometimes be artifactually induced by measurement noise. How does one then extract evidence for aliasing in psychophysical data, and how does one quantify it? The general method that we introduce here is based on modeling the performance that would be expected from the respective contributions of a periodically sampled process and a non-sampled (continuous) process to perception; the sampled process depicts an oscillation, characterized by its sampling frequency and variability, whereas the non-sampled process is reflected by a more classical psychometric function characterized by its threshold and slope. Each subject's performance can be fitted within such a model. To assess significance, the contribution of the sampled process to perception and the corresponding sampling frequency can be compared, within and across subjects, to surrogate values obtained under the null hypothesis of a classical psychometric function with additive measurement noise. We illustrate this new analysis method on a psychophysical task involving motion perception, which has previously been proposed to support temporal aliasing (cf. the continuous Wagon-Wheel Illusion).

Mentalistic Metatheory and Methodology

Donelson Dulany, University of Illinois

Sunday, June 7th, 17:30–19:30: Poster No. 94

When ASSC convenes near the birthplace of psychology as a science of consciousness, I especially want to outline for clarity and discussion what I believe to be a metatheory and methodology for a most productive science of consciousness. On mentalistic metatheory, consciousness is not a ''system'' or ''space''; it is a set of mental states interrelated by nonconscious neural operations within mental episodes to be theoretically analyzed. Conscious states carry symbols, functionally specified-that represent past, present, and future events out there and in our own consciousness, in the enormous range of significant mental activities, now widely recognized and investigated but with too little analysis of mental episodes interrelating conscious states. The metatheory and the theories it generates should then focus on what consciousness explains rather than on the explanation of consciousness, an explanation that emerges with recognition of the adaptive value of consciousness in carrying those symbols in those significant mental activities. Neural correlates can have value by enriching the theoretical networks, to the degree the brain imaging meets certain standards, but neither correlate can be "explained" by correlation alone. A conscious state lies at the intersection of a set of variables: agency, mode, and content, each of which can be assessed. Mental contents may be "identity symbols" or only ''literal symbols'' that precede and surround attentional identity symbols. Mental episodes may be deliberative, interrelating propositional states, or associative-activational (''evocative'') interrelating sub-propositional states. Mental episodes occur between transductions of sensory inputs and motor outputs and may activate and be activated by neural memory networks, all of which are non-conscious and non-symbolic. Methodologically, we should seek relations among phenomenal reports of specific conscious states, not as a first person data language, which cannot meet standard data language requirements, but as the most specific indices of conscious states described in first person theory language. Reports as indices of theoretical conscious states must meet certain validity standards: specificity, verbalizability, the memorability required for their higher-order representation. Bayes' theorem, Duhem-Quine thesis, and network theory together provide a methodology of competitive support for a theory and further validation of reports-as briefly illustrated.

Unconscious Priming Through Gaze-Contingent Substitution

Nathan Faivre, Laboratoire de Sciences Cognitives et Psycho
Sid Kouider, Laboratoire de Sciences Cognitives et Psychol

Saturday, June 6th, 17:30–19:30: Poster No. 3

We present a new paradigm called ''gaze-contingent substitution'' which aims at characterizing unconscious processes under conditions of inattention. This method offers the advantage of bypassing the limiting factors encountered in paradigms which use masking or the attentional blink, where the subliminal stimuli are presented very briefly, as well as in the interocular suppression method, in which the unreportable stimuli are suppressed very early on by the visual system (i.e. in subcortical pathways). Here, using an Eye-tracker, we presented stimuli that were long-lasting (e.g. the prime displayed for one second or more) but, crucially, as soon as the gaze diverged towards them from another distracting stimulus (the target), they were substituted by fillers. Unconscious influences were investigated by manipulating the prime-target relation and measuring priming effects. We report two studies, one using either congruent vs. incongruent arrows to measure sensori-motor priming, and the other using repeated vs. unrelated faces to measure perceptual priming. In both cases, we report robust priming under condition where the unattended stimuli remained impossible to identify. These results are interpreted in the theoretical framework of preconscious perception of supraliminal but unattended stimuli developed by Dehaene et al. (2006).

Subjective Versus Objective Simultaneity: A Magnetoencephalography (MEG) Study

Christine Falter, University of Oxford
Vincent Walsh, University College London
Sven Braeutigam, University of Oxford
Julian Kiverstein, University of Edinburgh
Anthony Bailey, University of Oxford

Saturday, June 6th, 17:30–19:30: Poster No. 19

Magnetoencephalography (MEG) is a non-invasive brain imaging technique that records magnetic fields associated with synaptic transmission in the human brain. MEG has good spatial resolution and extremely high temporal resolution. We used event-related MEG to examine the brain correlates of simultaneity perception in healthy adults. The objective of the current study was to test if the brain processes apparent simultaneity induced by non-simultaneous stimuli like objective simultaneity or like objective non-simultaneity. Two horizontal bars were presented simultaneously (onset and offset), or non-simultaneously employing two different onset delay conditions. The short delay separating the two bars was 17ms, which is below the subjective visual simultaneity threshold (i.e. apparent simultaneity), and the long delay was 117ms, which is well above the threshold (i.e. clearly non-simultaneous). All measurements were taken using a Neuromag-306 VectorViewTM system, providing a helmet-shaped array of 102 pairs of gradiometers. The responses were averaged separately for physically simultaneous stimuli and non-simultaneous stimuli in the two delay conditions. Significant differences between the evoked responses were sought using a time-dependent measure that takes into account the data from all sensors. As a general pattern in all three conditions we found more event-related MEG activity in the right hemisphere from a latency of about 300ms onwards, which was particularly confined to the parietal cortex in the case of non-simultaneity, but more distributed towards prefrontal and temporal areas in the simultaneity and apparent simultaneity conditions. These findings are broadly consistent with suggestions of a right hemispheric 'when' pathway. A widely distributed long-latency activation was apparent in the simultaneity condition but attenuated in the apparent simultaneity condition from about 600 ms onwards which suggests a difference in the processing of real and apparent simultaneity. Our results indicate a difference between conscious and non-conscious processing of temporal information.

Functional Connectivity Within the Default Mode Network in a Patient in Vegetative State

Davinia Fernandez-Espejo, University of Barcelona
Carme Junquè, University of Barcelona
Pere Vendrell, University of Barcelona
Eva Rivas, Hospital Clinic, Barcelona
Neus Fˆbregas, Hospital Clinic, Barcelona
Jose Maria Mercader, Hospital Clinic, Barcelona

Sunday, June 7th, 17:30–19:30: Poster No. 132

Introduction: Vegetative State (VS) is a clinical condition characterized by the maintenance of the arousal in absence of awareness. There have been several attempts of characterize the neural correlates of vegetative state and other disorders of consciousness. Early metabolic studies using positron emission tomography (PET) have shown dysfunctions and functional disconnections in fronto-parietal associative areas known as being part of the ''default mode network'' in VS patients. Methods: A patient with traumatic brain injury was scanned twice. The first scanner was performed 33 days after the injury when the patient was in VS. The second scanner was performed when the patient had recovered the consciousness at 13 months of evolution. We used resting-state functional connectivity magnetic resonance imaging analysis to study the changes in the default mode network that accompanied the recovery of the consciousness. SPM5 running in Matlab 6.5 was applied to detect brain regions showing a pattern of spontaneous fluctuation in blood oxygen level-dependent signal that correlated with the pattern of activity in the precuneus. Results: The analysis of the first fMRI dataset revealed a pattern of functional connectivity limited to two small regions in the left parietal lobe (Brodmann areas 7 and 40). The analysis of the second fMRI acquisition revealed a restored pattern of functional connectivity comprising most of the areas reported as being part of the default mode network: precuneus and left parietal regions, bilateral dorso-lateral and medial prefrontal regions, bilateral medial temporal regions as well as the left thalamus. Conclusions: Our results give further evidence of the alteration of the cortical and thalamo-cortical connectivity in VS patients and the importance of the functional restoration of this network for the emergence of the consciousness.

Out of the Body, Out of the Mind? Embodiment, Experience and Sensory Substitution Research

Joerg Fingerhut, Institut für Philosophie, Humboldt-Universität zu Berlin

Saturday, June 6th, 17:30–19:30: Poster No. 43

The contribution of the body for cognition can only be understood by assessing the organism's dynamic relations to the 'affordable' patterns of its environment. In my presentation, I propose a 'constitutive embodiment thesis' that focuses on the role of goal-driven bodily motor engagement in setting up perceptual experience. My thesis can be separated from a weaker thesis that bodily mediation merely causally triggers conscious episodes. On my account, the body is defined as enabling specific feedback loops that are constitutive for the qualitative experience an organism undergoes in establishing 'ecological control' (Clark 2008). At the heart of an enactive/embodied account of perceptual consciousness lies the claim that the intrinsic character of neurological events does not provide the means to explain experiential quality: "What determines and controls the character of conscious experience is not the associated neural activity."(No. 2009) Rather, its proponents refer to the larger setting in which those experiences occur and take seriously the coupling of the organism with the environment. Thus the alterable machinery of conscious experience is extended to bodily and environmental elements. On the basis of this assumption I oppose claims (e.g. Prinz 2009) that so far no elements outside the brain have been found that "co-vary with conscious states in content and time-course". Such claims, I argue, in limiting the scope of analysis to 'internal' events merely beg the question. In order to defend the 'constititutive embodiment thesis' and to establish a notion of sensorimotor identity, I draw on research on sensory substitution (Bach y Rita & Kercel 2003). This research supports the claim that goal-driven motor engagement constitutes specific conscious experience. Furthermore, it allows us to detach contributions of bodily and worldly elements. This, I argue, can be established by focusing on the organism's mechanisms of regulating the boundary conditions and by examining measurable shifts between the position of visual and auditory ego-centers. Evaluating new research on (minimal) TVSS and experiments done with a tactile third-eye positioned at the hand can help us to understand the specific role of the body in determining the intermodal differences and intramodal changes in perceptual experience.

Mindwandering Under Load

Sophie Forster, University College London
Nilli Lavie, University College London

Sunday, June 7th, 17:30–19:30: Poster No. 14

Perceptual load has been established as a key determinant of the processing of task-irrelevant external information (e.g., Lavie, 2005). Here we present a series of experiments examining the effect of perceptual load on the processing of task-irrelevant internally-generated information: task-unrelated thoughts (TUTs or “mind-wandering”). The frequency with which TUTs were reported during a visual-search task decreased as the task's perceptual load was increased. These effects were demonstrated on unintentional TUTs and in paradigms ruling out alternative accounts in terms of increased motivation or demands on responses or verbal working memory. In addition, a correlation was found between individual differences in load effects on internal (TUTs) and external (response-competition) distractors. These results suggest that exhausting attentional capacity in task-relevant processing under high perceptual load can reduce processing of task-irrelevant information from external and internal sources alike.

Learning to Be Social Through Dynamic Systems Control

Ana Franco, Universitè Libre de Bruxelles
Bruno Berberian, Universitè Libre de Bruxelles
Axel Cleeremans, Universitè Libre de Bruxelles
Bert Timmermans, Universitè Libre de Bruxelles

Saturday, June 6th, 17:30–19:30: Poster No. 117

This research aims to demonstrate how our capacity at social interaction can be acquired over time through an implicit process of reinforcement learning. As far back as 1954, Rotter wrote that the ''consequences of one«s behavior have an impact on one's motivation to engage in that behavior - we perform certain social actions that we hope will in turn cause other people to act in a desired way, such as smiling in order to get a smile in return.'' Recently, Wolpert & al. (2000, 2003) coined this process as a Social Feedback Loop, whereby an agent uses social feedback from the environment to adjust its internal forward model. Indeed, for a newborn, learning social interaction essentially comes down to learning links between its own behavior and the behavior of the mother in reaction to this, which essentially is reinforcement learning. Therefore we use the Dynamic Systems Control (DSC) paradigm, which is a form of reinforcement learning. In DSC tasks, participants manipulate one or more input values over a number of trials in order to obtain a target outcome value, while on each trial receiving (through the output) feedback on the efficacy of their manipulations. Berry & Broadbent (1984) used DSC to demonstrate a dissociation between performance and verbalizable knowledge, as participants typically optimize their task performance without being able to explain how they did so. In a number of experiments based on Berry & Broadbent's ''person interaction'' DSC paradigm, participants have to keep an artificial agent in a desired emotional state (e.g., ''friendly''), by providing inputs and observing their effect on the agent. The first experiment, in which participants provided emotional states as input, showed that participants indeed learn how to interact with an artificial agent. However, this paradigm didn't allow conclusions about whether learning was implicit. A new experiment disentangles implicit and explicit contributions by means of Process Dissociation Procedure (Jacoby, 1992), and renders learning more complex and ecologically valid by having participants learn how initially neutral actions, such as throwing a ball, influence the agent's emotional state. Results of this last experiment will be presented at the conference.

Computational Nonlinear Dynamics Model of Cognitive Bistability for Simulating Interrupted Stimulus

Norbert Fürstenau, German Aerospace Center

Sunday, June 7th, 17:30–19:30: Poster No. 98

Computational Nonlinear Dynamics Model of Cognitive Bistability for Simulating Interrupted Stimulus and Long Range Correlations. Simulation results of bistable perception due to ambiguous visual stimuli are presented which are obtained with a nonlinear dynamics model using delayed perception-attention-memory (PAM) coupling. Like the related Synergetic model of Ditzinger & Haken (1989) the present approach is based on attention fatigue within a perception-attention loop. In addition, however it includes a feedback delay time of 40 ms in agreement with Lamme(2003) in order to obtain estimates of absolute time scales. Percept reversals are induced by attention fatigue with an adaptive attention bias which balances the relative percept duration. The recursive PAM equations may be mapped to corresponding reentrant Thalamo-Cortical-Hippocampal loops representing the attentional top-down modulation according to Itty & Koch(2001) and memory effects. Periodic stimulus simulations as a function of stimulus off-time yields the reversal rate variation in surprisingly good quantitative agreement with classical experimental results of Orbach (1966) when selecting a fatigue time constant of 1 - 2 s. Coupling of the attention bias to the perception state introduces memory effects if the bias time constant < 20 s. They are quantified through the Hurst parameter H, exhibiting significant long range correlations (H > 0.5), i.e.the fractal character of the reversal time series (Fürstenau 2007, 2009) in agreement with recent experimental results of Gao Transition times of 150 - 200 ms between conscious perception states, mean percept dwell times of 3 - 5 s and the dwell time statistics (standard deviation/mean ca. = 0.5) exhibit good agreement with experimental values reported in the literature (e.g. Lamme(2003), Borsellino, Levelt (1967), Ito

How Unconscious Effect Information Modulates the Sense of Agency: an ERP Study

Antje Gentsch, Berlin School of Mind and Brain, Humboldt-Universität zu Berlin
Simone Schütz-Bosbach, MPI for Human Cognitive and Brain Sciences

Saturday, June 6th, 17:30–19:30: Poster No. 75

The sense of agency refers to the experience of control over one's actions and their sensory effects and is assumed to be generated by a comparison between predicted and actual action effects. It has been suggested that an internal forward model computes these predictions on the basis of efferent information which can be used to attenuate sensory effects of self-produced movements and in this way distinguish them from externally generated effects. On another view we infer the experience of control from observing covariances between thoughts and sensory events and independent of the efference copy of the motor command. Even though there is evidence for both accounts, their interrelation is so far poorly understood. We aimed at studying whether externally induced modulation of the experience of control over sensory effects does also rely on a sensory attenuation mechanism even though the effect representation was in fact independent of the motor system's execution commands. To this end we used event-related potentials to measure attenuation of self- versus externally generated sensory effects under different conditions of unconscious effect-priming. First results will be presented and their implications for the motor and inferential accounts of voluntary control will be discussed.

Causal Exclusion and Consciousness

Seli George, City University of New York Graduate Center

Sunday, June 7th, 17:30–19:30: Poster No. 38

State consciousness is that property in virtue of which a mental state, as opposed to a creature, is conscious. If it is not epiphenomenal, the property causally impacts cognition and behavior. There are many questions as to the effects the property might have, the means by which it has those effects, the cases in which it facilitates mental or physical performance, and so on. Yet a theory of ''conscious causation'' may well be a nonstarter if mental states themselves are causally inert: How can consciousness make mental processing more (or less) efficacious if it is not mental processing, but *neural* processing, that does the causal work? In this paper I examine the issue of conscious efficacy in the context of a more fundamental problem for mental causation known as ''causal exclusion''. Essentially, the generally accepted notion that physical events are part of a causally closed system entails that every neural event has a sufficient physical cause. So, barring overdetermination, mental events are unable to cause neural events. This result is a step in Jaegwon Kim's Causal Exclusion Argument (CEA). I argue that CEA poses a problem for conscious causation, and furthermore that proposed solutions to it do not necessarily secure the efficacy of state consciousness. For even if conscious mental events can be plausibly ''included'' in the etiology of neural events and behavior, their efficacy *qua being conscious* is not entailed. Thus, epiphenomenalism about consciousness can prove a more difficult thesis to debunk than epiphenomenalism about mental states in general. My procedure is as follows: First, I explain how the qua problem arises on an ontology that countenances events as causal relata. Next, I discuss CEA and various ways of preserving mental causation in spite of it, showing how these counterarguments need not establish that state consciousness is efficacious. Lastly, I argue that the qua problem obtains only on a certain kind of theory of consciousness, namely, where the property is construed as *intrinsic* to the mental event that instantiates it.

What Bistability Reveals About the Neural Basis of Perceptual Experience

Guido Gigante, Istituto Superiore di Saniti, Rome, Italy
Maurizio Mattia, Istituto Superiore di Saniti, Rome, Italy
Jochen Braun, Cognitive Biology, Otto-von-Guericke University
Paolo Del Giudice, Istituto Superiore di Saniti, Rome, Italy

Saturday, June 6th, 17:30–19:30: Poster No. 27

The instability of perception is one of the oldest puzzles in neuroscience. When visual stimulation is even slightly ambiguous, perceptual experience fails to stabilize and alternates perpetually between distinct states. The dynamics of this 'bistable perception’' have been studied extensively for decades. We have shown that many aspects of this dynamics are explained naturally by assuming a competitive dynamics among multiple neural populations that are independently meta-stable. Alternative models of bistable perception postulate a single recurrent network with (possibly noisy) oscillatory behavior. The advantage of multiple meta-stable populations is that their collective dynamics is decoupled from single neuron properties. This explains why the mean alternation rates of bistable phenomena range over two orders of magnitude, while the statistical variability of successive dominance periods (i.e., the coefficient of variation) remains essentially the same. We further propose that both competing visual inputs and alternative perceptual states are represented in a distributed manner through multiple meta-stable populations. This two-level representation accounts economically for memory effects such as the stabilizing influence of past perceptual states, which becomes particularly evident in intermittent displays. If perceptual experience is indeed based on multiple meta-stable populations, there must be a mechanism for ''reading out'' the collective activity of these populations and to thereby solve the so-called 'binding problem'. While visual memory could provide such a ''read-out'' for familiar visual scenes, unfamiliar scenes would require more generic mechanisms such as a saliency map. Accordingly, a theoretical analysis of the dynamics of bistable perception may offer an new perspective onto the nature of perceptual representations and the mechanisms of perceptual binding. Our results complement recent efforts to identify the neural correlates of perceptual experience with physiological approaches. By characterizing the distributed representation that underlies such experience, our analysis bears directly on the interpretation of results of multi-unit recording or functional brain imaging.

Anomalies of Temporal Perception: Evidence from Peri-saccadic Temporal Order Judgements

Francesco Giorlando, University of Cambridge
Roger Carpenter, University of Cambridge

Saturday, June 6th, 17:30–19:30: Poster No. 29

An essential part of the phenomenon of consciousness is the individual's awareness of time. However, a detailed theory of the neurophysiological mechanisms underlying temporal awareness remains elusive. Various alterations of temporal awareness occur in altered states of consciousness and are associated with psychopathology, it is estimated that 40–80% of psychiatric inpatients experience secondary comorbid dissociation which for some patients includes alterations in the ordering and flow of events. The authors present evidence of systematic inversion of temporal order judgements when flashed visual stimuli are presented up to 100 msec prior to a saccade. This anomaly of temporal perception was first reported by Morrone et al. 2005. In the current experiments, observers report ordering via a binary forced choice paradigm with randomised presentation of stimuli. As the onset and inter-stimulus interval is altered with reference to saccadic timing, inversions of perception on up to 80% of trials are reproducibly achieved, despite high confidence of judgements. The authors descibe these findings, characterising the typical extent of the effect as well inter-observer variation. A physiological model of the effect using the LATER model provides a description of how the phenomenon may arise from well understood neurophysiological mechanisms. In this model, the temporal inversion arises due to delays in the rise to activation of the percepts that are dependent on their relationship to saccadic timing. The findings are integrated with literature regarding similar temporal dynamics of activation in MT area in macaques (Ibbotson et al. 2006). Further characterisation of the effect with modulation by psychotropics is described (corticosteroids and ketamine). These anolmalies assist our understanding of the underlying processes leading to integrated temporal awareness and can act as a metric for comparing theories of how temporal estimation and ordering of events arise and what neural correlates are involved. The implications of the current research for models including oscillator-accumulator models and state dependent timing networks are discussed as well as implications for the alterations in temporal awareness that occur in psychosis and dissociation.

Internal Consciousness in Very Young Children: Memory, Planning, Self and the Babbling Stream

Alison Gopnik, UC Berkeley

Sunday, June 7th, 17:30–19:30: Poster No. 62

What is it like to be a baby or a very young child? In adults, conscious awareness is often organized in terms of a single, unified, coherent time-line. Adults conceive of a self, an inner observer, autobiographer and CEO who has been the subject of past experiences, as in autobiographical memory, and will be the subject of future experiences, as in prospective memory and planning. This unified, reflective, temporally coherent self has been identified with various types of frontal function and has sometimes been seen as a requirement for consciousness itself. I outline evidence that babies and young children do not have the same sort of coherent time-line. Both autobiographical and prospective memory appear to develop in the preschool years, and even three-year-olds fail to recognize their own past experiences or anticipate future ones. Nevertheless, children clearly have internal experience, though this experience is organized differently than it is in adults. I suggest that the phenomenology of “free association, hypnagogic sleep, and certain types of ''insight'' meditation may give us an adult approximation of preschool experience. Moreover, I suggest that these differing types of phenomenology are associated with different functions. For adults, the central cognitive agenda involves planning and executing future action, for babies and young children it involves exploring the possible ways that the world might be. These developmental data join several other recent developments which suggest that we can dissociate consciousness itself from self-conscious frontal functions.

Finding Correlations Between Subjective and Objective Measures of Awareness Using Masked Words.

Anastasia Gorbunova, University of Arizona

Sunday, June 7th, 17:30–19:30: Poster No. 12

The current study explores the idea proposed by Marcel (1983) that perceptual analysis can be performed in the absence of conscious awareness. In this experiment, participants were asked to perform two forced choice tasks based on the information gathered from a masked word presented for 50 milliseconds: (1) semantic discrimination in which participants had to choose between a word that was associated in meaning with the masked stimulus and an unrelated alternative, and (2) e-detection where participants responded yes if the masked word contained the letter e, or no if it did not. Performance on both tasks was better than chance, but performance on e-detection (mean % correct 67.2) was slightly better than on semantic discrimination (mean % correct 61.9). This provided an objective measure of partial awareness of the stimuli. For both tasks, participants were asked to rate their confidence after each trial. Confidence ratings ranged from 1 (complete guess), to 5 (identified the masked word), and provided a measure of subjective awareness. Pearson correlations were calculated to find that highly rated items generally constituted more correct responses than low rated ones (r = 0.549 for semantic discrimination, and r = 0.365 for e-detection). However, these correlations were far from perfect, suggesting that some information is gathered from masked words even when the participants feel that they are simply guessing. Possible mechanisms by which these tasks can be performed when confidence ratings are less than 4 and implications of these findings are discussed.

Abnormal Short Latency Afferent Inhibition in Disorders of Consciousness: Preliminary Findings

Olivia Gosseries, Coma Science Group, Cyclotron Research Center, University of liège
Natallia Lapitskaya, Coma Science Group, University of liège
Olivia Gosseries, Coma Science Group, University of liège
Victor De Pasqua, Centre Hospitalier Regional de la Citadelle
Alain Maertens de Noordhout, Centre Hospitalier Regional de la Citadelle
Jürgen Feldbek Nielsen, Hammel Neurorehabilitation and Research Centre
Gustave Moonen, University of liège
Steven Laureys, Cyclotron Research Centre, liège, Belgium

Sunday, June 7th, 17:30–19:30: Poster No. 128

Objectives: Peripheral input may have an inhibitory effect on motor cortex at short intervals (short latency afferent inhibition, SAI), thought to reflect cholinergic cortical modulation. SAI can be assessed by coupling electrical stimulation of the peripheral nerve with transcranial magnetic stimulation (TMS) of the motor cortex. We here evaluated SAI in patients with disorders of consciousness. Methods: Seven patients (mean 45 years, range 23-76) and nine healthy subjects (mean 32 years, range 23-59) were recruited. Patients had chronic brain injury (mean interval after injury 6 months, range 1-12) of traumatic (n=5) and non-traumatic (n=2) aetiology and met the criteria defining vegetative (n=2) or minimally consciousness state (n=5). SAI of the motor cortex was studied using a paired-pulse stimulation technique: conditioning electrical stimuli were applied to the median nerve at the wrist; test TMS stimuli were applied to the motor cortex. Interstimulus intervals (ISI) were determined relative to the latency of the N20 component of the somatosensory evoked potentials. We investigated ISIs minus 2 ms and plus 14 ms of the obtained N20 component latency (in steps of 2 ms). The amplitude of the conditioned motor evoked potentials (MEP) was expressed as the percentage of the amplitude of the unconditioned MEP. Results: In one vegetative patient neither sensory nor MEPs could be elicited. In healthy subjects the most prominent SAI was observed at the ISI of N20 plus 2 ms. In patients no inhibition was observed at this ISI compared with controls. Patients also tended to have a high resting motor threshold and less pronounced inhibition at other ISIs, but these differences were not significant. Conclusion: The assessment of motor cortical excitability in vegetative and minimally conscious patients may offer a better understanding of their underlying disordered cortical excitability. The presented pilot data suggest that SAI of the motor cortex, a putative marker of cholinergic cortical activity, is significantly reduced in patients suffering from chronic disorders of consciousness.

Abstraction of Action: The Involvement of the Right Inferior Frontal Gyrus in Action Interpretation

Pär Halje, Laboratory of Cognitive Neuroscience, EPFL
Silvio Ionta, LNCO-BMI-EPFL
Laurent Spinelli, University Hospital of Geneva
Margitta Seeck,
Olaf Blanke, Brain Mind Institute, EPFL, Switzerland

Sunday, June 7th, 17:30–19:30: Poster No. 78

Plenty of evidence point to the existence of shared brain mechanisms for language and motor function. This has led to speculations about whether the ability of the human brain to use abstractions has evolved from pre-existing mechanisms dedicated to motor functions, such as action imitation and interpretation. With data from a patient with intracranial electrodes we show that neural processes in the right inferior frontal gyrus (rIFG) distinguished between verbal descriptions of actions and non-actions ("I take the orange" vs. "It«s a sunny day"). An overlapping peri-sylvian area distinguished between the observation of meaningful and meaningless acts (grabbing an object vs. grabbing nothing). This was quantified with a linear classifier operating on the power spectrum of the local field potential. The classifier was used to naively probe the data for task-relevant sites, frequency bands and time periods. A classification performance of 93 % (cross-validated) was found in the upper beta band (24-28 Hz) for the verbal task. Classification between two non-action sentences at failed. For the observation task (meaningful vs. meaningless actions) the classification performance was 89 %, also in the upper beta band. These results show that the rIFG in our patient was involved in verbal description of actions as well as the interpretation of observed actions. This suggests that the rIFG supports a specific mechanism for representing and interpreting actions.

Why Make It Conscious? The Function of Consciousness in Therapeutic Change

Ida Hallgren Carlson, University of Gothenburg

Sunday, June 7th, 17:30–19:30: Poster No. 124

The function(s) of consciousness discussed in an evolutionary and cognitive/neuroscientific framework suggest that consciousness plays a crucial role in the comprehension of new information, in the learning of new tasks as well as allowing for flexible and non-automatic (voluntary) behavior. Consciousness has been suggested to involve large areas of the brain and may have an integrative function. Applied to theories about therapeutic work these suggestions, summarized by Baars (2002), call for some modifications of claims (or the lack thereof) made about the role of consciousness within different theories on therapy. Therapies influenced by traditional psychoanalytic theories do not seem to differentiate between conscious and self-conscious processes and hence give little attention to the role of the conscious information-processing that is never verbalized or made self-conscious. Consciousness research may also suggest that unconscious processes or material per se must not necessarily be made conscious for change to occur. Therapies stemming from the behaviorist tradition tend to leave out the role of consciousness altogether and do not allow for differences being made between simple fear-conditioning, that may take place without conscious processing, and the learning of more complex tasks that may involve relearning or restructuring of cognitions. Recent mindfulness-based therapies allow for a greater focus on conscious processing but lack clear definitions and do not make great efforts to explain what the functional role of ''being present'' or conscious might be. Many therapeutic traditions, e.g. anxiety-provoking short-term therapies, have given much weight to anxiety occurring in change processes, but have not made clear the causal role of anxiety. This paper suggests that the function of anxiety is to trigger and maintain a high degree of conscious processing that provides a unique capacity for integrating information in relearning. Further, the seemingly contradictory anxiety-based and mindfulness-based therapies are suggested to have a common aim, i.e. to give rise to conscious processing. However, anxiety might be an unfortunate but necessary ingredient where cognitive restructuring is necessary to allow for changed behavior. The theory of cognitive dissonance, showing how action-related inconsistencies provokes anxiety, might be of particular importance in this context.

Predictability of Free Decisions: Cognitive Load Limits Our Ability to Make Unpredictable Choices

Jakob Heinzle, BCCN, Charitè - Universitätsmedizin
Tatiana Usnich, BCCN, Charite - Universitätsmedizin
John-Dylan Haynes, BCCN, Charitè - Universitätsmedizin Berlin

Sunday, June 7th, 17:30–19:30: Poster No. 76

In certain situations humans can have the impression to be ''randomly'' choosing between several equal options. But previous research has suggested that humans find it very difficult to deliberately produce random behavior. Instead their choices in such unconstrained ''free choices'' follow characteristic patterns. In this study we were interested in cognitive influences on random behavior. We conducted a series of experiments to evaluate the randomness of choices and how this randomness was affected by a cognitive intervention that distracted subjects between successive choices. 24 healthy subjects generated random sequences of binary choices without any previous explanation of a mathematical definition of randomness. In each of 4 experiments the subjects had to select hundred times one of two identical squares (right or left). Truly random sequences are defined as there being no relation at all between successive choices. In order to manipulate the interaction between individual choices, we changed the delay duration and a task that had to be performed during the delay between subsequent selections (short pause, long pause, verbal task and calculation task). The sequences produced were tested for randomness by comparing the frequency of occurrence of all possible short sequences of up to 4 choices with the theoretical limit. These deviations were compared to a distribution defined by Monte-Carlo Simulation of the same task, which allowed us to compute the probability of getting an at least as high deviation randomly. The influence of the cognitive tasks was verified using paired Kolmogorov-Smirnov test on the distances and on the probabilities of being random. Only one third (32 of 96) of all sequences passed our test for randomness, i.e. could not be clearly distinguished from a real random sequence. Nevertheless, there was a significant difference in how random the subjects performed between the experiments with short pause without extra task (12 of 24) and the calculation task (4 of 24). Although subjects did try to perform the same random selection task in both conditions, a difficult task inserted between two choices corrupted their ability of behaving unpredictable. Thus, it seems that unpredictable behavior might require conscious deliberation.

Shared Temporal Accuracy of Action Execution and Sensory Perception

Tomomitsu Herai, Tokyo Institute of Technology
Ken Mogi, Sony Computer Science Laboratories

Saturday, June 6th, 17:30–19:30: Poster No. 73

Integration of an action and its sensory feedback is important in interacting with an uncertain environment and constructing a consistent model of the world. Studies have shown that the sensorimotor contingency affects our perceived timing of own action and its sensory feedback (Haggard et al., 2002; Tsakiris & Haggard, 2003, Engbert & Wohlschläger, 2007; Moore & Haggard, 2008). In the course of the flexible interaction with the environment, multisensory data need to be processed, in which audition and vision play important roles. The brain employs a mechanism which modulates the perceived timing of audiovisual stimuli depending on the distance between the perceiver and the source (Kopinska & Harris, 2004). In addition, the subjective simultaneity of audiovisual stimuli is affected by adaptation ( Fujisaki et al., 2004), spatial position (Zampini et al., 2003; Zampini et al., 2005) and attention (Zampini et al., 2005). Perception and action both occur in the stream of subjective time, characterized by respective temporal properties. To investigate the relation between the subjective simultaneity of audiovisual stimuli and action, we conducted an experiment in which the subjects’ actions affected the temporal patterns of resulting stimuli. Subjects pressed two keys simultaneously with the index fingers of their both hands. One of the keys generated beep, while the other generated flash. The delay between the generated beep and flash depended on the accuracy of the action of key pressing. The task of the subject was to press the two keys simultaneously and judge the simultaneity of beep and flash that follow. The modes of contingencies between action and stimuli were made variable by changing the relationship between the key pressings and stimuli. We found significant correlations between the accuracies of actions and the ''window'' of subjective simultaneity among subjects, although their task performances were widely varied. In addition, the correlation patterns were found to depend on the contingency between the key pressing and stimuli. These results suggest that the subjective simultaneity of audiovisual stimuli correlates with the accuracy of execution of action, indicating a common mechanism engaging the perception of subjective simultaneity in sensorimotor integration and action execution.

Phenomenal Variability and Introspective Reliability

Jakob Hohwy, Monash University

Saturday, June 6th, 17:30–19:30: Poster No. 41

Introspection is widely thought to be a very reliable guide to our own phenomenology. Recently, Eric Schwitzgebel has provided convincing and surprising evidence that the deliverances of introspection are uncertain and variable. He argues that the best explanation of this evidence is the unreliability of introspection in the domains of emotions and sensations, imagery, visual perception and thought. I discuss how to respond to this challenge to introspection and I use recent cognitive theory and neurobiological findings to suggest an alternative and better explanation of the evidence. On this explanation, it is mostly the phenomenal states themselves that vary, either chronically, or in testing conditions. I demonstrate this in different ways for the cases of emotion and sensation, imagery, visual perception, and thought. Thus the surprising evidence can be explained while allowing introspection to be reliable in most conditions. This application of these cognitive theories and empirical findings provides a novel perspective on introspection of phenomenal states.

Temporo-Parietal Cortex and Precuneus Encode Bodily Self-Location: Joining Robotics and fMRI

Silvio Ionta, LNCO-BMI-EPFL
Bigna Lenggenhager, Brain Mind Institute, EPFL, Switzerland
Michael Mouthon, LNCO-BMI-EPFL
Dominique Chapuis, Robotic Systems Laboratory-EPFL
Roger Gassert, Robotic Systems Laboratory-EPFL
Olaf Blanke, Brain Mind Institute, EPFL, Switzerland

Sunday, June 7th, 17:30–19:30: Poster No. 92

Aspects of bodily self-consciousness such as the conscious experience of where the self is experienced to be (self-location) can be experimentally manipulated by exposing people to conflicting multisensory input. Applying such conflict for body parts (e.g. rubber hand illusion) to the full body, recent experiments in virtual reality suggest that it can be extended to global aspects of self-consciousness (Ehrsson, 2007; Lenggenhager et al., 2007). Premotor, somatosensory, parietal areas and insula are involved in the localization of the illusory body part, but the neural underpinnings of more global illusions are yet to be investigated. Data from neurological patients with disturbed global self processes suggest that the temporo-parietal junction (TPJ) plays an important role in this localization process (Blanke et al., 2004). In the present study, we used fMRI and neuroscience robotics to measure the brain activity during experimentally induced changes in illusory global self-location. Subjects saw a video of a stranger's back or an empty room being stroked (visual input) while a robotic device stroked their back (tactile input). Direction and speed of the robotic stroking corresponded to (synchronous) or differed from (asynchronous) that of the video. Each block was composed of videoclips of one of the four conditions according to a 2x2 factorial design with Object (Body, No Body) and Synchrony (Synchronous, Asynchronous) as main factors. To evaluate the self localization, subjects imagined the falling time of a ball they were holding in their hand to the ground, which they indicated by pressing a button (Lenggenhager et al., in press). With the body videos, response times for the mental ball dropping were significantly longer in the synchronous with respect to the asynchronous stroking condition, suggesting an elevation in self-localization. With ''no-body'' videos there was no difference between the synchronous and asynchronous conditions. FMRI results showed bilateral activation of TPJ and precuneus with a significantly higher BOLD signal increase in the Synchronous/Body condition with respect to the other conditions. Other activated areas included sensorimotor and supplementary motor areas. These finding suggest an involvement of TPJ and precuneus in the experience of the conscious ''I'' as embodied and localized in space.

Do Dissociations Work?

Elizabeth Irvine, University of Edinburgh

Sunday, June 7th, 17:30–19:30: Poster No. 30

The basic method used to investigate the boundaries of conscious and unconscious perception is to find dissociations. While single dissociations are typically used to infer the existence of unconscious perception, this research is liable to the problems that face any use of dissociations (Reingold & Merikle, 1988, 1990). In response to this, Snodgrass et al. (2004) argue that there is evidence of a reversed association between direct and indirect measures of consciousness which avoids these problems (Dunn and Kirsner, 1988). However, there are major problems with Snodgrass et al.'s interpretation of the reversed association. Snodgrass et al.'s model of functionally exclusive (inhibitory) sets of conscious and unconscious perceptual influences is not a model that always holds (in contrast to a model involving functional independence), and the relation of inhibition between perceptual influences necessitates that they are dependent, which contradicts the standard inference of process/system independence. A further problem arises in the use of perceptual 'influences' as effects or end products of perception as the model is essentially a restatement of the empirical findings and adds little of theoretical value. Further problems arise in interpreting dissociations as they provide evidence from which to infer the existence of two (or more) functionally independent systems/modules, yet it is generally assumed that conscious perception is dependent on unconscious perception. The terms 'conscious processing' and 'conscious perception' are often used inappropriately as they attribute consciousness to a process or completion of a task, instead of to a state of a subject. Also, dissociations play a strange role in consciousness research as the categories to be dissociated are pre-defined and are assumed to exist above and below specific thresholds, so empirical results often do not guide theorising. Establishing dissociations in consciousness research is made more difficult as conscious and unconscious perception often rely on the same structures, brain areas, and often perform the same functions. It therefore appears that dissociation logic is of limited use in the context of unconscious perception research, and some implications for the science of consciousness will be drawn.

Trajectory Analysis of Search Behavior in Visual One-Shot Learning

Tetsuo Ishikawa, Sony Computer Science Laboratories, Tokyo Institute of Technology
Ken Mogi, Sony CSL

Saturday, June 6th, 17:30–19:30: Poster No. 119

Among the varied phases of human cognition, the very moment when an idea flashes into one's mind is one of the most dramatic instances of conscious experience. Throughout human history, there has been a rich accumulation of episodes of such insightful moments. For instance, the anecdote of Archimedes exclaiming ''Eureka!'' while he was taking a bath and found a new principle is well known (Vitruvius, ca. 27 B.C.). These are special cases of learning where the learning process occurs suddenly and is completed in a very short time (one-shot learning). In the field of visual perception, the perception of ''Mooney'' faces (Mooney, 1957), i.e high-contrast monochrome pictures of a human face, is a fascinating example of visual one-shot learning. When the subjects realize what is in the figure, synchronous activities have been found to spread over a wide area in the brain which correlate with the subject's perception (Varela et al, 1999). From a more general perspective, various kinds of objects can be hidden in bi-level quantised images in a similar way. Famous examples such as ''Dallenbach's cow'' (Dallenbach, 1951) and ''the Dalmatian'' (Gregory, 1970) are real puzzlers where subjects find it difficult to see what is in the figure. Here we study the cognitive process of visual one-shot learning in the interactive domain. The stimuli were two dimensional representations of 3-D objects designed in such a way that they were difficult to perceive from some angles, while easier to perceive from others. The subjects were presented with images from several different angles, and were asked to rotate them and search for the ''correct'' direction. By comparing with control tasks in which the same hidden figures were presented from the optimum direction, we clarify the importance of the intentional actions to get the epiphany. Analysis of temporalities involved reveals the nature of underlying dynamics of the phenomenon. We discuss the data in the context of hidden figure perception as a matching between the top-down and bottom-up processes, and explore the implications for the cognitive processes in general, conscious or otherwise.

Phenomenological Intentionality and Success

Blank Jared, CUNY Graduate Center

Saturday, June 6th, 17:30–19:30: Poster No. 39

There has been a recent growth in interest in developing a theory of intentionality that derives from phenomenological features of experience. We'll call such a view Phenomenological Theory of Intentionality (PI). PI theory represents a radical shift in the philosophical attitude to theorizing both about the nature of intentionality and consciousness. The usual approach is to divide and conquer each subject separately or, as representationalists do, to derive properties of consciousness from properties about representation. Theories concerning the fundamental representational properties of mental states have been pursued in terms of reductive naturalistic theories that posit some kind of naturalistically kosher relation between thinker and environment. PI theorist subvert this latter approach and attempt to give a derive a notion of intentionality, the so-called aboutness of mental states, out of a notion of phenomenology which concerns subjectively felt and intrinsically specified aspects of experience. Though the PI theorist has attempted to account for traditional features of mental representations, like reference and the contribution external factors play in the determination content, it's unclear whether she can genuinely explain them. The objection I want to press is in this vein but concerns a somewhat deeper explanatory goal of the theory of intentionality, one that, I claim, PI theorists lose sight of: they cannot account for success. One of the fundamental reasons we're concerned with intentionality is that it explains the coordination between the content of our beliefs and desires: true belief generally lead to the satisfaction of our desires. I argue that PI theorists simply don't have the resources to account for this basic explanatory goal of the theory of intentionality.

Conscious Expectancies vs. Automatic Adjustments in a Stroop Task

Luis Jimenez, University of Santiago
Amavia Mendez, University of Santiago

Sunday, June 7th, 17:30–19:30: Poster No. 122

Congruency effects such as those found in a typical Stroop task are subject to a sequential modulation: they are larger after a congruent trial than after an incongruent trial. Egner (2007) has recently pointed to three potential factors affecting these sequence congruency effects: (1) conflict-driven adjustments in cognitive control, (2) episodic repetition effects, and (3) conscious expectancies. Although there is some evidence indicating that only consciously detected conflicts may result in an adaptation effect over the next trial (Kunde, 2003), other results suggest that control can be exerted without a conscious feeling of effort (Naccache et al. 2005). In the present study, we adapt a procedure first developed by Perruchet et al. (2006) to dissociate automatic vs. strategic effects, and we apply this method to dissociate the effects of adaptation to conflict from those of conscious expectancies. By assessing expectancies after several runs of congruent or incongruent trials, we found that participants expectancy follows the gambler fallacy, thus expecting a change after a large run of either congruent or incongruent trials. However, even though participants declare to be expecting an incongruent trial after a series of congruent trials, yet the difference in RT in favor of congruent trials yields its maximum value in this case, as predicted by an automatic conflict-adaptation account. In contrast, after a series of incongruent trials, when participants declare to be expecting the next one to be a congruent trial, responding to another incongruent trial results in only a minimum cost. In a series of follow-up experiments we manipulated the proportion of transitions between congruent and incongruent trials, by favoring either the repetitions of the same type of trial of the alternation between them. We found that these manipulations do not alter the conflict adaptation effects as observed in RT performance, but they alter the expectancies in accordance with the training schedule. These results are interpreted as showing that even though conscious expectancies can arise from training in this task, conflict-driven adjustments are better thought of as adjustments produced automatically in response to previous trials.

The Forward-Looking Nature of Consciousness as a Basis for Embodied Communication

J. Scott Jordan, Illinois State University

Saturday, June 6th, 17:30–19:30: Poster No. 91

Recent research indicates (1) actions are consciously planned in terms of the distal effects they are to produce, and (2) planning and perception share common neural resources. As a result, perceiving the effects of an action (e.g., hearing the tones produced by striking a piano key) activates the pre-motor cortical centers one would use while consciously planning to produce such effects oneself. In short, conscious perception takes place within an intentional context that emerges from the action-effect contingencies one learns over the life course (Jordan, in pressa,b; Kinsbourne & Jordan, in press). Given this notion of intentional contexts, the present talk will present experiments that do the following: (1) support the assertion that perception is altered in a forward-looking manner as one learns action-effect contingencies, (2) examine the conditions that allow one to develop intentional contexts, and (3) examine how one’s own conscious perceptions are altered as one learns to cooperatively generate an intentional context with another (e.g., play a video game). The experiments to be presented examine intentional contexts in spatial perception. Specifically, research indicates participants perceive the vanishing point of a moving stimulus beyond the actual vanishing point. Jordan and Hunsinger (2008) found that those who have experience controlling the stimulus' movements perceive the stimulus to vanish further ahead than those who do not. Also, one can learn the action-effect contingencies that give rise to such forward displacements (FD) by simply observing another control the stimulus. However, larger FD only occurs if one perceives the effects of the model's actions, as well as the actions themselves. Finally, Jordan and Knoblich (2004) found that FD increases as one attempts to control the movements of the stimulus co-operatively with another participant. These data support the assertion that body consciousness and perception entail forward-looking, intentional content derived from planning. They further imply that as agents continuously generate distal effects together (e.g., play a video game), the resultant group effect (i.e., the changes in the game produced by their collective actions) becomes part of one's own action-effect contingencies (i.e., one's own intentional context) (Knoblich & Jordan, 2003).

No Virtual Mind in the Chinese Room

Christian Kaernbach, Christian-Albrechts-Universität zu Kiel

Saturday, June 6th, 17:30–19:30: Poster No. 95

The Chinese room thought experiment of John Searle militates against strong artificial intelligence, illustrating his claim that syntactical knowledge by itself is neither constitutive nor sufficient for semantic understanding as found in human minds. This thought experiment was put to a behavioural test, concerning the syntax of a finite algebraic field. Input, rules and output were presented with letters instead of numbers. The set of rules was first presented as a table but finally internalized by the participants. Quite in line with Searle's argument, uninformed participants mastered the syntax but did not explicitly report semantic knowledge. In order to test the virtual mind reply to the Chinese room argument, the reaction time pattern of the participants was compared to that of an informed control group. The correlation was quite high but could be traced back to memory load and response priming, i.e. to syntactical factors. No trace of tacit semantic knowledge of the task could be found in the experimental group.

Decoding the Expected Value of Multi-attribute Objects from the Human Brain

Thorsten Kahnt, Berlin School of Mind and Brain
John-Dylan Haynes, BCCN, Charitè - Universitätsmedizin Berlin

Sunday, June 7th, 17:30–19:30: Poster No. 80

People are permanently faced with situations in which they have to choose between different options and actions. How do people decide between different alternatives? Economic theories and reinforcement learning theory assume that subjects choose that option with the highest expected value. On the other hand, it has been suggested, that conscious deliberation or unconscious deliberation are two fundamental different modes of decision. Previous research has shown that each mode can be superior depending on the complexity of the decision problem at hand. Conscious deliberation seems to be better for decisions about simple objects whereas unconscious deliberation is more efficient for decisions about complex, multi-featured objects (Dijksterhuis et al. 2006). Here, in a first step we used fMRI and multivariate pattern analyses to uncover the representation of the expected value of complex, multi-attributed objects. Prior to scanning subjects learned the association between different attributes and reward. During fMRI acquisition subjects saw multi-attribute objects and reported the integrated value of these objects. Distributed patterns of activity in the ventromedial prefrontal cortex (VMPFC) and the amygdala predicted the integrated reward value of multi-attribute objects. Functional connectivity analyses revealed an increased functional coupling between both regions during object presentation. During the rating phase, however, functional connectivity between VMPFC and supplementary motor cortex increased relative to the stimulus phase. These results suggest that the expected value of multi-attribute objects is represented in the VMPFC and amygdala and used by the VMPFC to guide decisions. In next experiments, we plan to compare the neural representation of expected value while subjects use different decision modes to make their choice.

Conscious Monitoring of Motor Performance During Locomotion in a Virtual Reality Setting

Oliver Kannape, Laboratory of Cognitive Neuroscience, EPFL
Lars Schwabe, ARS, University of Rostock
Tadi Tej, Laboratory of Cognitive Neuroscience, EPFL
Olaf Blanke, Brain Mind Institute, EPFL, Switzerland

Sunday, June 7th, 17:30–19:30: Poster No. 64

Recent research (Lenggenhager 2007, Ehrsson 2007) suggests that awareness for one«s entire body (ownership) and ownership for one«s body parts rely on similar multisensory mechanisms. In the present setup we extended this line of research in order to investigate motor contributions to the awareness of the entire body during locomotion (motor awareness). In study 1, we asked 9 participants to walk towards 4 different target positions while their body movements were tracked via optical motion capture. Movements were mapped to a virtual body and played back, in real-time, on a projection screen. The body movement and position of the virtual character could systematically be deviated from the participants' movements by ±5°, ±10°, ±15°, or ±30°. Motor Performance (MP) and Motor Awareness (MA) were measured. Results show that participants are unaware of angular biases of ~10° (MA defined as the point of subjective ambiguity) despite participants' motor behaviour (significantly deviated walking paths in the direction opposite to the deviation; p<0.001). In study 2 (N=14) we investigated the influence of a) the virtual body's orientation (upright/inverted) and b) its walking direction (back/front) on MP and MA. We further analysed the relationship between MA and MP describing the likelihood of errors in MA with respect to MP (Motor Awareness Index, MAI). We found a significant orientation x direction interaction for MP (p=0.016). MA was at 13.8-15.8° but not significantly different across conditions (p>0.18). Analysis of the MAI revealed that participants were more likely to make MA errors in the inverted conditions (p=0.016). These data confirms that full-body MA depends on angular deviation as well as orientational incongruency (inversion) between physical and virtual body. These data illustrate that humans consciously monitor the position of their full-body and its locomotion with a high degree of uncertainty and that the degree of uncertainty can be manipulated experimentally. This extends previous data on MA for arm movements to the moving full-body that are associated with distinct functional consequences - our setup displaces the participant’s centre of conscious experience in space by up to 10º.

Predicting Visual Unconscious Processes from EEG Single Trial Analysis

Lisandro Kaunitz, CIMEC, University of Trento
Brian Murphy, CIMEC, University of Trento
Massimo Poesio, CIMEC, University of Trento
David Melcher, CIMEC, University of Trento

Sunday, June 7th, 17:30–19:30: Poster No. 134

Previous electrophysiological studies have produced evidence for specialized cortical regions involved in the processing of different object categories, specially between animals and tools (Kiefer, 2001). In all these studies the results have been obtained by averaging the physiological signals coming from many trials across subjects. To investigate the neural signatures of conscious and unconscious perceptual categories by analyzing EEG single trials we used algorithms developed for categorizing epochs based on the spectrum of the EEG signal (Dalponte, 2005), data mining techniques and a support vector machine. In the first experiment with visible (unmasked) stimuli we were able to predict what class of object-animals or tools-participants had seen on each trial with an accuracy that ranged from 73% to 86%. In the second experiment, we used Continuous Flash Suppression (CFS: Tsuchiya, 2005) to present pictures of animals or tools to our subjects while they undergo two conditions: a visible condition -in which objects are consciously perceived- and an invisible condition -in which participants are not aware of the presence nor the identity of the suppressed stimuli. Our first results show that on the conscious conditions we could correctly classify meaningful images from their scrambled counterparts (74% accuracy, n=2, p<0.001) and animals images versus tool images (59% accuracy, n=3, p<0.001). On the unconscious conditions we could separate the meaningful targets from the control images (57% accuracy, n=2, p<0.001) but we obtained no significant results for the unseen animals versus unseen tools (52% accuracy, n=3, p=0.10). The goal of this project is to accurately predict whether a subject has unconsciously seen either a meaningful image or a meaningless one by analyzing EEG single trials data. References: Dalponte, Michele (2005). Sistema innovative per l’analisi e la classificazione di segnali EEG per applicazioni di Brain-Computer interface. Tesi di Laurea. Markus Kiefer (2001). Perceptual and semantic sources of category-specific effects in object categorization: Event-related potentials during picture and word categorization. Memory and Cognition, 29(1):100–116. Tsuchiya, N., and Koch, C (2005). Continuous flash suppression reduces negative afterimages. Nat Neurosci. 8, 1096-1101.

Finding McGurk: Localisation of the Source of the McGurk-Effect and Related Oscillatory Activity

Julian Keil, University of Konstanz
Niklas Ihssen, University of Bangor
Nathan Weisz, University of Konstanz

Saturday, June 6th, 17:30–19:30: Poster No. 85

The McGurk-effect is a perceptual illusion, which demonstrates an interaction between auditory and visual sensory systems in the perception of speech. If a mismatch between the perception of a sound and the accompanying visual input occurs, it has been observed that the unified perception of both modalities fuses to a novel percept that neither exactly matches the sound nor the sight. A number of studies reported, that this illusion does not appear in every trial, but rather in a proportion of trials (~60–70%). The present study was designed to shed light onto the conditions under which this effect occurs and the identification of cerebral sources associated with it. Specifically we are interested in the potential influence of ongoing ("background") brain oscillations associated with varying perception. In the present study, videos of an actor articulating a series of syllables (e.g. 'aba') were dubbed with the audio tracks of different syllables (e.g. 'aga'). The subjects thus were presented with a sequence of 390 videos, part of which contained the original audio tracks and part of which contained mismatching tracks. As an indicator of the perception, the subjects had to report via button press what syllable they had perceived. In this way, trials containing a novel perception could be contrasted with trials containing the original perception. MEG was used to record event related and oscillatory activity during this stimulation. Previous studies have shown an increase in gamma band activity related to incongruent stimuli. FMRI-studies have shown an involvement of superior temporal gyrus in speech perception. We thus firstly propose that the source of the McGurk-effect lies within the supratemporal auditory cortex and could be reflected by increased gamma oscillations. Moreover, further analysis will scrutinize the question regarding interareal synchronization particularly between auditory and visual areas. We hypothesize an increased long-range synchronization between these areas to reflect the crossmodal interaction.

Understanding Libet«s Urge

Jean-Remi King, Institute of Neurology, University College London

Sunday, June 7th, 17:30–19:30: Poster No. 74

In the famous experimental paradigm of Libet et al (1983), aiming at demonstrating the unconscious source of motor intentions, participants are asked to move their finger by letting "the urge to act appear on its own at any time". However, the motor cognition literature never explicitly investigated what this sensation might in fact reflect. In this poster, the author demonstrates the ambiguity of ''urge'' by highlighting its implicit assumptions among the addiction, tic disorder models and motor cognition literature. Two main aspects are proposed: a prediction sensation and an inclination experience. These two concepts are then considered from a phenomenological approach to a neuro-cognitive perspective. The specific roles of the Supplementary Motor Area (SMA) and the parietal cortex appear to be importantly related to the former feature, whereas the latter seems to relate to dompaminergic pathways, action selection and inhibition mechanisms, notably implemented by the prefrontal cortex and the basal ganglia. Finally several research directions are suggested, with a special emphasis on the need of cross studies comparing drug addiction, tic disorders (TD) and motor cognition neuro-physiological, cognitive and behavioural data and models.

A Solution to the Puzzle of Temporal Experience

Michal Klincewicz, Graduate Center, City Univerisity of New York

Saturday, June 6th, 17:30–19:30: Poster No. 51

The temporal dimension of conscious experience raises an important challenge. As Sean Kelly puts it, ''How is it possible for us to have experiences as of continuous, dynamic, temporally structured, unified events given that we start with (what at least seems to be) a sequence of independent and static snapshots of the world at a time?'' Kelly claims that the only significant attempt at a solution is found in the retention theory of Edmund Husserl. Husserl argues that the contents of past experiences are retained in a kind of short-term memory, where they play a role in determining the content of a present experience. Kelly faults this model on phenomenological grounds. He observes that experiences don't present the contents of past experiences. If they did, listening to a melody would produce a successively more complex chord; similarly, a visual experience of a moving object would present a trail. Since experience normally does not present things in that way, Kelly argues that the retention account must be rejected. I argue that the mechanism responsible for experiencing temporal extension is something very much like Husserl«s retention model, and that Kelly's argument rests on the controversial premise that all mental states are conscious. Husserl's notion of retention can be accounted for in terms of unconscious mental states. On the view proposed, temporal experience is comprised of (1) a mental state that's conscious in virtue of the individual's being aware of it in a suitable way and (2) the retention process, which is conscious in virtue of the individual«s having a distinct awareness of that very process. The mental states that constitute the retention process are not themselves contents of the awareness of the process. As such, the process does not contribute to the experience as one is aware of it, and so does not produce the unusual phenomenology Kelly argues the retention theory implies. Arguably, mental states are conscious in virtue of the individual having a higher-order thought about them. On this view, the mental states constituting the retention process are not contents of the relevant higher-order thoughts.

Conscious Detection Under Different Types of Load

Nikos Konstantinou, Institute of Cognitive Neuroscience, University College London
Bahador Bahrami, Institute of Cognitive Neuroscience, University College London
Jean-Remi King, Institute of Neurology, University College London
Nilli Lavie, Institute of Cognitive Neuroscience, University College London

Sunday, June 7th, 17:30–19:30: Poster No. 108

It is an established finding that high perceptual load reduces the processing of task-irrelevant stimuli. Recently, Macdonald & Lavie (2008) demonstrated that high perceptual load reduces detection sensitivity for task-irrelevant stimuli, demonstrating a phenomenon of Load Induced Blindness. Here we report a series of experiments examining the effects of visual short-term memory (VSTM) load on detection of task-irrelevant stimuli, and comparing these effects to those of perceptual load and verbal working memory load. Subjects searched for designated target letters while also asked to detect the presence or absence of a critical stimulus (CS; a contrast increment) in the periphery. This task was performed during the retention interval of either a delayed-matched-to sample VSTM task or a verbal working memory task. We varied the level of either perceptual load in the visual search task (by varying the search set size), or VSTM load or working memory load (by varying the memory set size). CS detection sensitivity (d’) was consistently reduced under high, compared to low, VSTM load and perceptual load but was unaffected by the level of working memory load. In subsequent experiments we demonstrate a significant negative correlation between memory capacity (measured with Cowan's K) and detection sensitivity. The clear trade-off between VSTM capacity and detection sensitivity indicates shared resources between VSTM and detection. We discuss these findings in relation to neural and behavioral evidence that VSTM and conscious perception are intimately linked and utilize common sensory cortices whereas working memory draws on frontal cortices associated with cognitive control.

The Explanatory Gap Problem: How Neuroscience Might Contribute to Its Solution

Daniel Kostic, Institut für Philosophie, Humboldt Universität zu Berlin

Sunday, June 7th, 17:30–19:30: Poster No. 44

According to explanatory gap problem there is an unbridgeable gap in our understanding of consciousness, because, it has been argued, of leaving out the qualia from any kind of empirical explanation of consciousness. This claim has been corroborated by arguments according to which if phenomenal properties were possibly causally or functionally inert (if they were not causally or functionally co-varying with the neurobiological properties in a counterfactual scenario) then it is taken as strong evidence that they are not functionalizable-cannot have functional roles within an empirical theory of consciousness (Chalmers 1996; Jackson 1982; Kripke 1972; Levine 1983, 2001; etc). However, if phenomenal properties were causally or functionally inert, as proposed by the explanatory gap argument, then a nomological connection between them and physical properties cannot be established, i. e. empirical evidence for psychophysical laws could not be verified, they cannot be detected from the third person perspective. Furthermore, if variations of phenomenal properties have no causal or functional consequences altogether, i.e. if they were causally or functionally inert, then they cannot be detected even from the first person perspective (Pauen 2006). If this is so, then anti-functionalization arguments that rely on the causal or functional inertness of phenomenal properties are incoherent. This argument leaves only one option: the explanatory gap does not make sense assuming causal inertness of phenomenal properties, thus phenomenal properties should be considered functionalizable. This option seems very plausible especially taking into account that it implies an empirical commitment of actually showing how this strategy applies to the available results from neurosciences. For example by distinguishing genuine dissociation syndromes of affective and sensory components of pain (Grahek 2007, Hardcastle 1999) or utilizing the quality space model of color vision (Clark 1993). Functionalization in this framework amounts to showing how neurobiological concepts generalize over phenomenal properties, i. e. how phenomenal properties fit into the functional roles within a possible empirical theory of consciousness.

Evidence from a PRP-Paradigm Points to a Perceptual Locus of the Negative Compatibility Effect

Daniel Krüger, University of Göttingen
Uwe Mattler, University of Göttingen
Susan Klapötke, University of Göttingen

Saturday, June 6th, 17:30–19:30: Poster No. 81

Visual stimuli (primes) that are made invisible by masking can affect motor responses to a subsequent target stimulus. When a prime is followed by a mask which is followed by a target stimulus a negative compatibility effect (NCE, or inverse priming effect) has been found: Responses are slow and frequently incorrect when prime and target stimuli are congruent, and responses are fast and accurate when prime and target stimuli are incongruent. To functionally localize the origins of the NCE, we applied the psychological refractory period (PRP-) paradigm which assumes a perceptual level of stimulus analysis, a central bottleneck of response selection, and a level of motor execution. Two dual-task experiments were run with the PRP-paradigm to localize the NCE relative to the central bottleneck. Results from the locus-of-slack procedure point to a perceptual locus of the NCE, because the NCE disappeared when a tone-task blocked the processing of the visual stimuli before the bottleneck. Results from the effect-propagation procedure point to a perceptual or central locus of the NCE, because the NCE was transferred to the tone-task which had to wait until visual processing had passed the bottleneck. Together, the pattern of results suggests a perceptual locus of the NCE.

Dissociating Intentional Non-Action from Daydreaming by Means of fMRI Pattern Classification

Simone Kuehn, MPI for Human Cognitive and Brain Sciences
Nils Bodammer, MPI for Human Development Berlin
Marcel Brass, University of Ghent

Saturday, June 6th, 17:30–19:30: Poster No. 77

Most juridical systems recognize intentional non-action - e.g. the failure to render assistance - as intentional acts by regarding it as in principle culpable. This raises the fundamental question whether intentional non-action can be distinguished from simply not doing anything on the basis of objective evidence. By means of pattern classification on fMRI data we demonstrate that it is possible to distinguish these two mental states on the basis of brain activity. Our results show that not doing anything is best predicted from resting state brain areas; whereas choosing not to act involves a network that is usually associated with motor simulation and motor preparation. Hence our data support the implicit assumption of legal practice that voluntary non-action shares important features with overt voluntary action.

Auditory Capture of Touch and Bodily Self-consciousness

Tom Lavanchy, EPFL - Swiss Federal Institue of Technology
Jane Aspell, Brain Mind Institute, Ecole Polytechnique Fèdèrale de Lausanne
Bigna Lenggenhager, Brain Mind Institute, EPFL, Switzerland
Olaf Blanke, Brain Mind Institute, EPFL, Switzerland

Saturday, June 6th, 17:30–19:30: Poster No. 89

Bodily self-consciousness can be disrupted by employing conflicting visual-somatosensory bodily input (Lenggenhager et al., 2007; Ehrsson, 2007). These behavioural findings were recently linked to the spatial representation of visuotactile stimuli by measuring crossmodal congruency effects (CCEs) for the full body (Aspell et al., 2009). This study revealed that changes in the mapping of tactile stimuli could be associated with predictable changes in bodily self-consciousness. We were interested in whether bodily self-consciousness as indexed by the CCE could also be modified by conflicting auditotactile stimuli, and as a first step here we used the crossmodal congruency task to study audiotactile integration in the bodily space.We presented auditory stimuli behind the backs of subjects either in far or near bodily space, while tactile stimuli were applied to subjects' backs (n=11). On each trial a target tactile stimulus was presented on the subjects' backs at one of four locations while distractor auditory stimuli were presented at corresponding locations. Subjects had to make speeded elevation discriminations of the tactile targets while ignoring auditory distractors. In visuotactile CCE studies subjects perform worse when the distractors are presented at incongruent elevations (Spence et al., 2004). We did not find audiotactile full-body CCEs with this set-up: congruent and incongruent distractors did not (p>0.05) affect performance differently. In a second study we tested whether also seeing one's body during the experiment would affect audiotactile CCEs. Subjects (n=6) viewed their body (via a camera and head-mounted display) as though two metres in front while audiotactile stimuli were presented as before. This stimulation revealed audiotactile full-body CCEs – i.e. incongruent auditory stimuli impaired performance relative to congruent stimuli (p<0.05). This suggests that audiotactile CCEs are only found when subjects are also able to view their body, even though this visual information is task irrelevant. Audiotactile spatial interactions may therefore be facilitated by vision of one's own body.

When Do You Detect the A? Perception in Grapheme-Color-Synesthesia on Different Visibility Levels

Anna Maria Leugner, Max Planck Institut for Brain Research
Corinna Haenschel, Max Planck Institut for Brai Research
Notger Müller, Magdeburg-Universitätsklinik für Neurologie
Wolf Singer, Max Planck Institut for Brain Research
Lucia Melloni, Max Planck Institute for Brain Research

Saturday, June 6th, 17:30–19:30: Poster No. 87

In this study we investigated whether Synesthetes can detect graphemes earlier and for longer time periods, when aided by color experience, than non-synesthetic subjects. Subjects were instructed to grade the visibility of graphemes embedded in noise via a button press. A letter, a number or a symbol increased to full visibility and then decreased again. Symbols were chosen such that they did not elicit any color experience in the Synesthetes. The Synesthetes and the age and education matched control subjects performed two conditions. In the first condition the stimulus presentation was black and white. In the second condition the background color matched the Synesthete’s experience of the letters and numbers. We could show that letters and numbers were more often reported to be fully seen than symbols in both groups. While the visibility of letters or numbers and symbols was not significantly different between groups, the distinction that Synesthetes made between the percept of the two stimulus types was in specific visibility levels bigger than the distinction that controls made. However, the Synesthetes were able to perceive letters or numbers at lower visibility levels than symbols in both cases, when the visibility increased and when it decreased again. Controls did not vary in the detected visibility levels for the two stimulus types. Remarkably, for the Synesthetes the difference in judgment of letters and numbers versus symbols was bigger in the colored condition than in the black and white condition. In Controls no condition effect was found. It can be concluded that the main difference in the synesthetic percept as compared to the non-synesthetic percept is how a stimulus that evokes a synesthetic experience is perceived relative to one that does not elicit a color impression. Synesthetes perceive graphemes that elicit a synesthetic experience earlier and for longer time periods than those which do not. Additionally, we observed that matching the background color to the synesthetic color of the appearing letter or number provides support of the subject's synesthetic abilities.

Asyncronic Comparison of Qualia

Chia-Hua Lin, Institute of Philosophy of Mind & Cognition
Thomas Benda, Institute of Philosophy of Mind & Cognition

Sunday, June 7th, 17:30–19:30: Poster No. 56

Dennett (1988) has presented a dissonance which is caused by our intuition that one is infallible to his own sensory experiences. This dissonance, in particular, is that while we attribute infallibility to one«s statement of his current coffee taste, we might nonetheless hesitate to attribute the same infallibility to one«s comparison of his coffee taste experiences over time. In other words, the assumption that we have coffee taste qualia is not only helpless but also misleading. Therefore, Dennett suggests, the assumption of qualia should be thrown away. What Dennett has overlooked is that the assumption of the ability to compare qualia simultaneously is different from the assumption of the presence of qualia: while the former should be thrown away, the latter remains useful in the investigation of the mind. This paper on the one hand resorts to the irretrievability of any sensory experience and consequently the impossibility of simultaneous comparison of qualia. On the other hand a hypothesis of the memory of senses is proposed for the asyncronic comparison of qualia. Dennett has raised, contrary to what he was intended, that there is still much to know about experience: if qualia are irretrievable, why we seem to be able to tell the difference between two sensory experiences and why the more one is trained the more detail of distinction one could make? The hypothesis of memory of senses answers to the difficulty of the asyncronic comparison of qualia: for a sensory experience to be memorized and used in later comparison it has to be tagged with a cue or several cues. These cues enable a subject to report the discrepancy or consistency in two sensory experiences even though the comparison is not done, and is not possible to be done, by recalling the previous experience to compare with the present one. Reference: Dennett D. C. (1998) "Quining Qualia" in D. Chalmers (ed.)Philosophy of Mind: Classical and Contemporary Readings, Oxford University Press

Neural Mechanisms of Repetition Priming

Chun-Yui Lin, University of Arizona
Lee Ryan Ryan, University of Arizona

Sunday, June 7th, 17:30–19:30: Poster No. 110

Repetition priming refers to the change in speed, accuracy or bias in processing a stimulus following prior exposure to the same stimulus. It is thought to be a form of unconscious or implicit memory. Recent neuroimaging studies have found that behavioral priming is typically accompanied by reduced neural activity (i.e. repetition suppression, or RS) in several cortical regions (e.g., fusiform gyrus and left prefrontal cortex) and it is hypothesized that RS may be the underlying neural mechanism of behavioral priming. However, it is still under debate whether RS in all these brain regions make similar contributions to behavioral priming. Several recent studies examined the correlation between the magnitudes of behavioral priming and RS across subjects. They generally found that only the RS in the left prefrontal region was correlated with behavioral priming. No strong evidence has been reported yet regarding the correlation between RS in posterior perceptual regions and behavioral priming. This contradicts some classical theories that suggest that priming may rely heavily on the posterior perceptual cortices. One possible reason for this finding is that studies have only examined conceptual priming tasks which would rely more heavily on the frontal areas but not the posterior perceptual cortices. We hypothesized that if perceptual priming tasks are used we may find a correlation between behavioral priming and RS in posterior brain regions. The present functional magnetic resonance imaging (fMRI) study examined this hypothesis. Participants performed several perceptual tasks (symmetry judgments on pictures of novel shapes or known objects, and picture naming) and conceptual priming tasks (natural or man-made classification). Our results indicated that RS in posterior perceptual regions could indeed be correlated with behavioral priming depending on the tasks used. We also demonstrated that within-subject correlations between RS and behavioral priming are more sensitive to this relationship than between-subject correlation coefficients. Our data suggested that priming may not only rely on the frontal cortices but also on other posterior cortical regions that are related to the specific task used.

Music of the Hemispheres: Correlating Phenomenology and Brain Function Through Data Sonification

Dan Lloyd, Trinity College

Saturday, June 6th, 17:30–19:30: Poster No. 133

Consciousness is polyphonic (O'Brien & Opie, 1998), comprising multiple simultaneous, interacting dimensions in only partial registration with immediate sensory/motor environments. This is unsurprising, arising from a variable recurrent dynamical neural system. Yet the science of consciousness tends to be monophonic and static, highlighting one or few aspects of brain function, tracking a narrow band of experimental conditions, and regarding neural responses as stationary (i.e., similar with each repetition within an experiment; Lloyd, 2002, 2004, in press). Presenting data pictorially implicitly reinforces these limitations. Thus, for presenting neurophenomenology, another sense may be preferred. Hearing is inherently multivariate (discriminating simultaneous intensities at multiple frequencies), temporally sensitive, and specialized for tracking dynamic change. Auditory acuity is especially evident in music perception. In this presentation, we convert fMRI data into polyphonic, musical sound. Briefly, distributed brain regions are segregated using Independent Component Analysis. Each region is assigned a tone; its loudness changes as regional activation varies over time. The compiled varying tones produce a polyphonic soundtrack for the experiment (which is then accelerated to typical musical tempos, and visualized with animated ''piano roll notation'', indexing soundtracks with experimental timelines). Many phenomenological saliencies of the fMRI data are audible. In a simulated driving experiment, differing experimental conditions manifest different harmonic keys and tempos, even though no single component corresponds to any specific task condition. Distinctive chords mark condition onsets. In a second experiment, schizophrenia patients and healthy controls can be readily discriminated by sound alone. In both cases, the heard properties correspond to confirmable statistical features of the data. In conclusion, data sonification is recommended as a discovery tool in neurophenomenology. Note: The talk can be scaled from 20 to 50 minutes. The music is fascinating; this could occur afterhours, in the tradition of ASSC cross-disciplinary entertainment. Lloyd, D. (2002). Functional MRI and the study of human consciousness. J Cogn Neurosci, 14(6), 818-831.(2004). Radiant Cool: A Novel Theory of Consciousness. MIT Press (in press). Through a glass darkly: Schizophrenia and functional brain imaging. Philosophy, Psychiatry, and Psychology. O'Brien, G., & Opie, J. (1998). The Disunity of Consciousness. Australasian J Phil, 76, 378-395.

Do Vestibular Signals Influence Illusory Self-Attribution of a Rubber Hand?

Christophe Lopez, Laboratory of Cognitive Neuroscience
Bigna Lenggenhager, Brain Mind Institute, EPFL, Switzerland
Olaf Blanke, Brain Mind Institute, EPFL, Switzerland

Sunday, June 7th, 17:30–19:30: Poster No. 66

The vestibular contribution to bodily self-consciousness is suggested by neurological observations. Artificial stimulations of the peripheral vestibular system may modify the experience of phantom limb sensations in amputees and may also alter self-attribution of body parts in neurological patients suffering from somatoparaphrenia. In the present study, we investigated whether galvanic vestibular stimulation (GVS) - that evoke an illusory sensation of self- and/or environment motion - interferes with the mechanisms underlying body part localization and self-attribution in healthy participants using the so-called rubber hand illusion. The (unseen) left hand of the subject was stroked synchronously or asynchronously with a (seen) left rubber hand for 1 minute. Synchrounous stroking of the subject's hand and the rubber hand induced an illusory self-attribution of the rubber hand and a mislocalization of the subject«s left hand, that was shifted towards the rubber hand. We compared whether localization and self-attribution during the rubber hand illusion was influenced by GVS (anode on the right or left mastoid process, < 2 mA) applied during the 1 minute stroking, with respect to a baseline without any galvanic stimulation and sham stimulations (on the neck). The localization of the subject's left hand was measured immediately after the stroking using a ruler. Self attribution of the rubber hand was mesured with questionnaires. Our preliminary data show that left and right anodal GVS did not modify the illusory self-attribution of the rubber hand (similar ratings of the illusions measured with questionnaires) and did not alter the localization of the subject's left hand. These data stress the robustness and consistency of the rubber hand illusion in healthy subjects even during artificially induced illusory self-motion. Although some clinical observations showed that vestibular stimulations influence self-attribution of one's own hand in some neurological patients (stimulation-induced recovery of ownership), our data in healthy subjects did not reveal a vestibular influence on the illusory self-attribution of a fake hand. As vestibular signals code for the motion and orientation of the whole body, we hypothesize that the vestibular contribution to bodily self-consciousness may be less important for body parts than for the whole body.

The Signal Processing Architeture of Conscious and Unconscious Perception

Brian Maniscalco, Columbia University
Wei Guo, Columbia University
Hakwan Lau, Columbia University

Saturday, June 6th, 17:30–19:30: Poster No. 97

What distinguishes conscious from unconscious processing? Many believe that unconscious processing has its own special substrate or ''channel'' in the brain. For instance, the neural substrate of blindsight and other unconscious processes has been hypothesized to be subcortical, whereas conscious processing is often held to be restricted to the cortex. We call these ''dual channel models''. Alternatively, conscious and unconscious processes may both occur in a single information processing channel. On this view, representations become conscious due to their overall level of activation, or due to how they are processed at higher stages. We call these single channel models or hierarchical models. Distinguishing between these two kinds of models is of great importance as many neurobiological and philosophical theories rest on assuming that either of the models is correct. Previously we reported a visual psychophysics paradigm that dissociates changes in objective task performance from changes in subjective visibility (Lau & Passingham 2006, PNAS), capitalizing on some properties of metacontrast masking. These data present a stringent modeling challenge: which class of models best captures the observed dissociation between objective information processing and subjective, conscious experience? We used formal model comparison techniques to evaluate each model class's ability to fit the data. The results strongly favor a hierarchical model wherein conscious experience is determined by late stages of information processing. We then evaluate each models' ability to fit corresponding fMRI data (collected in a new experiment) in order to confirm our analysis. These results have implications for our neurobiological and conceptual understanding of conscious experience and its neural correlates. On the hierarchical model supported by the results here, what determines whether a representation becomes conscious is not whether it is processed in some privileged neural circuit, or how it is processed at early sensory stages (cf Ned Block's recent discussions). Rather, a representation becomes conscious due to the nature of information processing that occurs at late processing stages. Philosophically, this model supports some versions of the higher-order theory of consciousness.

Cognitive Media and Self-interpretation

Lars Marstaller, Berlin-Brandenburg. Akademie d. Wissenschafte

Saturday, June 6th, 17:30–19:30: Poster No. 59

Verbal reports of subjective experience play a crucial part in establishing a link between the explanandum of consciousness and the explanans of brain activity. But how do we come to know about our own experience? Evidence from neuroscience, comparative anthropology, and developmental theory increasingly emphasizes the socio-cultural dimension of mind. The default system of the brain is largely overlapping with the areas active during social cognition (Schilbach et al. 2008). Basic mind-reading abilities occur very early in the development that humans share with other primates (Tomasello et al. 2005) but is subsequently superseded by use of language and immersion in narrative practices (Gallagher & Hutto 2008). It thus seems not unlikely that knowing about one's own experience is similar to social cognition and mind-reading. I therefore argue that the contents of conscious experience are individuated by socially mediated practices of self-interpretation. Such a claim stands in conflict with computational views that individuate representations according to their systemic function. But I claim that this need not be the case if cognition is seen as embodied, situated (Thompson 2007) and extended (Clark 2008). Human cognition is dependent on interaction with a socio-culturally shaped environment and the human brain's plasticity to invent, use and incorporate external cognitive media that change the information processing regimes for sucessful action in complex environments (Clark 2008). In this sense, language constitutes the paradigmatic cognitive medium. It is both a means of communication governed by social norms and a means of altering cognitive processing demands. I argue that by emphasizing the sensorimotor loop and the internalization of external linguistic cognitive media (Clark 2008, Fernyhough 2008, Clowes 2007), representation and interpretation can be combined. I first review some of the evidence that the social dimension of cognition is fundamental for mind. Then, I argue that contents of experience are individuated through self-interpretation and discuss the possibility of combining representational and interpretational theories of content. Finally, I speculate on the implications for the role of subjective experience in a science of consciousness.

Importance of EEG Frequency Bands in the Assessment of Depth of Anesthesia

Vasile Moca,
Bertram Scheller, Clinic for Anesthesiology, Frankfurt am Main
Raul Muresan, Center for Cognitive and Neural Studies
Michael Daunderer, Clinic for Anesthesiology, Munich
Gordon Pipa, Max Planck Institute for Brain Research

Saturday, June 6th, 17:30–19:30: Poster No. 131

Different brain states have been associated with the expression of various oscillation frequencies in electroencephalographic (EEG) recordings. Of particular importance is the assessment of depth of anesthesia (DOA) in patients undergoing surgical interventions. The proper quantification of brain activity by means of EEG recordings during anesthesia is helpful both for obvious medical reasons and for understanding how conscious states emerge from brain dynamics, and for that matter, how oscillatory activity correlates with conscious/unconscious states. Here, we recorded EEG signals from 62 patients undergoing surgery, under deep anesthesia induced with a cocktail of substances. Auditory stimulation was applied through short clicks repeating with a frequency of 9.1 Hz. After cleaning and filtering the signals in the band of 0.5 to 600 Hz, we divided the data into segments having a length of 100 seconds. These segments were categorized into 5 DOA classes (ranging from 5: awake, to 1: deeply anesthetized) by 2 human expert anesthesiologists that relied on mid-latency auditory evoked potentials (MLAEP) and the notes of the attending anesthesiologist in order to classify. Next, we designed an artificial classifier that extracted time-domain descriptors from the original raw EEG signal with a Time Encoded Signal Processing And Recognition (TESPAR) technique. These descriptors were then classified by means of multi-layer perceptron (MLP) classifiers that had to learn the mapping of EEG segments on the 5 DOA classes, from examples provided by the human experts. The artificial system could learn this mapping almost perfectly, i.e. its classification performance almost equaled the self-consistency of the human experts. Finally, we selectively filtered the EEG signals knocking out low, medium, or high frequency components or keeping various combinations of these, and then we reclassified filtered signals to identify which components contributed most to correctly identifying each DOA class. We found that the gamma band is especially relevant for identifying awake states, and that most other states can be detected relying on a combination of various frequency bands. Results suggest that brain states cannot be simply identified on the basis of single frequency bands, hence, various brain states are characterized by intimate relations among multiple frequency bands.

The Petrified Self and Anosognosia in Alzheimer«s Disease

Daniel Mograbi, King's College London
Prof. Richard Brown, King's College London, Inst. of Psychiatry
Robin Morris, King's College London, Inst. of Psychiatry

Saturday, June 6th, 17:30–19:30: Poster No. 63

Anosognosia - lack of awareness concerning a disease or loss of cognitive function - is a common feature in Alzheimer's disease (AD). We will explore the relationship between the preservation of the self and anosognosia in AD, suggesting that one potential explanation for this phenomenon is the lack of update of personal information due to the memory impairments characteristic of this illness. For this purpose we will briefly review the literature about the neural correlates of the self, the relationship between self and memory and the profile of memory impairments in AD, suggesting the hypothesis of a petrified self as the cause for anosognosia. We will show how this is in accordance with evidences from studies that point to an outdated self-evaluation in Alzheimer's disease and previous accounts of anosognosia, such as the views of Ramachandran (1999), Weiskrantz (1997) and Morris (Agnew & Morris 1998, Morris & Hannesdottir 2004).

Decoding Covert Intentions in Time-based Prospective Memory

Ida Momennejad, Berlin School of Mind and Brain & BCCN, Humboldt-Universität zu Berlin
John-Dylan Haynes, BCCN, Charitè - Universitätsmedizin Berlin

Saturday, June 6th, 17:30–19:30: Poster No. 71

Our sense of 'volitional control' partly depends upon our ability to pursue 'distal intentions' or intended commitments to perform a task at a later point in time. Such distal intentions need to be pursued even in face of distracting short term tasks. This requires a capacity to maintain the prior intention while performing other tasks, and later retrieve and execute the intention once we think the appropriate conditions are met. Here we investigated a) whether it is possible to dissociate the spatiotemporal patterns of brain activity associated with encoding, maintenance, and retrieval of covert distal intentions and b) whether it is possible to decode distal intentions during a conflicting ongoing task. To answer these questions, we used a modified task-switching paradigm in combination with functional magnetic resonance imaging. Subjects had to perform one task (color judgement) while at the same time memorizing one of two other possible delayed tasks (magnitude or parity judgement). The subjects were required to switch to the delayed task in a self-paced fashion after a given time of 15s, 20s or 25s. A support vector machine was trained to recognize or ''classify'' the brain patterns associated with the future intention during the maintenance phase before the switch to the delayed task. When tested on novel imaging data, our method could successfully predict the covert delayed intention up to 20 seconds before the switch, and with up to 85% accuracy in single subjects. The information concerning the prospective task-set could best be decoded from the spatiotemporal patterns of activity in the premotor and prefrontal cortices. The maintenance condition revealed more rostrolateral and dorsomedial patterns of activation in the PFC whereas the retrieval phase activated more rostromedial-PFC regions. Notably, during the maintenance phase high decoding accuracies were obtained in the premotor cortex (FEW,p<0.001), DLPFC, and orbitofrontal areas which can be ascribed to memory retrieval, temporal processing, task-set-conflict and inhibition. In conclusion, information about a covert prior intention to perform a task-set tens of seconds later can be decoded from ongoing brain activity even while one«s currently executed intention (''intention-in-action'') is to perform a highly demanding conflicting task-set.

Single Neurons Projection of Thoughts Onto a Computer Screen: Modeling the Data

Cerf Moran, California Institute of Technology

Saturday, June 6th, 17:30–19:30: Poster No. 139

Look at the computer, concentrate on thinking of your favorite person, and make his or her image appear on the screen. Try harder. Harder... For people who do not possess super-hero powers, this experience is bound to fail. However, we conducted such a experiment on eleven subjects, enabling them to control the appearance of an image on a computer screen with their mind. We recorded from neurons in the medial temporal lobe (MTL) in patients with pharmacologically intractable epilepsy implanted with chronic electrodes to localize the seizure focus for possible surgical resection. Patients were presented with four different images (familiar personalities, animals etc) that elicited specific firing activity in four single units selected in a prior screening session. The patients were asked to fade in and out one of two images simultaneously present on the screen by attending to one or the other overlapping images for 10 sec. Patients reached a control accuracy level of 71% after a few trials. We adapted a Diffusion-Decision model to quantitatively describe the activity of these 4 single neurons using Poisson rates ramping up to a threshold and an attentional modulation of the neuron’s firing rates. This enabled the patient to learn how to utilize this real-time closed-loop circuit to improve conscious mental control over fading (by differentially attending to one or the other images). This implies that subjects can selectively and differentially control the firing activity of subsets of MTL neurons.

Hypnotic Susceptibility, Sleepiness and Subjective Experience

Levente Móró, University of Turku
Valdas Noreika, University of Turku
Antti Revonsuo, University of Skövde
Sakari Kallio, University of Skövde, Sweden

Saturday, June 6th, 17:30–19:30: Poster No. 103

Albeit being an interesting topic in consciousness research, relationships between different altered states of consciousness are rather insufficiently explored. In particular, hypnosis research remains isolated from studies of other altered states of mind. The early view of hypnosis had been deeply influenced by its observable resemblance with sleep - even the phenomenon was named after the Greek god of sleep, Hypnos. Since those times a great number of experimental results had pointed out that hypnosis is neither the usual waking state nor any of the sleep stages. However, theoretical and experimental work suggests that the phenomena of hypnosis and sleep still share many similarities which should not be rejected. In our study, we investigated the relationships between hypnotic susceptibility, self-reported sleepiness, and the subjective experience of receiving hypnotic suggestions. Based on previous work, our initial hypotheses held that: 1) Hypnotic susceptibility correlates positively with self-reported daytime sleepiness; 2) Higher hypnotic susceptibility is associated with higher hypnotic inertia that manifests as increased drowsiness after the hypnotic procedure, and 3) Subjective experience gives a more valid measure of hypnotic suggestibility than behavioral scores alone. In order to test the hypotheses, we administered hypnosis-related and sleep-related scales to 90 volunteer subjects. After a standard group hypnosis session, the subjects were asked to fill out the Harvard Group Scale of Hypnotic Susceptibility Form A (HGSHS:A), the Karolinska Sleepiness Scale (KSS), the Epworth Sleepiness Scale (ESS), the Pittsburgh Sleep Quality Index (PSQI), and our self-developed the Questionnaire on Subjective Hypnotic Experiences (Form E). HGSHS:A scores and item difficulties were checked against previously established Finnish and international norms, and all scale results were analyzed by multiple statistical methods. Our findings indicate that hypnotic susceptibility correlates with both habitual daytime sleepiness and instantaneous sleepiness after the hypnotic procedure. Furthermore, our results lend also support to the third initial hypotheses about subjective scales being useful tools when evaluating experiences of hypnosis and sleep. For both these cases of altered consciousness, findings support the idea of common underlying psychological mechanisms, such as the ability to quickly relax the mind-body, or to get immersed into less externally driven mental content.

Vision by Inference: Visual Recognition Under Uncertainty

Raul Muresan, Center for Cognitive and Neural Studies
Ioana Tincas, Center for Cognitive and Neural Studies
Vasile Moca, Center for Cognitive and Neural Studies
Lucia Melloni, Max Planck Institute for Brain Research

Sunday, June 7th, 17:30–19:30: Poster No. 26

When presented with images of isolated objects, human subjects can provide almost instant recognition, indicative of apparently feed-forward like processing of visual scenes. When objects are occluded or embedded into a set of distractors, recognition is more difficult and a visual search task must be first engaged. A third case, that was much less investigated however, is when the visual system handles very limited amounts of information about the objects in the visual scene. When elementary visual features are lacking and subjects are provided with barely discriminable information, they engage into a hypothesis-driven visual exploration process, giving either a wrong answer or the correct answer but after a very long exploration time. To study such cases, one needs to be able to control the amount of information that is provided to the visual system and to prevent elementary visual features to be detectable in the stimulus (in order to prevent instant recognition). We developed a stimulation paradigm that takes advantage of a deformable grid of black dots on a white background. The grid is progressively deformed towards points of maximum information content belonging to the object to be categorized. When the grid is the less deformed subjects have a very hard time perceiving the underlying target object, while for high levels of deformation towards informative points, the subjects can instantly recognize the object. By manipulating a single parameter, one is able to provide the visual system with progressively more information without directly exposing the subject to elementary visual features of the object. Object-related information is retrieved from the statistical deviation of grid points from the regular lattice towards informative points belonging to the object. Therefore, subjects engage into an active visual inference rather than feed-forward recognition process. We show with psychophysical data collected from 9 subjects that our stimulation paradigm can provide a continuous and controllable transition from unperceivable to easily perceivable representations of objects. Therefore, this stimulation paradigm may be especially useful to study how the visual system handles uncertainty, i.e. visual inference.

From Implicit to Explicit Self-Representation: No Self-Consciousness Without Consciousness of Other Minds

Kristina Musholt, Berlin School of Mind and Brain, Humboldt-Universität zu Berlin

Sunday, June 7th, 17:30–19:30: Poster No. 42

Self-consciousness can be defined as the ability to think of oneself as oneself (e.g. L. R. Baker, 1998). Here, I will argue that self-consciousness in this sense requires an explicit self-representation, which needs to be distinguished from the implicit self-related information that is a necessary structural component of conscious experience. This distinction can be elucidated with the help of Perry's (1993) theory of 'unarticulated constituents' and Recanati's (2007) framework for a relativist semantics, and can be further spelled out by distinguishing different levels of explicit representation (Dienes & Perner, 1999; Karmiloff-Smith, 1992). Given that self-consciousness requires an explicit self-representation, it follows that it is crucial for any theory of self-consciousness to explain the transition between states that contain implicit self-related information (i.e., information that is in fact about the subject) to states that contain explicit self-related information (i.e., information that is represented as being about the subject). I will argue that an important part of this transition consists in the acquisition of a Theory of Mind, that is, the ability to ascribe mental states to others and to oneself. If this is right, it suggests a symmetric relation between self-consciousness and consciousness of other minds; in other words that the same cognitive mechanisms required for attributing mental states to others are necessary for attributing mental states to oneself (cf. Frith and Happé, 1994). I will present theoretical as well as empirical data that lend support to this "symmetry thesis", that is, the thesis that self-consciousness is constitutively linked to an awareness of other minds.

Individual Differences in Subjective Experience: The Case of Visual-Gustatory Interactions

Sagiv Noam, Centre for Cognition and Neuroimaging, Brunel University, West London
Jessica Stock, Brunel University, West London, UK
Sagiv Noam, Centre for Cognition and Neuroimaging, Brunel University, West London

Sunday, June 7th, 17:30–19:30: Poster No. 84

Cross-modal interactions are prevalent in perception. Such interactions can influence perception, performance, and brain activity. However, it is becoming increasingly clear that there are substantial individuals differences in the degree to which information processed in one sensory modality can alter processing in another. The wide range of synaesthesia variants known to science demonstrates this well. In the present study we examined whether visual stimuli can induce a gustatory experience in some individuals. We find that a minority of individuals do report vivid synaesthesia-like gustatory imagery induced by images of food. We will discuss how these interactions depend on familiarity and taste preference, whether they are induced by the food image or the sight of another individual eating it, and whether the phenomenon depends on empathy and the presence of other types of synaesthesia. We speculate that the phenomenon may be mediated by hyper-activation of the mirror neuron system at least in some of the individuals we identified.

The Subjective and the Objective Duration of Static NREM Sleep Dreams

Valdas Noreika, University of Turku
Valtteri Arstila, University of Turku
Christine Falter, University of Oxford
Julian Kiverstein, University of Edinburgh
Jennifer Windt, Johannes Gutenberg-Universität Mainz
Antti Revonsuo, University of Skövde

Sunday, June 7th, 17:30–19:30: Poster No. 104

Anecdotal evidence suggests that dreams may distort the subjective experience of time. The relationship between the subjective and the objective clock duration of dreams has been debated for many years. Several researchers have argued that people dream throughout the night, while others have claimed that dreams are formed instantaneously at the moment of awakening. Yet, only a few empirical studies, based on post-awakening reports, incorporation of external stimuli into dream content, or counting tasks performed in lucid dreams, have aimed to explore the average duration of dreams. These studies have focused on the duration of typical complex REM sleep dreams, leaving NREM sleep dreaming aside. We aimed to investigate the duration of static NREM sleep dreams, which usually consist of just a few comparatively mundane experiences. Such simple dreams, which can be considered as minimal states of sleep consciousness, are usually forgotten, unless reports are elicited in the sleep laboratory. 10 participants spent 4 experimental non-consecutive nights in the sleep laboratory and were awakened, following the early-night serial awakenings paradigm, 8 times per night. Awakenings took place during NREM sleep Stages 2 and 3 as well as during sleep onset REM sleep, and were followed by free dream reports and answers to a detailed questionnaire on different aspects of dreaming, including 5 questions about the subjectively estimated duration and the speed of time passage in the dream. Post-awakening answers revealed that the subjectively estimated duration of dreams differs between sleep stages even when the duration of sleep was controlled, with REM dreams reported as having been longer than NREM dreams. Yet, even the simplest NREM dreams were often experienced as extending into 30 s, 1 min or even longer time intervals. By contrast, the subjective speed of time passage did not differ between dreams of different sleep stages and was typically scored as resembling waking life. The duration of minimal dreams was also tentatively explored by EEG contrast between the reports of dreaming and dreamless sleep from the same sleep stages. This novel approach might eventually provide new means for determining the objective rather than the subjective duration of dreams.

Measuring Flexible Control in Artificial Grammar Learning.

Elisabeth Norman, Haukeland University Hospital, University of Bergen
Mark Price, University of Bergen
Emma Jones, University of Bergen

Saturday, June 6th, 17:30–19:30: Poster No. 115

One operational criterion for assessing whether knowledge is consciously available is that it can be applied in a controlled, flexible manner (Baars, 1988). This study represents an attempt to apply this criterion to knowledge acquired in an Artificial Grammar Learning (AGL) experiment. We present data from one experimental study (N = 80) that applied a novel procedure for measuring flexible control in AGL. In the learning phase, participants were trained on two different artificial grammars, Grammar A vs. Grammar B (Dienes, Altmann, Kwan, & Goode, 1995; Wan, Dienes, & Fu, 2008). In a pure test condition, grammaticality was judged with respect to one grammar throughout the block. In a mixed test condition, instructions to classify with respect to either Grammar A or Grammar B varied randomly from trial to trial. The pure test condition corresponds to the procedure of Dienes et al. (1995). The mixed test condition was a novel procedure assumed to require a higher degree of conscious, flexible control. The experiment used either traditional letter strings written in black ink or letter strings in which the colour and font of individual letters varied randomly between trials to increase the perceived complexity and thus reduce the influence of explicit strategies. The ability to classify the test strings according to the grammatical rule was above chance in both the pure and mixed test conditions. Confidence ratings indicated that participants in all conditions showed metacognitive awareness of the acquired knowledge. There was evidence of flexible control in all conditions. However, in the simple stimulus condition, there was a relative advantage of pure blocks over mixed blocks. This advantage was not found in the complex condition. Flexible control and metacognitive awareness indicate that learning was clearly not implicit, even in a condition of increased perceived complexity. We discuss which additional measurements are needed to assess whether learning in this situation is fully explicit or associated with conscious feelings reflecting implicit learning, e.g., fringe consciousness.

Effects of Object Substitution Masking on Different Qualities of Perceptual Awareness

Kathrin Ohla, Centre Hospitalier Universitaire Vaudois, University of Lausanne
Micah Murray, Centre Hospitalier Universitaire Vaudois
Ulrike Toepel, Centre Hospitalier Universitaire Vaudois
Niko Busch, Universitè Paul Sabatier, Toulouse

Sunday, June 7th, 17:30–19:30: Poster No. 6

Object substitution masking (OSM) occurs when a briefly presented target is surrounded by a spatially distinct mask that outlasts it. This delayed mask offset has been shown to abolish an accurate perceptual representation of the target prior to semantic analyses. According to the re-entrant hypothesis of OSM, conscious perception emerges from recurrent communication between low level and high level visual areas. The trailing mask is thought to create a mismatch between low level information and feedback representations to the effect that processing of the object is interrupted, rendering only the mask available for conscious report. Interestingly, most previous studies have focused on OSM of simple geometric objects. Here, we investigated the impact of OSM on the perception of complex naturalistic objects. To achieve this, we first studied effects of set size and trailing mask duration on the perception of masked targets. These behavioral results confirm and extend previous findings of stronger masking with increasing set size and longer mask duration. We then conducted a high density electroencephalographic study and performed electrical neuroimaging analyses to identify the brain mechanisms that underlie different qualities of the formation of object representations. More specifically, we contrasted trials in which target objects were localised within the display but could not be identified with trials in which objects were fully identified and with trials in which objects were neither localised nor identified. While some EEG effects were suprisingly independent of how accurate information about the target object could be reported, other effects were specific for consciously perceived targets. We discuss the results in the context of the re-entrant hypothesis of OSM.

The Visible Trace of Invisible Elements in Human Vision

Thomas Otto, Laboratoire Psychologie de la Perception, Universitè Paris Descartes
Haluk Ögmen, University of Houston
Michael Herzog, Ecole Polytechnique Fèdèrale de Lausanne

Saturday, June 6th, 17:30–19:30: Poster No. 5

One of the most frequently used tools in conscious research is visual masking in which a target is rendered invisible by a following mask. Usually, it is assumed that the target is suppressed by the mask and, hence, does not reach awareness. Here, we show that even though the target itself is rendered invisible by masking, its features can still be clearly perceived at other locations. This result raises the question whether or not the target is conscious. We used a sequential metacontrast masking paradigm in which a central Vernier is followed on either side by sequences of flanking lines. These lines are consecutively shifted further away in space creating the percept of two diverging motion streams. Because of the masking the Vernier is invisible. Still, when the Vernier is slightly offset to the left or to the right, observers perceive a corresponding offset within the stream of lines, which are in fact straight. Hence, although the Vernier itself is invisible, its offset is visible. To investigate how the Vernier offset is processed, we presented a second offset to one of the flanking lines and measured the interaction between the two offsets. Our results show that the offsets are very precisely processed although observers cannot consciously access the individual offsets within the motion streams. Importantly, the two offsets are integrated only if they are grouped within the same motion stream. Therefore, we propose that unconscious feature processing is achieved by grouping. In this framework, the Vernier itself is not conscious, not because of suppression but because of grouping.

An Enactive Account of the Mutual Scaffolding of Concepts and Experience

Joel Parthemore, COIN Lab, Informatics Research Center
Anthony Morse, COIN Lab

Sunday, June 7th, 17:30–19:30: Poster No. 36

Specifying the content of experience, including the conceptual, is central to understanding consciousness, yet current theories of concepts do not seem up to the job. We suggest an empirically testable model of the mutual scaffolding of concepts and experience, where neither is ultimately primary but each gives rise to the other. Such a model can then serve as the anchor to an enactive theory of concepts more generally. Concepts are most naturally understood in symbolic and representational terms, while much of experience, in contrast, is non-symbolic and non-representational. We begin by proposing that concepts are grounded in a combination of (a) sensorimotor engagement between the agent and its environment, (b) a parsimonious set of proto-conceptual primitives, and (c) an ability to discover and assign salience to regularities between action and perception. This is in keeping with neuroscientific evidence and contemporary enactive approaches. The discovery of the most basic of regularities alters the interaction of the agent with its environment not only by shaping and directing the discovery of further such regularities between action and perception, but of regularities among the regularities themselves. So there are both sensorimotor contingencies and further contingencies that do not directly involve sensorimotor engagement but only relate indirectly back to such engagement. As this process of finding patterns and patterns within patterns snowballs, the result is increasingly more abstract and crucially more recognizably representational and symbolic content - precisely to the extent that it becomes more invariant and therefore to some practical extent independent of particular moments of experience or particular contexts of application. Even such ''high-level'' conceptual content has impact down to the lowest levels of cognition, as recent empirical work e.g. in visual processing attests. Critically, conceptual content is never independent of its roots; it lies toward one end of a continuum grounded in and ultimately inseparable from basic sensorimotor contingencies. In parallel with this, we conclude that the aspects of cognition most naturally described as symbolic - i.e., the conceptual ones - arise from mechanisms best understood in sub-symbolic and associational terms.

Incidental Learning of Interactions Between Motor and Linguistic Sequences

Antoine Pasquali, Universitè Libre de Bruxelles
Stephanie Chambaron, Universitè Libre de Bruxelles
Dominique Ginhac, Universitè de Bourgogne
Axel Cleeremans, Universitè Libre de Bruxelles

Sunday, June 7th, 17:30–19:30: Poster No. 112

In order to investigate interactions between different levels of processing under a paradigm of Serial Reaction Time (SRT) task, we develop a dual task in which the usual dots stimuli are replaced by words forming meaningful sentences. In this new task, a motor (SM) and a linguistic (SW) sequence continuously and concurrently attract participant's attention. In a first experiment, SM is second-order-conditional (SOC), of length 12 and among 4 items, whereas SW is a fairy tale. Learning phase consists of 15 blocks of 100 words. During the transfer block (13th), we either randomize the motor sequence (TM), the words' order (TW), or both (TMW). Transfer effect in TMW appears to be exactly equal to the sum of TM and TW effects. In a second experiment, we introduce new instructions in order that participants dedicate their attention either to the SM (CM) or to the SW (CW). A post-test SM recognition task is applied, completion of trigrams in both inclusion and exclusion, but also for specific words presented in bold in the SW. As expected, TM effect is stronger for CM, and TW is stronger for CW. Moreover only CW performs well in the SW recognition task. However, CM does not score better in the SM recognition task, as if forced-reading were interfering with participants' ability to form explicit knowledge about the sequence. We introduce in a last experiment an abstract interaction between SM and SW by building sentences such that every word has a specific position depending on its type (1: article/pronoun; 2: adjective/adverb; 3: noun; 4: verb). In the transfer block, we present conditions in which, either we break the interaction between the two sequences (TI), or we change the two sequences' construction but keep their interaction (TC), or both (TIC). All participants are to respond as fast as possible (CI), but additionally, half of them must also pay attention to the positions of the words (CE). During a post-test recognition task, we ask participants to discriminate old and new words' positions. Results should confirm an incidental learning of the abstract interaction between MS and LS.

History Dependence in Multistable Perception Highlights the Role of Noise in Visual Perception

Alexander Pastukhov, Otto-von-Guericke Universität
Jochen Braun, Cognitive Biology, Otto-von-Guericke University
Pedro Ernesto Garcia Rodriguez, Mathematics Researh Center
Gustavo Deco, Pompeu Fabra University

Sunday, June 7th, 17:30–19:30: Poster No. 24

If our visual system is confronted with an ambiguous image it lapses into spontaneous alternations, a phenomenon dubbed multistable perception. A classical statistical property of these oscillations is statistical independence of successive dominance percepts (''sequential independence''). This feature is hard to reconcile with susceptibility of perceptual dominance to adaptation. To investigate this puzzling contradiction we assessed the predictive power of cumulative history for future dominance periods. Cumulative history was computed convolving all preceding dominance intervals with an exponential decay (time constant _;decay). We found that it has a significant influence on dominance durations, transition times, as well as transition direction. First, we observe a significant correlation between cumulative history and the next dominance time (correlation coefficient 0.2-0.5) for 0.3< _;decay/ Tdom<1. Consistent with adaptation, observed correlation was negative for dominance of the same and positive for dominance of the other percept. Cumulative history of clear percepts also influences transition periods: their durations as well as the likelihood of return transitions peaked when histories were balanced, revealing the noise driven nature of perceptual reversals. We also carried out computational simulations of a simple rate model to assess its ability to replicate experimental findings. Additional statistical properties, revealed by the cumulative history approach, resulted in narrower regions of suitable values for adaptation and inhibition strength variables. Critically, they show that a system should operate close to the bifurcation line in a bi-stable rather than oscillatory regime. Former one implies that noise is indispensable for the phenomenal alternations even in the presence of adaptation dynamics. In summary, multistable perception depends significantly on history, even for continuous displays. This dependence was hitherto overlooked, as it is evident only with an integral measure of history. The modest degree of this dependence, extended transitions in the event of balanced histories, lack of correlation between estimated adaptation time-constants and spontaneous alternations rates of observers and results of simulation all point out at the importance of noise as a principal driver of perceptual transitions.

Personal Autonomy and Intentional Action Control

Jan Prause-Stamm, Berlin School of Mind and Brain, Humboldt-Universität zu Berlin

Saturday, June 6th, 17:30–19:30: Poster No. 57

The capacity to consciously control one«s own actions is an essential aspect of autonomous agency. However, based on Libet's famous experiments (Libet et. al. 1983) some philosophers claim that human agents lack the capacity for conscious action control. Thus, they conclude autonomy must be an illusion. I argue that these skeptical conclusion about the possibility of conscious action control and autonomous agency rest on two faulty assumptions. First, it is presupposed, on a conceptual level, that only actions that are under immediate conscious control can be understood as autonomous. Secondly, it is maintained that the empirical findings support the claim that it is impossible to consciously control one's own actions. In contrast to these assumptions I argue, firstly, that unconscious action control can be a part of autonomous agency as long as it is still the agent which controls his actions. Secondly, I develop a framework of intentional action control which emphasizes the different functions of consciously forming and holding an intention. I conclude that autonomous agency is not threatened by recent neuropsychological findings about mechanisms which initiate and sustain actions without being under immediate conscious control. Quite the opposite, an empirically based picture of autonomous agency emerges. Of particular importance for my own account of autonomous agency is the notion of an intention conceived of as a mental state that modulates the mechanisms which initiate and sustain actions in order to achieve the intended goal. Apart from philosophical accounts of natural agency (Bishop 1989; Bratman 1987, 2006; Mele 1995) I put emphasis on recent neuropsychological approaches towards executive functions (Miller & Cohen 2001) and the role of conscious intention formation for actions which are not under immediate conscious control (Goschke 2003; Hommel 2003). In my opinion understanding the interdependencies between conscious and unconscious action control is one of the major challenges for the development of a naturalistic account of autonomous agency. My discussion will clarify some important conceptual issues about intentional action control and autonomy and I will use empirical data about intentional action control in order to elaborate a detailed account of autonomous agency.

What Kind of Mental Images Are Synaesthestic Spatial Forms?

Mark Price, University of Bergen

Sunday, June 7th, 17:30–19:30: Poster No. 82

For a substantial minority of people, thinking about members of certain sequences (e.g., months, week days, numbers) is accompanied by the additional sensory experience that the concept member occupies a precise location in imaginal or peripersonal space. This location is in turn part of a spatially extended pattern which can be very complex and idiosyncratic. These spatial forms remain a poorly understood individual difference in everyday mental imagery. Outstanding questions include whether spatial forms should be considered as a variety of synaesthesia, or as an extreme on the continuum of everyday mental imagery, whether they are associated with benefits or disadvantages, and what type of visuo-spatial representation they are. Both behavioural and self-report data will be presented to address these issues. First, by self-report, people with spatial forms show above average everyday mental imagery. In addition, scores on a questionnaire that taps spatial form experience in the general population also correlate strongly with self-reported general imagery. Second, mental imagery instructions can induce control groups to show similar behavioural patterns to people with spatial forms in laboratory experiments. Third, spatial forms do not confer strong advantages in tests that might be expected to benefit from inspectable visuo-spatial models of the layout of the forms. In these tests, loading visuo-spatial components of working memory also fails to selectively impair performance of spatial form groups. It is tentatively concluded that spatial forms may be vivid automised visual images at the end of the spectrum of individual differences in mental imagery, but do not seem to be spatial cognitive maps of the type that some introspective reports would initially suggest.

Is Sequential Knowledge Essentially Statistical in Nature?

Fu Qiufang, Institute of Psychology, CAS
Sun Huiming, Institute of Psychology, CAS
Zoltan Dienes, University of Sussex
Fu Xiaolan, Institute of Psychology, CAS

Sunday, June 7th, 17:30–19:30: Poster No. 114

We investigated whether the sequential knowledge acquired in sequence learning is essentially statistical in nature, as many have argued, or can include more abstract properties, and whether such knowledge can be unconscious. We adopted three types of stimuli in probabilistic sequence learning. Compared with the standard stimuli that followed a second order conditional (SOC) sequence, transfer stimuli followed a different SOC sequence (same abstract properties, different n-grams); deviant stimuli followed neither SOC sequence (both abstract properties and n-grams different), which contained mainly runs and alternations. The three types of stimuli all appeared in each block. The probability for standard stimuli was .833 or .667 depending on experiments and correspondingly the probabilities for transfer and deviant stimuli were either same (both .083), or different (.25 for transfer and .083 for deviant). Subjects were presented two generation tests after either a short or long training phase, while indicating either their confidence in their judgment (on a 50-100% scale) or the basis of their judgment as it subjectively seemed to them: random, intuition, rules or memory. Subjects had to generate sequences that either followed the training regularities (inclusion) or failed to follow them (exclusion). In training, subjects showed faster responses to standard stimuli than transfer stimuli, showing procedural learning of n-grams, but also faster responses to transfer stimuli than deviant stimuli, showing procedural learning of abstract features of the sequential structure beyond n-grams. After training, subjects could, depending on training conditions, generate more standard than transfer responses, indicating control over the use of knowledge of n-grams, and more transfer than deviant responses, indicating control over the use of abstract knowledge. In both cases, control could be exerted when subjects were subjectively unaware of their knowledge (i.e. they felt they were guessing or using intuition). The results were inconsistent with the claim that subjects can only implicitly learn n-grams (fragments or chunks), i.e. only engage in first-order statistical learning, and suggested instead that sequential knowledge could be based on abstract rules. The data challenge chunking or fragment models of unconscious learning.

Inattention Boosts Subjective Visibility: Implications for Inattentional and Change Blindness

Dobromir Rahnev, Columbia University
Brian Maniscalco, Columbia University
Elliott Huang, Columbia University
Linda Bahdo, Columbia University
Hakwan Lau, Columbia University

Saturday, June 6th, 17:30–19:30: Poster No. 9

Why is it that the whole visual scene in front of us appears to produce uniform subjective visibility, when current research seems to suggest that we can only process a small part of it effectively (e.g. experiments of change and inattentional blindness)? Here we report a paradoxical finding that can explain this phenomenon. Subjects detected strongly and weakly attended gratings whose contrasts were adjusted to produce the same detectability (d). We found that subjects were conservative in detecting the strongly attended gratings and liberal in detecting the weakly attended gratings, i.e. they produced higher hit rate when there was a lack of attention. We confirmed that this was still the case even when subjects were instructed to be unbiased and were given feedback after each trial. The effects were robust and easy to replicate. In order to test whether the results were merely driven by the difference in stimulus contrasts between the attended and unattended conditions, in a separate experiment we used multiple levels of contrasts for both conditions. The results demonstrated that more attention leads to lower hit rate even when the effect of stimulus contrast was independently accounted for. We reasoned that these effects were due to weakly attended stimuli producing disproportionately high phenomenology, and we tested this hypothesis directly in another experiment. We matched for subjects' ability to discriminate between two orientations of a grating, and required subjects to give subjective visibility ratings after the discrimination. As expected, participants gave higher ratings for the weakly attended gratings. Thus, surprisingly, our results demonstrate that the subjective visibility associated with weakly attended signals is higher than what would be warranted by the quality of these signals. This effect can explain people«s experience of a uniformly visible visual field and their surprise after performing badly in inattentional and change blindness experiments - the lack of attention to the periphery may paradoxically give rise to an inflated sense of phenomenology. Our results provide further support to the idea that attention and awareness are fundamentally different processes.

Attention and Biased Competition in Multi-Voxel Object Representations

Leila Reddy, Universitè de Toulouse, UPS, CerCo, France
Nancy Kanwisher, Mc Govern Institute, MIT
Rufin VanRullen, CNRS,Centre de Recherche Cerveau et Cognition

Sunday, June 7th, 17:30–19:30: Poster No. 2

The biased-competition theory accounts for the effects of attention at the neuronal level. Two hallmarks of this model are 1) the neuronal response to simultaneously presented stimuli is a weighted average of the response to isolated stimuli, and 2) attention biases the corresponding weights in favor of the attended stimulus. However, perception is not a property of single neurons, but probably relies instead on the activity of larger populations of neurons, which could be reflected in fMRI patterns of activity. Because several non-linearities can influence the pooling of single-neuron responses into BOLD signals, the fMRI effects of attention need not exactly mirror those observed at the neuronal level. Here we ask 1) how simultaneous stimuli are combined in multi-voxel patterns of representation and 2) how this effect depends on stimulus category, the brain region under consideration, and the allocation of attention. We considered data from an fMRI study in which four object categories (faces, houses, shoes and cars) were presented in four conditions: in isolation, or in pairs such that each category was attended, unattended, or attention was divided equally between the two. Unlike traditional analyses that collapse the response across all voxels in a region of interest, the response in each condition was represented in a multi-dimensional space where each voxel defined a dimension. In this high-dimensional space, the BOLD response to two simultaneously presented categories was well described as a weighted average of the response to individual stimuli. The weights were biased towards the preferred category in category-selective regions (FFA and PPA for faces and houses, respectively). Independently of this category-specific effect, and consistent with the biased competition theory, attention shifted the weights in favor of the attended stimulus, and the magnitude of this shift (30%) was quantitatively consistent with previous reports in single neurons.

Language the Mediation of Shallow and Deep Interpretation and Its Role in Conscious Mental Life

Clowes Robert, Centre For Research in Cognitive Science

Sunday, June 7th, 17:30–19:30: Poster No. 54

In recent years the philosopher Andy Clark (1998; 2006) is one of several who, against previous cognitive science orthodoxy, have argued that human language plays a central role in human thinking. Clark's ecological account of the cognitive role of language hinges on language playing a role in constraining, structuring and scaffolding non-linguistic cognition, rather than mirroring it. Clark's view hangs together well with current embodied and ecological approaches to mind. Clark's account seems to hinge on what I have called shallow interpretation (Clowes, 2007): that is, words are used as ecological pivots around which other cognitive abilities turn. This account argues for a special role for language which appears contrary to other 'embodied' theories of thinking, specifically those posited by some cognitive linguists (Fauconnier & Turner, 2002; Lakoff & Johnson, 1999). The cognitive linguists see distinctly human abilities depending on what might be called 'deep interpretation': the use of unconscious metaphor systems facilitated in part by language. This talk will consider how human conscious mental life depends upon the ability to shift between shallow and deep interpretation and what this indicates about our conscious mental lives more generally. References: Carruthers, P. (1996). Language thought and consciousness: an essay in philosophical psychology. Cambridge: Cambridge University Press. Clark, A. (1998). Magic Words: How Language Augments Human Computation. In P. Carruthers & J. Boucher (Eds.), Language and Thought. Interdisciplinary Themes (pp. 162 - 183). Oxford: Oxford University Press. Clark, A. (2006). Material Symbols. Philosophical Psychology, 19(3), 291-307. Clowes, R. W. (2007). The complex vehicles of human thought and the role of scaffolding, internalisation and semiotics in human representation. Paper presented at the Adaptation and Representation, virtual platform @ Fauconnier, G., & Turner, M. (2002). The Way we Think: Conceptual Blending and the Mind's Hidden Complexities. New York: Basic Books. Lakoff, G., & Johnson, M. (1999). Philosophy in the Flesh. New York: Basic Books.

V1 and V2 are not the Direct Neural Correlates of Visual Awareness: Evidence from the TMS-study

Niina Salminen-Vaparanta, University of Turku
Mika Koivisto, University of Turku
Valdas Noreika, University of Turku
Simo Vanni, Helsinki University of Technology
Matti Itkonen, University of Turku
Antti Revonsuo, University of Skövde

Sunday, June 7th, 17:30–19:30: Poster No. 8

The role of V1 and V2 in visual awareness remains controversial. Lamme (2000, 2006) proposed that visual information is processed twice in V1: when the visual input arrives to the striate area, and when it comes back after a feedforward and feedback loop in the ventral visual stream. The recurrent processing to V1 was suggested to be sufficient for visual awareness to arise. In contrast, Koch (2003) argued that V1 is prerequisite for visual awareness, and processes visual information largely unconsciously. The present study aimed to test the hypothesis that if Lamme's model is accurate, then TMS-stimulation disturbs visual awareness during feedforward activation ~ 50-100 ms post stimulus (Bullier et al. 1996) and feedback activation after ~ 140 ms post stimulus (Koivisto et al. 2005). In the previous TMS-studies of early visual areas only the accuracy of the responses, but not the subjective awareness itself, has been measured. Consequently, in the present study the accuracy of responses and subjective awareness of stimulus features and occurrence were studied. Six participants performed a letter-identification task by giving a forced-choice response and an evaluation of their experience of the stimulus while receiving magnetic stimulation. Single TMS-pulses were delivered after the stimulus onset either to V1 or V2, by combining the detailed retinotopic maps of V1 and V2, acquired with multifocal fMRI (Vanni et al. 2005), with a modelling of the electric field distribution in occipital lobe during TMS-stimulation. In both V1 and V2 stimulation, awareness of stimulus features was suppressed in relatively long time window (between 40-120 ms post stimulus), whereas awareness of stimulus occurrence was impaired between 40-80 ms after stimulus onset. Because V1 and V2 stimulations revealed only single time windows for suppression of awareness, without any later suppression after 140 ms, the information processing in early visual areas seems to be a prerequisite for visual awareness (e.g., feedforward processing), rather than a direct neural correlate of it. However, it remains possible that the late parts within the suppression time windows involve rapid local recurrent processing.

Performance Feedback and Event Related Potentials in Blindsight

Jessica Sänger, Heinrich-Heine University
Petra Stoerig, Heinrich-Heine-University of Düsseldorf

Sunday, June 7th, 17:30–19:30: Poster No. 116

Lesions of the primary visual cortex cause visual field defects in the contralateral hemifield. Although patients are unaware of stimuli presented within absolute defects, they may still detect, localize, and discriminate them when forced choice methods are used. As blindsight improves with practise, we here asked whether trial-by-trial feedback would enhance detection performance. A disk (3.2° in diameter, 80 ms, -.5 log contrast) was presented in pseudo-random alternation with a blank stimulus to the hemianopic field of two patients. Both target and blank stimuli were announced by a brief beep, and the patients indicated 'blank' or 'target' by button press. In half of the 1,200 trials, each response was followed by visual feedback (an icon showing a thumb pointing upward or downward) to signal response correctness. To learn how stimuli, responses and feedback are reflected in the electroencephalographic brain activity, we recorded event-related potentials (ERPs) while the participants performed the task. Although the patients reported no stimulus awareness, even without feedback detection was better than expected on the basis of uninformed guessing. However, feedback improved detection so that d' consistently increased from 1.8 to 3.4 (across both patients) over the course of 10 blocks. Electrophysiologically, target but not blank stimulus presentations produced a delayed ipsilesional N1-component. Interestingly, response-related ERPs show an enhanced negativity at parietocentral electrode sites for correct compared to error responses. Furthermore, the error feedback, which occurred in about 25-30% of the trials, yielded a frontocentral P3-like positivity. Together the results indicate that trial-by-trial feedback improved the detection of 'blind-seen' targets that differentially modulated ipsilesional brain activity. The unusual negativity that followed correct more than false responses shows that the brain integrated stimulus and response information in the absence of stimulus awareness on the part of the patient. Possibly by virtue of being relatively rare, the error feedback after the false responses initiated a pronounced potential that may play a role in improving the implicit performance.

A Modality Independent Monitoring Network Modulates Conscious Experience of Agency

Fabian Schäbe, Division Med. Psych, University Hospital Bonn
Henrik Walter, University of Bonn
Knut Schnell, University Hospital Bonn, Division Med. Psych

Saturday, June 6th, 17:30–19:30: Poster No. 65

Introduction Predictive self monitoring is crucial for conscious experience of agency [1]. Only if there is a mismatch in prediction and actual feedback, feedback is experienced fully conscious. Recently, we have demonstrated a fronto-parietal monitoring network underlying these predictions in a visuo-motor-task using fMRI [2, 3]. Here, we researched if this network is modality independent. Experiment: We investigated 20 healthy subjects in a 3 Tesla MRI-scanner. In two different paradigms, participants had to monitor temporal incongruence between their own actions and sensory consequences. In the first experiment, participants performed a simple visuomotor task (steering a car) and had to detect intermittent takeover of control by the computer. In the second, auditory-speech task, participants read words aloud in scanning pauses receiving auditory feedback via headphones and had to detect asynchronicity between their own actions and auditory feedback. Both experiments contained a control condition (CC) in which no monitoring had to be performed, a solitary monitoring condition without (sine) actual incongruence (MCsIn) and a monitoring condition with (cum) actual incongruence (MCcIn). Results: In both experiments, epochs of action monitoring comprising incongruence events (MCcIN>CC) induced a significant (p<0.01 FDR corrected) activation of a network consisting of bilateral inferior parietal lobes, inferior frontal/insular and anterior cingulate cortices, ventral striatum, cerebellar hemispheres and right dorsolateral prefrontal cortex. Additionally, modality specific activation in this contrast was found in respective sensory cortices i.e. in left temporal auditory areas in the auditory-speech task and in temporoparietal junctions in the visuomotor task. In both experiments, solitary action monitoring (MCsIn>CC) events induced activation of the right inferior frontal/ insular cortex. Discussion: Our experiments prove the existence of a modality independent network for sensory-motor self-monitoring. Within this network the transition between the right inferior frontal gyrus and anterior insula is already activated in the absence of actual incongruence events and might therefore be related to self-awareness in general [4]. [1] Blakemore SJ, Frith C. Curr Opin Neurobiol 2003; 13: 219-24. [2] Schnell K et al. Neuroimage 2007; 34: 332- 341. [3] Schnell K et al. Brain 2008; 131: 2783-97. [4] Craig AD. Curr Opin Neurobiol 2003; 13: 500-5.

Is Free Will Saved by Perceptual Biases?

Stefan Schauber, Technical University Berlin
Manfred Thüring, Technische Universiät Berlin

Saturday, June 6th, 17:30–19:30: Poster No. 69

The results of Libet et al. (1983) indicate an unconscious neuronal initiation of free conscious decisions. In this classic study a new method was introduced to determine mental acts. This paradigm was immediately criticized by several authors (e.g. Breitmeyer, 1985) as it used a fast rotating spot circling around a clock face. In particular the representational momentum, first described by Freyd & Finke (1984), was discussed to be a possible influence on the timing of subjective events. Research on this effect indicates that judgements on the location of moving objects are biased by the direction of the movement. It therefore might serve as an alternative explanation of Libet's findings. Nevertheless, as the representational momentum is highly sensitive to influences like object speed, size, or context, possible influence of the effect is hard to determine. The current study tried to fill this gap. Combining Libet's clock paradigm with a typical representational momentum experiment enabled us to investigate if the results of Libet et al. (1983) could be explained by the mentioned effect. Our findings indicate that the representational momentum is weakened by the use of a clock face - as employed by Libet - and the observed antedate of the rotating spot is relatively small. Hence, the representational momentum offers no sufficient alternative for explaining Libet's findings.

TT, ST, IT, NPH... How Do We Understand Others?

Tobias Schlicht, Center for Integrative Neuroscience Tübingen
Albert Newen, Institut für Philosophie, Ruhr-Universität Bochum

Sunday, June 7th, 17:30–19:30: Poster No. 58

There is a longstanding controversy about how we know about the mental states of others (mindreading). According to Theory-Theory (TT), when we ascribe mental states to a person, we employ a folk-psychological theory similar to a scientific theory (Gopnik, Meltzoff 1997). Some proponents of TT postulate an innate neural mechanism designed to understand others (Baron-Cohen et al. 1985). According to Simulation Theory (ST), we create pretend mental states in ourselves and project them onto others, thus simulating in our own minds what others might be thinking (Goldman 2006). Gallagher's (2005) Interaction Theory (IT) rejects an assumption shared by those views, namely, that we have a problem of gaining access to other minds in the sense that they have mental states 'hidden' behind their (otherwise meaningless observable) behavior. The central claim is that we can, in most cases, directly perceive what other people are up to. Finally, according to Hutto's (2008) Narrative Practice Hypothesis (NPH), our direct encounter with stories about people acting for reasons, being supplied in interactive contexts, provides us with the forms and norms of folk psychology. The controversy results from the fact that these views are typically introduced as characterizations of the 'most pervasive' or even 'only' and thus exclusive way of understanding other minds. In this paper, it is argued that such strong claims are unconvincing. For example, Goldman's (2006) Simulation Theory does not work partly because it does not illuminate how we come to know which pretend mental states we have to create. Here, assumptions borrowed from TT are necessary. In the constructive part of the paper, it is argued that a more differentiated framework is needed according to which we have different mindreading strategies at our disposal. Whether and when we employ them depends (1) on our prior relation to the person we wish to understand, (2) on their behavioural patterns which we observe and (3) on the context of the situation in which the observed person displays these behaviours. In the sketch of such a framework, it is shown how elements of TT, ST, IT, and NPH can be integrated.

What's Wrong with Mr Hyde? Personal Identity, 'Multiple Personality' and the Unity of Consciousness

Ulla Schmid, Philosophisches Seminar, Universität Basel

Saturday, June 6th, 17:30–19:30: Poster No. 33

It has been argued repeatedly that the phenomenon of Multiple Personality actually confronts us with the existence of different persons in their own rights, all located within one human body (e.g. Rovane 1998). This interpretation is based on the limited accessibility between the different streams of consciousness of a 'host personality' and her various 'alter personalities': affected people for example 'lack time' or can't remember their actions. Further, this view is based on the assumption that a unified consciousness is a necessary condition for there being a person or, in other words, that the identity of a person (her 'self') is tied to the scope of those mental states she can consciously recall. I find that this view is misleading in at least three respects. On the one hand, the view sketched so far confounds the identity of persons with that of personalities, or better: different facets of a person's emotional/psychological life. Whereas the term 'person' refers to a human being emphasising her (conceptual) capability as a rational and moral agent, the term 'personality' refers roughly to the particular character a concrete person develops during her life. Therefore, the unity of consciousness is not the foundation of the unity of a person, but rather derivative from it. On the other hand, consciousness is understood single-sidedly as basically consisting in mental 'states' which 'occur' 'within' the human being. This employs the Humean picture of the mind as theatre of changing impressions and neglects that large parts of a person's mental life are constituted by the person herself as the subject that generates them. Finally, the constitution of an individual's personality or 'self' is not understandable atemporally, i.e. without consideration of her biography, and in particular her actions. In this sense, the dissociation of different realms of experience can be seen as an instrument for a certain purpose, e.g. a reaction to traumatising events, which is performed rather than suffered by the affected individual and presupposes rather than undermines the unity of the different personalities.

Feature-Based Attention Modulates Priming Effects in a Primed-Pointing Paradigm

Filipp Schmidt, Justus-Liebig-Universität Giessen
Thomas Schmidt, Justus-Liebig-Universität Giessen

Saturday, June 6th, 17:30–19:30: Poster No. 13

We introduced two experiments to study the influence of feature-based attention with color and shape on the time course of speeded primed pointing responses. Color or shape targets were preceded by prime stimuli that triggered either the same or opposite response as the targets on the same location. Before each trial the relevant target pair was indicated by an attentional cue stimulus. Time intervals between presentation of cue and prime as well as prime and target stimuli were varied systematically. Prime visibilities were explored in separate behavioral tasks. Pointing trajectories showed strong priming effects that were amplified by attentional direction on the relevant feature at optimal cue-prime intervals. In conclusion, visual feature-based attention seems to modulate the earliest phases of visuomotor processing. This modulation was independent of visual awareness of the primes, strongly supporting the notion of distinct processes underlying visual awareness and attention.

Accessing Preconscious Stages of Lightness Processing

Thomas Schmidt, Justus-Liebig-Universität Giessen
Sandra Miksch, Justus-Liebig-Universität Giessen
Lisa Bulganin, Justus-Liebig-Universität Giessen
Florian Jäger, Justus-Liebig-Universität Giessen
Felix Lossin, Justus-Liebig-Universität Giessen

Saturday, June 6th, 17:30–19:30: Poster No. 21

We demonstrate qualitative dissociations of lightness processing in visuomotor priming and conscious vision. Speeded keypress responses to the brighter of two luminance targets were performed in the presence of preceding dark and bright primes (clearly visible and flanking the targets) whose apparent lightness was enhanced or attenuated by a lightness illusion. Response times to the targets were greatly affected by consistent vs. inconsistent arrangements of the primes relative to the targets. These priming effects could systematically contradict subjective lightness matches, such that one prime could appear brighter than the other but prime as if it was darker. Instead of following this illusion, priming effects eclusively depended on the raw local contrast of flankers to the immediate background. Our findings indicate that visuomotor priming effects, as opposed to conscious perceptual judgments, access an early stage of lightness processing exclusively based on local image contrast and still devoid of lightness constancy.

The Intentional Content and Structure of Motor Experience

Michael Schmitz, Forschergruppe "Grenzen der Absichtlichkeit", Fachbereich Philosophie, Universität Konstanz

Sunday, June 7th, 17:30–19:30: Poster No. 32

While there has been a burgeoning interest in the phenomenology of action in recent years, the basic motor experience of acting is still somewhat neglected, often ignored or at least not clearly distinguished from other forms of actional experience. After drawing that distinction, an account of the structure and intentional content of motor experiences is proposed. It is argued that they present their subjects (rather than the experiences themselves, as suggested by Searle (1983) - as the cause of bodily movements; that they are constitutive of an elementary form of action; and that they therefore (pace e.g. Bayne 2009) should not be considered to be perceptions of actions. Then a case is made for the view that the intentional content of motor experiences is nonconceptual according to criteria such as their fine-grainedness, analog character and context-dependence. It is further argued that these elementary action (re)presentations are intention-independent in a sense that is parallel to the way in which perceptual states are belief-independent. Finally, it is shown how this conceptual framework can account for empirical results (e.g. Kammers et al. 2006) showing a divergence between assessments of location manifested in grasping behavior and in verbal judgment, and for other phenomena such as 'utilization behavior', in terms of the relative independence of conceptual and nonconceptual representations.

Losing Your Self: Neural Correlates of Frontotemporal Dementia

Matthias Schroeter, MPI Human Cognitive and Brain Sciences
Karolina Raczka, MPI Human Cognitive and Brain Sciences
Jane Neumann, MPI Human Cognitive and Brain Sciences

Saturday, June 6th, 17:30–19:30: Poster No. 67

Recent studies suggest that frontotemporal lobar degeneration is the second most common diagnosis of dementia in individuals younger than 65 years. The most common subtype, frontotemporal dementia (FTD), is characterized by alterations in behavior and personality, namely decline in social interpersonal conduct, impairment in regulation of personal conduct, emotional blunting, and loss of insight ('diagnostic core features'). Although FTD is clinically well characterized and several hints for specific cognitive deficits were recently published, there is still controversy with regard to the neural basis of the disease. Accordingly, the aim of our study (Schroeter et al., 2008) was to focus neural impairments in FTD to specific brain regions and discuss the disease in a framework of cognitive neuropsychiatry. We conducted a systematic and quantitative meta-analysis including morphometric studies with magnetic resonance imaging (MRI) and functional imaging studies applying (18F) fluorodeoxyglucose positron emission tomography (FDG-PET). Seven significant above-threshold clusters were identified. They were located in several frontomedian regions (anterior medial frontal cortex, pregenual anterior cingulate cortex, subcallosal/septal area, gyrus rectus), medial thalamus, left superior frontal sulcus, and right anterior insula. Our study characterizes FTD as the frontal variant of frontotemporal lobar degeneration. More specifically, it indicates that FTD is mainly characterized by impairments in frontomedian networks. This brain region has been previously suggested as the key region for theory of mind (ToM) or 'mentalizing', where mental states have to be attributed to self and other people, and which enables social cognition. Moreover, it has been related to self-related processing, and it was suggested that permanently processing self-referential stimuli constitutes the self of a person. In sum, our study suggests that FTD selectively affects neural networks related to social cognition, finally leading to the aforementioned clinical symptoms. Hence, the disease FTD enables an understanding of the ''condition humaine'', the neural networks that make us human. References: Schroeter ML, Raczka K, Neumann J, von Cramon DY. Neural networks in frontotemporal dementia - A meta-analysis. Neurobiology of Aging 2008;29:418-26.

Stability of Pattern Vectors Distinguishes Conscious from Non-Conscious Neural Information

Aaron Schurger, CEA - NeuroSpin (France)
Francisco Pereira, Princeton University
Anne Treisman, Princeton University
Jonathan Cohen, Princeton University

Sunday, June 7th, 17:30–19:30: Poster No. 16

One consistent finding among neuroimaging studies contrasting conscious and non-conscious perception is a significantly stronger response to consciously perceived stimuli in content-specific areas such as the FFA or V5/MT. By contrast, we assert that the transition of neural information from ''non-conscious'' to ''conscious'' depends as much on the nature of the distributed pattern of neural information as it does on the specific loci and absolute level of neural activation. Based on the theory of Marcel Kinsbourne, we predicted that, compared to non-conscious perception of the same content, conscious perception would be associated with (1) a more consistent and distinct spatial distribution of information from one stimulus presentation to the next, and (2) a more consistent pattern of activation within that configuration, without necessarily any difference in intensity. We parametrically manipulated the visibility of face and house stimuli using dichoptic color masking. Behavioral discrimination accuracy ranged from ~100% (''visible''), to ~70% (''threshold''), to chance (50%, ''invisible''). A Gaussian naive Bayes classifier was trained to discriminate the stimulus category at each level of visibility based on voxels in the temporal lobes, with nested cross-validation used for voxel selection. The classifier was significantly above chance (12 subjects) at both the highest and lowest levels of visibility (but, surprisingly, not at the threshold level). When the classifier was trained on "visible" patterns and tested on ''invisible'' and vice-versa, accuracy was indistinguishable from chance, and there was little overlap in the space of voxels that was maximally informative at each level of visibility. In order to test our predictions, we treated each set of ''informative'' voxels as a high-dimensional space, and the BOLD activation patterns as vectors within that space (treating the vector angle as a reflection of the information content, and the norm as a reflection of the intensity of the representation). Our results confirm our predictions, suggesting that conscious states are (relatively) stable states. We are able to rule out an explanation based solely on differences in signal-to-noise ratio. Our results also indicate that conscious information is more sparsely and narrowly distributed, challenging the idea of a ''global broadcast''.

The Influence of rTMS over the Dorsolateral Prefrontal Cortex on Hypnotic Response

Rebecca Semmens-Wheeler, University of Sussex
Zoltan Dienes, University of Sussex

Sunday, June 7th, 17:30–19:30: Poster No. 100

Higher cognitive functions have traditionally been assumed to require consciousness and attention (e.g. Norman and Shallice 1986), involving dorsolateral prefrontal cortex (DLPFC), bilaterally: the left dorsolateral prefrontal cortex (DLPFC) for creating accurate higher order thoughts (HOTs) and conscious awareness (Lau, 2008), and the right DLPFC for top-down attentional processes (Vanderhasselt et al, 2007). Current neurophysiological (Gruzelier, 1998) and sociocognitive theories of hypnosis (Dienes and Perner, 2007) also suggest possible role of DLPFC for hypnotic suggestibility by assigning distinct roles to each DLPFC. Specifically, it would be expected that disruption of left DLPFC function would increase hypnotic susceptibility (by making it harder to create accurate HOTs and thus easier to form intentions to respond to hypnotic suggestions without awareness). Conversely, we predicted that disrupting the right would lead to a decrease in hypnotic suggestibility. The present study used repetitive transcranial magnetic stimulation (rTMS) to investigate the role of both - left and right - DLPFC on hypnotic suggestibility. The function of the left DLPFC was temporarily disrupted using rTMS; the vertex was stimulated as a control (sham) site. We found that the participants' subjective ratings of hypnotic phenomena were significantly increased for left DLPFC stimulation, as compared to control stimulation. This effect accounted for 25% of the variance, controlling for expectancy. Objective and involuntariness ratings did not differ significantly across conditions. On the other hand, the influence of the TMS on the right DLPFC showed a decrease in hypnotic suggestibility, as expected (the results obtained by a previous pilot study). These results are in line with previous studies (Dyer, Dienes and Hutton, 2007; Lau and Passingham, 2006; Vanderhasselt et al, 2007) and provide strong support for the role of the left DLPFC in subjective feelings of hypnosis. The findings thus provide support for the assertion of cold control theory that hypnotic responding can be resultant from inaccurate HOTs. However the lack of differences observed for objective and involuntariness ratings suggest that other factors may play an important role in hypnotic processing (e.g. socio-cognitive factors). Altogether, this study offers novel insight into the mechanisms underlying hypnotic processing and the nature of subjective consciousness experience.

Self-Other Distinction in the Perception of Actions Performed in Synchrony with Music

Vassilis Sevdalis, MPI for Human Cognitive and Brain Sciences
Peter Keller, MPI for Human Cognitive and Brain Sciences

Saturday, June 6th, 17:30–19:30: Poster No. 61

Previous research has shown that point-light displays of bodily movements can provide essential information about the identity of an acting agent. Observers can distinguish between their own and others' actions when depicted as point-light displays. Nevertheless, the self-other distinction in point-light displays has been little studied when action information is presented in different sensory modalities simultaneously and in music related contexts. The aim of this study was to investigate the self-other distinction in point-light displays in which actions are presented in visual or audiovisual modalities. Specifically, is agent recognition facilitated by audiovisual presentation for actions that were originally performed with musical pacing stimuli? In a first session, participants executed different ecologically valid actions ranging in complexity (clapping, walking, and dancing) in time with three musical pieces. Performances were recorded by a motion capture system. In two subsequent sessions the same participants watched point-light displays of their own or another participant's actions. The task was to identify the agent depicted (self vs. other). Various manipulations were applied to the visual information (degradation from fifteen to two point-lights) and to the auditory information (self generated claps vs. externally generated music vs. no sound). Results indicate that recognition accuracy was significantly better than chance in all action and degradation conditions and was significantly increased when more kinematic information was available, both between different actions and within the same action. Auditory information, whether self or externally generated, did not contribute reliably to performance even for the most ambiguous visual displays. The absence of a benefit of audiovisual displays suggests that information about the temporal relationship between agents' movements and the music made no compelling contribution to the perception of action identity. Overall findings suggest that judgements of action agency depend predominantly on motor cues. The potency of such cues is evidenced by the fact that agents were recognizable in the case of simple actions, even when minimal visual information was provided.

Modular Small-World Neural Networks as a Substrate for Consciousness

Murray Shanahan, Imperial College London

Sunday, June 7th, 17:30–19:30: Poster No. 96

This presentation proposes the unification of two influential approaches to the scientific study of consciousness, namely neurodynamics and global workspace theory. 1) A number of researchers pursuing a neuroscientific understanding of consciousness have advocated the idea that the conscious condition depends on a balance of segregated and integrated activity in the brain, and several formal complexity measures have been defined that attempt to quantify this balance (Edelman, Tononi, Seth, and others). Meanwhile, anatomical and functional studies have shown that the nervous systems of many animals exhibit small-world connectivity properties at multiple scales (Sporns and others). Recent computer modelling work where complexity measures are applied to modular small-world neural networks has shown that they are capable of supporting dynamically complex activity, wherein segregation and integration are duly balanced (Shanahan, 2008b). So, from the standpoint of those who advocate a neurodynamical approach, such networks look like a promising candidate for the neural substrate of consciousness. 2) According to global workspace theory (Baars), the mammalian brain instantiates an architecture comprising a set of parallel specialist processes and a global workspace. Parallel specialists (or coalitions of processes) compete to influence the global workspace, which in turn exercises a systemic influence on the full cohort of specialists. The conscious/ unconscious distinction then aligns with the distinction between localised processing and processing mediated by the global workspace. The long-range connections of the cerebral white matter have been hypothesised as the basis of a possible global neuronal workspace (Dehaene and others), and computer models show how they might exhibit the required a succession of broadcast, reverberating states (Shanahan, 2008a). From a graph-theoretical point of view, localised regions of cortex form densely clustered modules, while long-range white matter pathways confer small-world properties on cortex as a whole. In short, modular small-world networks also look like a promising candidate for the neural substrate of consciousness from the standpoint of global workspace theory. Shanahan, M.P. (2008a). A spiking neuron model of cortical broadcast and competition. Consciousness and Cognition 17(1), 288-303. Shanahan, M.P. (2008b). Dynamical complexity in small-world networks of spiking neurons. Physical Review E 78, 041924.

Functional Connectivity Study of fMRI "Resting Brain Activity" in Vegetative State

Andrea Soddu, Coma Science Group, Cyclotron Research Center, University of liège
Melanie Boly, Cyclotron Research Centre, Belgium
Quentin Noirhomme, Coma Science Group, CRC, University of liège
Audrey Vanhaudenhuyse, Coma Science Group, Cyclotron Research Center, University of liège
Mario Stanziano, Second University of Naples
Smadar Ovadia, Weizmann Institute of Science
Pierre Maquet, Cyclotron Research Centre, Belgium
Michele Papa, Second University of Naples
Rafael Rafael, Weizmann Institute of Science
Steven Laureys, Cyclotron Research Centre, liège, Belgium

Sunday, June 7th, 17:30–19:30: Poster No. 130

Introduction: The aim of this study is to assess fMRI resting-state cerebral connectivity in vegetative state patients by means of a user-independent method. Resting baseline (or ''default mode'') activity is thought to be related to awareness of the internal world (i.e., mind wandering, daydreaming, mental imagery, inner speech etc) and encompasses posterior-cingulate/precuneal, anterior cingulate/mesiofrontal and temporoparietal junction cortices. Methods: We here present a novel clinical application for a user-independent ''default mode'' network analysis. Resting state data were acquired on 12 vegetative state (age range 27-87 y) and 26 healthy subjects (21-60 y). Patients« diagnosis was based on Coma Recovery-Scale assessment prior and following scanning. Data were pre-processed and analyzed using independent component analysis ICA as implemented in Brain Voyager. Connectivity studies employed 13 target regions of interest (10x10x10 mm) defined on an average ''default mode'' map calculated in controls. Resting state connectivity was assessed by calculating the number of functional connections within the ''default mode'' map for each subject. Next, student-T tests compared patients to controls at the group-level (p<0.05). Results: Compared to controls, vegetative patients showed a lower total number of edges (i.e., connections; 46 +/- 15 and 24+/-9, p=5*10^-6) and less functional connections with the precuneus (9+/-2 and 4+/-3; p=7*10^-4). The ''default mode'' network shows a reduced connectivity in vegetative patients as compared with controls mainly in posterior brain areas encompassing the precuneus and posterior parietal cortices. Conclusions: The presented connectivity ICA method permits a user-independent identification of the ''default mode'' network connectivity in the vegetative state. Comparison with healthy control data emphasizes the importance of precuneal and posterior parietal functional disconnections in pathological loss of consciousness. References: Boly, M (2008), 'Consciousness and cerebral baseline activity fluctuations', Human Brain Mapping, vol. 29, no., pp. 868-74. Fair, DA (2008), 'The maturing architecture of the brain's default network', Proc Natl Acad Sci USA, vol. 105, no. 10, pp. 4028-32. Fox, MD (2005), 'The human brain is intrinsically organized into dynamic, anticorrelated functional networks', Proc Natl Acad Sci USA, vol. 102, no. 27, pp. 9673-8.

Awareness of Fearful Faces Requires Attention

Timo Stein, Princeton University
Marius V. Peelen, Princeton University
Johanna Ritter, LMU Munich
Katharina N. Seidl, LMU Munich

Saturday, June 6th, 17:30–19:30: Poster No. 17

Fearful faces receive privileged access to awareness relative to happy and non-emotional faces. We investigated whether this advantage depends on currently available attentional resources. In an attentional blink paradigm, observers detected faces presented during the attentional blink period that could depict either a fearful or a happy expression. Attentional load of the blink-inducing target was manipulated by increasing flanker interference. For the low-load condition, fearful faces were detected more often than happy faces, replicating previous reports. Importantly, this advantage for fearful faces disappeared for the high-load condition, during which fearful and happy faces were detected equally often. These results suggest that the privileged access of fearful faces to awareness does not occur automatically, but instead requires attentional resources.

The Explanatory Role of Computational Models in Cognitive Neuroscience

Catherine Stinson, University of Pittsburgh

Sunday, June 7th, 17:30–19:30: Poster No. 50

This paper reevaluates the role of computational models in Cognitive Neuroscience in light of recent work on Mechanisms. This reevaluation has implications for the purported autonomy of psychological theories, and clarifies the sense of emergence appealed to in PDP work. The basic goal of Cognitive Neuroscience is to integrate theories of cognitive function with knowledge about neural structures. This consists neither in ontological identifications across levels, nor in reducing the laws of one level to the laws of the other. Rather this integration occurs through explanations of cognitive functions by neural structures. Further, these explanations are typically mechanistic. This paper concentrates on the role of computational models in these explanations. The explanatory role played by computational models cannot be filling in the details of Marr's computational level for a given algorithmic-level cognitive theory. The filling in of details would not constitute an explanation, since this sort of strategy already assumes that the algorithm can be instantiated (in multiple ways), and that the theory is autonomous from any particular instantiation. For computational models to be explanatory, the status of the cognitive theory must not be autonomous from the explanans being sought. Computational models are not meant to be lower level implementations of cognitive theories, as should be obvious from the fact that they exhibit behaviour analogous to cognitive functions. Computational models are better understood as simulations. Simulations can serve both as proofs of concept for cognitive theories, and as a method for discovering mechanisms. PDP models in particular, which do not have cognitive modules hard-coded in, but rather gradually develop intermediate structures, can be seen as a method for discovering novel mechanisms. Seeing computational models as simulations clarifies talk of emergence in the PDP literature. Bedau introduces a notion of ''weak emergence as underivability except by simulation''. I argue that the sense of emergence most applicable to PDP models is related to this one, although behaviours need only be underivable in practice, not in principle. In this sense, saying that a behaviour emerges from a computational model means exactly that the simulation has discovered a mechanistic explanation for the behaviour.

Object Recognition Through Image-to-Sound-Based Sensory Substitution of Sight

Petra Stoerig, Institute of Experimental Psychology, Heinrich-Heine University
Michael J. Proulx, Institute of Experimental Psychology
Tessa van Leeuwen, F.C.Donders Center for Cognitive Neuroimaging

Saturday, June 6th, 17:30–19:30: Poster No. 83

Our knowledge of objects is largely derived through our senses. The visual one is of particular importance because it allows us to gain information about complex scenes and silent objects we cannot explore tactually. Blindness prevents this immediate access, and sensory substitution attempts to (re-) establish it by transforming visual images into a format that an intact sensory modality can process. To learn whether blind and blindfolded subjects learn to interpret sound patterns that represent everyday objects, we here used an image-to-sound converting sensory substitution system with a small head-based video camera. The objects belong to three categories (bathroom, kitchen, food); five objects were presented per category and session. To further generalization across object classes such as cups, brushes, and carrots, different exemplars were used in each of 15 sessions. Participants listened to the sound patterns generated by viewing the objects from different perspectives, and tried to categorize (''kitchen'') and identify (''cup'') them. Before and after the training, participants were asked to describe the features that allow recognition of the objects. Also before and after the training, participants underwent functional magnetic resonance imaging to reveal how the brain responds to the sound patterns and how the training affects activation patterns. Results show marked improvements in both categorization and identification performance; they were accompanied by increased activation in multimodal superior parietal regions. Particular features (like handles) and materials (like metal) are especially relevant for recognition. Despite the necessity to adapt to the properties of the unusual ''eye''; the training enabled fast recognition for prototypical exemplars in a subgroup of participants. In a developmental sighted ''audiovisual''; synaesthete, the auditory format that represents visual information generated visual phenomena unrelated to the objects, however, object-related visual representations were described by a late blind participant with deprivation-induced synaesthesia. Supported by the Volkswagen-Foundation (I 80/742 & 743)

Epistemic Restraint: An Antidote to Zombie Poison

Carolyn Suchy-Dicey, Boston University

Sunday, June 7th, 17:30–19:30: Poster No. 140

Zombies haunt philosophers of mind now more, perhaps, than Halloween movie-goers. The implications of Cartesian dualism made it easy to laugh it off until David Chalmers presented materialists with a red pill: materialism cannot make sense of the difference between conscious beings and philosophical zombies, where the latter are functionally indistinguishable from the former but unconscious. Moreover, not only are materialists unable to explain the existence of consciousness, but even if we take the existence of consciousness as a primitive, materialists are unable to explain why individual conscious experiences have the characteristics they do. In fact, the materialist is unable to distinguish not only the zombie from the conscious being, but the subject with normal vision from the subject with an inverted color spectrum. The zombie problem, or so-called "hard problem," has now made its way into the science of consciousness as it suggests that researchers cannot study consciousness by studying its material underpinnings. This "hard problem" of consciousness studies comes in several forms, but all of these point to the same epistemic gap: despite our valiant efforts, we cannot get "all the way" to consciousness. I argue in this paper that the hard problem is actually an impossible problem. However, I also argue that the problem is not as rich as it first appears, and that it is no greater than the necessary limitation on our knowledge of the real world or the existence of other minds. Thus, epistemic restraint dissolves the hard problem in the same way that it dissolves these other metaphysical worries: by setting our sights a little lower than epistemic infallibility, we can re-open the portals of consciousness studies. This need not be equivalent to ignoring the problem, or taking the blue pill, as we can admit levels of certainty in knowledge, where the highest level is not attainable for any objective claims. In my view, the hard problem is a colorful illustration of an epistemic limit that, because it is a necessary feature of our experience, makes no practical difference.

Dream Remembering: An Exploration of the Ways Dreams Penetrate Consciousness

Connie Svob, University of Alberta
Don Kuiken, University of Alberta

Saturday, June 6th, 17:30–19:30: Poster No. 105

While it may seem intuitively clear that dreams have the power to influence waking thoughts and feelings (e.g., the distress that follows severe nightmares), surprisingly little research has explored the ways in which dreams enter waking consciousness. In general, the literature on dream recall has been limited to dream recall frequency (Blagrove & Akehurst; 2000, Schredl, 2008; Schredl & Reinhard, 2008; Schredl, 2007). By focusing on the frequency with which dreams are recalled, the specific details, nuances, and processes by which dreams are remembered have been overlooked. We address specifically how dreams may penetrate waking consciousness, including the ways in which they continue to linger for minutes, hours, days, and, sometimes, even years later. To investigate the phenomenon of dream remembering beyond frequency per se, we designed a new questionnaire to measure and explore the dimensionality of dream remembering. We anticipated that a uni-dimensional model would be insufficient to capture the complexity of dream remembering. Rather, we expected that a multi-dimensional model would more accurately depict the different forms of dream remembering. In an initial study, we administered a dream remembering questionnaire to 512 participants. Using exploratory factor analysis (Principal Components, Varimax rotation), we identified nine dimensions of dream remembering. The nine-factor model includes the following dimensions: (1) vivid dream revisualization (including recall frequency), (2) affective and motivational carryover, (3) the frequency of disturbing dreams (e.g., nightmares), (4) liminal dream re-entry, (5) dream/reality blending, (6) cued daytime dream recall, (7) lingering sensory acuity, (8) persistent paralysis, and (9) dream recall consistency. Based on the findings of this study, we propose a theoretical shift that allows for the articulation of a multi-dimensional model of dream remembering. Furthermore, a re-conceptualization of what is meant by ''remembering'' carries implications for the domain of scientific consciousness research. By deepening our understanding of the components involved in remembering dreams, greater insight may be gained into components and processes of memory, boundaries between states of consciousness, as well as the purpose and function of dreaming.

Modality-Independent Modulation of Unpredictable Event on Visual Perceptual Stability

Kohske Takahashi, The University of Tokyo
Katsumi Watanabe, The University of Tokyo

Sunday, June 7th, 17:30–19:30: Poster No. 20

In visual competition, perception for invariant visual patterns changes stochastically from one percept to the other. This perceptual alternation is known to be susceptible to various internal (e.g. attention or top-down control) or external (e.g. stimulus blanking or visual transient stimulation) factors. Recently we have shown that both auditory and visual transient stimulation increased the frequency of immediate perceptual alternation (IMRF, 2008), suggesting that stability of visual perception was reduced by modality-independent transient events. Here we report that unpredictability of the external event is critical for the perceptual destabilization. Participants observed bistable apparent motion (''quartet dots'') wherein two dots could be seen as moving vertically or horizontally, and kept reporting the perceived direction of motion by button press. Task-irrelevant visual (100ms flash of background display) or auditory (100ms beep sound) transient stimuli, which the participants were explicitly instructed to ignore, were present at either random (Experiment 1; 4-12 sec) or constant (Experiment 2; 4 sec) intervals. In separate sessions, the flash and the beep were presented either in isolation (flash-only or beep-only conditions), synchronously (synchronous condition), or asynchronously (asynchronous condition). Each session lasted for 4 minutes and was repeated 2 times (8 sessions in total). Results showed that the increase in perceptual alternation occurred mostly within the period of less than 2 seconds after the visual and auditory transient in Experiment 1 (random interval). The effect magnitudes were similar irrespective of modality and correlated within participants between visual and auditory stimulation. In addition, the synchronous condition did not promote the destabilization further, suggesting that the destabilization is not additive among modalities. Interestingly, the perceptual destabilization was not observed in Experiment 2 (constant interval). Since the increase in perceptual alternation occurred less than 2 seconds after the transient stimulation, the shorter fixed interval (4 seconds) in Experiment 2 was not likely to account for the discrepancy. These results suggest that the destabilization mechanism exerts modality-independent unpredictable (or salient) events to increase the probability of perceptual alternation. Alternatively, the stabilization mechanism is prone to disturbance by modality-independent unpredictable events. The functional role of the process will be discussed.

Implicit Learning of Mirror Symmetries

Jinghua Tang, East China Normal University
Xiuyan Guo, East China Normal University
Zoltan Dienes, University of Sussex

Sunday, June 7th, 17:30–19:30: Poster No. 118

Research into implicit learning shows that people can implicitly learn the n-gram structure of sequences. A type of structure that goes beyond n-grams is symmetry. Yet detecting symmetry allows shorter (hence faster) encodings of stimuli and may be evolutionarily useful in e.g. detecting good genes. Thus, we reasoned implicit learning may become sensitive to symmetry, specifically mirror symmetries, above and beyond n-gram structure. In the first experiment, people were exposed to a series of numbers and for each number indicated whether it was larger or smaller than previous one. For each block of 12 numbers, the second half was a retrograde (or, in a different group, an inverse-retrograde) of the first half (the ''far'' rules), while during each half, the second quarter was an inverse-retrograde (or retrograde) of the first quarter (near rules). Over sixty-four blocks, we compared reaction times of mirrored parts to the initial random parts they mirrored. Participants' reaction times showed they learnt the inverse retrograde symmetry between quarters (near rules) but not the symmetries between halves of blocks (the far rules). That is, people could learn to detect symmetries between sets of three trials but not sets of six trials, which may have exceeded the memory buffer of the implicit learning system. In the second experiment, we explored the unconscious learning of symmetry in movement. Subjects walked around a circle with eight points, walking out sequences of seven movements, apparently as an exercise in Zen meditation. The last three movements of each sequence were the retrograde (or inverse) of the first three. After 24 training sequences obeying the retrograde (or inverse) rules, subjects were asked to either rate their liking or else classify 20 further walking sequences. Subjects liked the symmetry they were trained on even though they could not classify which sequences obeyed the training structure. Both experiments indicated that people could learn mirrored rules, under certain conditions, consistent with an ability to implicitly learn more than n-grams (in fact, being able to learn an “operation over variables', difficult for some computational models of learning).

Differential Patterns of Spontaneous Phenomenological Response to a Hypnotic Induction

Devin Terhune, Lund University
Etzel Cardela, Lund University

Sunday, June 7th, 17:30–19:30: Poster No. 102

Previous research indicates that individuals exhibit heterogeneous patterns of spontaneous experiences following a hypnotic induction. Such variability is explained in part by hypnotic suggestibility but is also present among highly suggestible individuals. Typological models have proposed that highly suggestible individuals are comprised of dissimilar types of respondents who differ in multiple dimensions of hypnotic responding. This study sought to discern phenomenological classes from the spontaneous experiential responses to a hypnotic induction and to assess whether experiential variability among highly suggestible individuals conforms to a typological pattern. Six hundred and forty individuals experienced the Waterloo-Stanford Group Scale of Hypnotic Susceptibility, Form C (WSGC; Bowers, 1998), which consists of a hypnotic induction followed by a series of suggestions. A two-minute resting epoch was embedded within the WSGC. Participants retrospectively completed the Phenomenology of Consciousness Inventory (PCI; Pekala, 1991) in reference to their experiences during the resting epoch. Five phenomenological state factors derived from the PCI (dissociated control, positive affect, negative affect, visual imagery, and attention to internal processes) were submitted to a latent profile analysis. The fit indices and likelihood ratio tests of multiple models were contrasted and participants were assigned to a class on the basis of the posterior probabilities of the best fitting model. A four-class model exhibited the strongest fit to the data. The first and second classes were comprised of individuals from all levels of hypnotic suggestibility, whereas the third and fourth classes were comprised of only low and medium suggestible individuals. Highly suggestible individuals were divided evenly between the first and second classes, which differed in negative affect, dissociative experiences, and attentional absorption. These results indicate that highly suggestible individuals do not exhibit a homogeneous experiential response pattern to a hypnotic induction and provide support for typological models of high hypnotic suggestibility. References Bowers, K.S. (1998). Waterloo-Stanford Group Scale of Hypnotic Susceptibility, Form C: Manual and response booklet. International Journal of Clinical and Experimental Hypnosis, 46, 250-268. Pekala, R.J. (1991). The Phenomenology of Consciousness Inventory. West Chester, PA: Mid-Atlantic Educational Institute, Inc.

Binding, Attention and Consciousness in the Multisensory Brain

Thomas Thesen, New York University
Mark Blumberg, New York University
Chad Carlson, New York University
Werner Doyle, New York University
Amy Trongnetrpunya, New York University
Ruben Kuzniecky, New York University
Orrin Devinsky, New York University
Charles Spence, Oxford University
Eric Eric, UC San Diego

Sunday, June 7th, 17:30–19:30: Poster No. 88

Information about events in our environment reaches the brain via several distinct sensory channels that, at the peripheral level, convert quite different physical forces into neural impulses. Our brain has the ability to integrate these inputs across sensory modalities to create a coherent and unified object representation. The processing of these signals and, consequently, their interpretation by the brain, may depend on brain state modulated endogenously by internal motivations. The current study investigates the neurophysiological basis of feature binding, attention and task demands across the visual and tactile modalities by means of intracranial EEG recordings directly from the human brain with high temporal and spatial resolution. Subjects were stimulated with brief tactile taps on the thumb and index finger with simultaneous LED flashes at the same locations. Each task employed stimulus conditions that consisted of bimodal congruent, bimodal incongruent or unimodal tactile or visual stimulation. Attention to modality varied between blocks. In Experiment I, subjects made speeded button responses to any stimulus in the target modality, irrespective of location. In Experiment II, subjects were instructed to maintain a different task set and made speeded elevation discrimination responses to stimuli in the target modality. Data from intracranial electrodes were analyzed in the time and frequency domains to compute ERPs and event-related power changes across multiple frequency bands. We show the spatio-temporal dynamics of feature binding across sensory modalities and their neurophysiological modulation by task demands and attention, shedding light on the neural mechanisms underlying top-down influences on perception and the unity of conscious experience.

Implicit Learning of Subliminal Material

Bert Timmermans, Universitè Libre de Bruxelles
Anne Atas, Universitè Libre de Bruxelles
Axel Cleeremans, Universitè Libre de Bruxelles
Sid Kouider, Laboratoire de Sciences Cognitives et Psychology

Saturday, June 6th, 17:30–19:30: Poster No. 111

This research aims to establish implicit learning of subliminally presented material using different implicit learning paradigms. In a typical implicit learning paradigm - the serial reaction time task - subjects are presented with a sequentially structured series of stimuli and are required, on each trial, to press the key corresponding to one of four possible stimuli. Reaction times decrease progressively during learning, and increase when the pattern of the stimuli changes. Performance in this task has been taken to indicate the existence of dissociations between ability to learn about the sequential contingencies between successive stimuli and awareness of these same contingencies. Typical results, however, systematically indicate some awareness of the relationships between successive stimuli. To address this issue so as to provide a more convincing demonstration of unconscious learning, we explored whether it is possible to learn about sequential contingencies between subliminally presented stimuli. The first two experiments aimed at influencing implicit learning of a supraliminal sequence through presentation of backward- and forward-masked subliminal primes before the onset of each visible stimulus. Crucially, while the supraliminal sequence remains the same throughout the experiment, the subliminal sequence of primes changes in a transfer block. Results show substantial decreases in performance when the subliminal sequence is changed, which suggests that the primes were indeed processed and modulated learning of the supraliminal sequence. In a third experiment, participants respond to a sequence consisting alternatively of visible and invisible items. The visible items do not constitute a regular sequence, but are completely predicted by the combination of the previous visible and invisible item. Again, participants show decreased performance when the subliminal items change. In a fourth experiment, participants are presented with a visible sequence, each item of which is presented together with a subliminal item. Unbeknownst to them, the subliminal items also constitute a sequence. The transfer block consists of visible and invisible sequences switched. We predict that exposure to the sequence in a subliminal manner will positively influence performance when the sequence is made visible.

Decoding Implicit Consumer Choices from Brain Signals

Anita Tusche, Bernstein Center Computational Neuroscience
John-Dylan Haynes, MPI for Human Cognitive and Brain Sciences, Leipzig

Saturday, June 6th, 17:30–19:30: Poster No. 135

In everyday life decision makers often face situations in which they lack choice-relevant information and are pressured for time. Even when sufficient information is given people are not very good in computing ''right'' choices by thorough, conscious deliberation. Instead a period of ''unconscious thought'' after information presentation was recently found to lead to sound choices among complex products (Dijksterhuis et al., 2006). Does that mean that the human brain does not require conscious deliberation for making choices among products? Do automatic brain processes guide a subsequent choice without the involvement of conscious deliberation and even attention directed to the decision? To address this question we implemented two independent experiments using functional magnetic resonance imaging (fMRI). In a first experiment, we investigated the neural processing of consumer products while subjects actively evaluated the products. Male participants were instructed to rate the attractiveness of each car they were presented with. In a second experiment, we studied the predictability of choices from brain activity while products were presented outside the focus of attention. Instead of evaluating the products, the participants were now engaged in a demanding visual fixation task while task-irrelevant images of cars were presented to the background of the screen. After the scanning session participants of both experiments were asked about their willingness to buy each of the presented cars. Importantly, during the experiments subjects were not informed that they would have to make such a purchase decision later on. Multivariate pattern classification was then applied to the imaging data in order to search through the whole brain for information predicting these subsequent consumer choices. We found that neural activation patterns in medial prefrontal cortex and insula predict subsequent consumer choices in both experiments. Interestingly, the amount of predictive information encoded in these areas was just as high even when subjects were not attending to the product. These findings suggest that the brain can evaluate consumer products even in the absence of explicit deliberation and attention to products and purchase choices. Moreover, the results indicate that information encoded this way is sufficient to reliably predict consumer choices later on.

Mechanisms of Masked Priming: A Meta-Analysis

Eva Van den Bussche, Campus Kortrijk, University of Leuven
Wim Van den Noortgate, University of Leuven - Campus Kortrijk
Bert Reynvoet, University of Leuven

Sunday, June 7th, 17:30–19:30: Poster No. 120

The extent to which unconscious information can influence behavior has been a topic of considerable debate throughout the history of psychology. A frequently used method for studying subliminal processing is the masked priming paradigm. The current meta-analyses focus on studies using this paradigm. Our aim was twofold: first, to assess the magnitude of subliminal priming across the literature and to determine whether subliminal primes are processed semantically; second, to examine potential moderators of priming effects. We found significant priming in our analyses, indicating that unconsciously presented information can influence behavior. Furthermore, priming was observed under circumstances in which a non-semantic interpretation could not fully explain the effects, suggesting that subliminally presented information can be processed semantically. Nonetheless, the non-semantic processing of primes is enhanced and priming effects are boosted when the experimental context allows the formation of automatic stimulus-response mappings. Our quantitative review also identified several moderators that influence the strength of priming.

Age Effects on Attentional Blink Performance in Meditation

Sara van Leeuwen, Max-Planck-Institut für Hirnforschung
Notger Müller, Magdeburg-Universitätsklinik für Neurologie
Lucia Melloni, Max Planck Institute for Brain Research

Sunday, June 7th, 17:30–19:30: Poster No. 126

Here we explore whether mental training, in the form of meditation, can aid in overcoming age related attentional decline. To investigate this, we compared performance on the attentional blink task between three populations: A group of long-term meditation practitioners within an older population; a group of age and education-matched control subjects and a group of young control subjects who had never undergone any form of meditation practice. In the attentional blink task subjects have to detect two targets embedded among distracters. Upon detection of the first target, subjects show a substantial reduction in accuracy for detecting the second target when it appears between 200 and 500 ms after the first. This drop in performance is referred to as the attentional blink and is thought to arise due to an inability to efficiently distribute limited attentional resources. Research has shown that the blink is increased in that it is both steeper and broader in aging populations. The broader blink effect in aging populations is thought to reflect a reduced ability to sustain attention over time. Meditation is expected to counteract these aging effects on attentional blink performance in that the practice of meditation lays much emphasis on the training of attention. Our results show that long term meditation practice leads to a reduction in the attentional blink. Furthermore, meditation practitioners taken from an older population, performed comparably to a control group taken from a younger population, whereas the control group, age-matched to the group of meditation practitioners, does not only exhibit deficits in performance for shorter delays, typically known as the AB phenomenon but also at longer delays. On the contrary, meditation practitioners do not show any of these deficits in performance. These results support the hypothesis that meditation practice can: (i) alter the efficiency with which attentional resources are distributed and (ii) help to overcome age-related attentional deficits both in the ability to distribute attentional resources in the temporal domain and in the sustaining of attention over time.

Modulation of Sensory and Pain Processing with Hypnosis: a Thulium-YAG Event Related fMRI Study

Audrey Vanhaudenhuyse, Coma Science Group, Cyclotron Research Center, University of liège
Melanie Boly, Cyclotron Research Centre, Belgium
Evelyne Balteau, Cyclotron Research Center,University of liège
Caroline Schnakers, Cyclotron Research Centre, liège, Belgium
Gustave Moonen, University of liège
Andrè Luxen, Cyclotron Research Center,University of liège
Jean-Francois Brichant, Department of Anesthesiology
Pierre Maquet, Cyclotron Research Centre, Belgium
Steven Steven, Cyclotron Research Centre, liège, Belgium
Marie-Elisabeth Faymonville, Clinic of Pain, University Hospital of liège

Saturday, June 6th, 17:30–19:30: Poster No. 101

Background: The neural mechanisms underlying hypnosis and especially the modulation of pain perception by hypnosis remain unclear. Using fMRI, we here describe how the hypnotic state alters the perception of sensory and painful stimulation. Methods: We used a parametric single-trial thulium-YAG laser fMRI paradigm in 13 healthy volunteers. Two sessions were performed on different days: one during normal wakefulness and one during hypnotic state. During each session, 200 laser stimuli with an intensity ranging from 300 to 600 mJ were administered on the dorsum of the subjects' left hand as described elsewhere (Boly et al, PNAS 2007). After each stimulus, subjects rated their sensory perception on a five-point scale (P0, no perception; P1, perceived not painful; P2, mild; P3, moderate and P4, severe pain). A repeated measures ANOVA compared sensation scores in normal wakefulness and hypnotic state, separating stimuli into low sensory range (i.e., <450 mJ) and high noxious range (i.e., ³; 450 mJ) intensities (using SPSS 14.0). Cerebral activations induced by the intensity-matched laser stimuli in normal wakefulness and in the hypnotic state were compared using statistical parametric mapping (SPM5). Results: As compared to normal wakefulness, subjects' ratings of perception decreased during the hypnotic state for both sensory (mean ± standard deviation 0.5±0.2; 0.3±0.2; p<0.001; respectively) and noxious stimuli (1.8±0.4; 1.3±0.4; p<0.001). Laser induced cerebral activation was decreased during hypnosis in anterior cingulate, right prefrontal and right premotor cortices, as well as in downstream regions encompassing brainstem, right thalamus, bilateral striatum, bilateral insula and right primary somatosensory cortices. Conclusion: These findings point to an important role for both the cortical pain neuromatrix and hierarchically ''lower-level'' brain areas in a hypnosis-induced decrease of sensory and affective aspects of stimulus perception. It reinforces the idea that not only pharmacological but also psychological strategies for relieving pain can modulate the network of cortical and subcortical regions that participate in the processing of noxious external stimuli.

Learning to be Conscious

Astrid Vermeiren, Universitè Libre de Bruxelles
Axel Cleeremans, Universitè Libre de Bruxelles

Saturday, June 6th, 17:30–19:30: Poster No. 11

The neural workspace model, often discussed in the consciousness literature, is relatively inflexible and has little or nothing to say about the relationship between learning and consciousness. Yet, anecdotal reports abound, for instance, that people who have become expert in conducting subliminal perception experiments actually consciously experience the very primes that normal participants fail to perceive! Expertise can thus modulate availability to awareness. We investigated this issue by exploring whether extended training on processing subliminally presented materials improves performance and awareness. We performed a training experiment where participants learned to identify subliminally presented numbers for five days. The numbers were masked and there was a varying time interval between the number presentation and the postmask presentation. On each trial, participants had to identify the number stimuli and indicate on a four-point PAS scale (Ramsy & Overgaard, 2004) what was the visibility of the stimulus. We observed not only that objective identification performance improved but also that the subjective feeling of visibility increased with training. This suggests that training can decrease the threshold for conscious perception. We discuss these results in the perspective of the radical plasticity thesis, stating that ''consciousness emerges in systems capable not only of learning about their environment, but also about their own internal representations of it''. We argue that learning increases the quality of representations, making them more easily available to awareness.

Why Phenomenal Consciousness Cannot be Explained by Content

Gottfried Vosgerau, Institut für Philosophie, Ruhr-Uni Bochum

Saturday, June 6th, 17:30–19:30: Poster No. 47

The phenomenal character of conscious experience is often explicated by some special kind of content: phenomenal content, the content of a higher-order thought, non-conceptual content, etc. It is then said that the proposed kind of content is able to explain the phenomenal character (sometimes among other factors). However, since mental representations are introduced to explain complex behavior, the content of mental representations is ultimately determined by the target object of the behavior to be explained (whatever additional criteria are formulated). Importantly, the target object plays an essential role for the acquisition of contents but not for the possession: Once a specific content is acquired, it is possible to have a mental representation with this content without there being the represented object (these are cases of misrepresentations). Thus, the content of a given mental representation is fixed by the behavioral pattern it explains, where the behavioral pattern is individuated by its usual target object. The difference between conscious and unconscious representations does not lie in the target of the behavior (resp. the behavioral pattern) it helps to explain - this can be demonstrated with a lot of different pathological impairments of consciousness. For example, patients suffering from visual form agnosia are able to act according to the ''perceived'' form of an object although they are unable to consciously register this form. Hence, the content of their unconscious representation is the same as the content of the conscious representation in healthy subjects (both can engage in a behavior targeting at the form). Hence, the content of a mental representation (whatever kind of content it is) cannot explain the phenomenal character of conscious experiences. In other words: there is no such thing as ''phenomenal content''. This finding implies that a lot of classical arguments and theories in philosophy are not sound (e.g. the argument for p-consciousness by Block, the higher-order thoughts theories). Moreover, it shows that the investigation of the content of conscious representations does not help us in understanding phenomenal consciousness. Rather, empirical research has to focus on the processing of mental representations in order to shed light on phenomenality.

Blaming the Brain? A Neurophilosophical Perspective on the Neuroscience of Aggression

Henrik Walter, University of Bonn

Sunday, June 7th, 17:30–19:30: Poster No. 34

Interpersonal violence and aggression are serious problems in almost every society. Apart and above from political, sociological and psychological approaches, neuroscience has made considerable progress in understanding its underpinning brain mechanisms [1, 2]. But does a neurobiological understanding of violence implicates excusing it? Can we blame our brains for our aggressive acts? These questions have gained even more weight, as in the last decade the problem of free will has gained considerable interest not only in philosophy, but also in science and in the public [3,4]. Some approaches to free will argue that a neuroscientific account of volition leaves no room for personal responsibility (hard determinism), whereas others, while acknowledging the scientific findings, argue that they have no impact for moral responsibility (conservative compatiblism). I will argue that both approaches are flawed: Hard determinism is wrong, because even if accepting its premises, there is room and reason to distinguish between different types of violent acts calling for a revised version of responsibility. Conservative compatibilism is wrong, because even if none of our practices in fact will change, as claimed by this position, the justifications for ascribing responsibility should change, which is a revision in its own. The position I defend, is called ''revisionist compatibilism'' [5]. I will use aggression as a prime example for it, present a survey of recent neurobiological findings, and will show why a revisionist position is best suited to deal with them.[1] Siever J. Neurobiology of violence and aggression. Am J Psychiatry 2008; 165:429-442. [2] Buckholtz J & Meyer-Lindenberg A. The neurogenetic architeture of aggression.TINS 2008; 31: 120-9. [3] Walter H. Neurophilosophy of free will. MIT Press 2001. [4] Haggard P. Human volition: the neuroscience of will. Nat Rev Neurosci 2008; 9:934-946. [5] Walter H. Neurophilosophy of Moral Responsibility: The Case for Revisionist Compatibilism. Philosophical Topics 2004; 32: 477-503.

The Experience of Attention

Sebastian Watzl, Columbia University

Sunday, June 7th, 17:30–19:30: Poster No. 40

The phenomenology of perceptual experience is shaped by attention. Imagine listening to a jazz band. Your experience has one kind of phenomenology when you are focusing on the sound of the piano, and a different phenomenology when you are focusing on the trombone. Or, to consider a visual example, imagine being in the subway: you have one visual phenomenology when you are focusing on your newspaper and a different one when (maybe without having moved your eyes) you are focusing on your neighbor. But what exactly is the contribution of attention to perceptual phenomenology? Often attention is said to highlight a certain part of our perceptual experience, or to make it more prominent. Yet what does this prominence amount to? Does your environment necessarily appear different when you focus on one aspect rather than another, as if the object of your attention gets "colored" with prominence? The answer, I believe, is no: the phenomenology of attention cannot be understood in terms of a particular way our environment appears to be (a certain environmental content). I argue against four different versions of such a view: (a) that attention enables conscious experience tout court (which implies that all contents of experience are attentional contents) (b) that attention enables the presentation of objects, (c) that attention presents relatively determinate or specific contents, (d) that attention presents a primitive or relational property of prominence. As alternative I suggest that the phenomenal shaping of experience by attention consists in part in an experience of attention itself: we are perceptually aware of ourselves and our own active mental attending. I end by suggesting different ways of conceptualizing that self-awareness and by an outlook into the consequences of my conclusion.

Awareness of Own and Observed Actions: Social Role Influences Feelings of Control

Dorit Wenke, Max Planck Institute for Human Cognitive and Brain Sciences
Roland Pfister, Julius Maximilian University Würzburg
Sukhvinder Obhi, Wilfrid Laurier University, Canada
Antje Holländer, MPI for Human Cognitive and Brain Sciences
Wolfgang Prinz, MPI for Human Cognitive and Brain Sciences

Sunday, June 7th, 17:30–19:30: Poster No. 72

The experience of causing an event alters subjective time: actions and their effects are subjectively perceived as temporally shifted towards each other. In the present study we investigated temporal binding in a social setting in which two participants performed actions at different points in an action sequence: One participant (the ''leader'') initiated the sequence by pressing a key at a time her choosing. The leader's action produced a tone that served as a go-signal for the other participant (the ''follower'') to press his key. Participants either estimated the length of the interval between the first (leader's) keypress and the tone, between the tone and the second (follower's) keypress, or between the follower«s keypress and the ensuing second tone. First results indicate that all intervals are perceived to be shorter when judged from the leader perspective than when judged from the follower perspective. This result suggests that leaders experience more control over the events in the action sequence they initiate than do followers who merely respond.

Measures of Knowledge Access in Artificial Grammar Learning Task

Michal Wierzchon, Jagiellonian University, Krakow
Maciej Taraday, Jagiellonian University, Krakow
Anna Hawrot, Maria Curie-Sklodowska University, Lublin
Michal Wierzchon, Jagiellonian University, Krakow
Dariusz Asanowicz, Jagiellonian University, Krakow

Saturday, June 6th, 17:30–19:30: Poster No. 121

The aim of the presentation is to compare the access consciousness measures applied in implicit learning studies. Most of the researchers have agreed that the accuracy of the indirect tests performance reflect in fact both conscious and unconscious influences of memory. To assess the ability of conscious access to the information stored in the memory multiple methods were proposed, such as confidance ratings or more recently post-decision wagering. In the presentatation the results of those two methods will be compared with the feeling of warmth scale data. It was assumed that feeling of warmth, as a more sensitive measure, would be correlated with the classification accuracy to more extend than other scales. To address the problem, the two artificial grammar learning experiments were conducted. In the classification phase of both experiments, the declarative ratings of confidence, feeling of warmth and post-decision wagering were collected. The results indicate the classification above the chance level in all experimental groups. The higher classification accuracy was related with the higher feeling of warmth declared by the participants and high confidence ratings in both experiments. The results of post-decision wagering were not related to the classification performace. The results of the studies seem to confirm that warmth scale is an sensitive measure of knowledge access. The scale is also more understandable for paritcipants. Finally the results showed significant difference between the post-wagering scale and other two scales results suggesting, that those measures could reflect other aspect of access consciousness.

Consciousness and Targetless Higher-order Thoughts

Jonah Wilberg, University of Essex

Saturday, June 6th, 17:30–19:30: Poster No. 35

In this paper I defend higher-order thought (HOT) theories of consciousness against a prominent objection. The central claim of HOT theory is that a mental state is conscious only if one has, or is disposed to have the HOT that one is in that state. According to the objection, HOT theory is unable to account for cases in which the relevant HOT misrepresents, or occurs in the absence of its target. In response, I show that there is a coherent account HOT theory can offer of such cases. I begin by evaluating different ways of drawing the distinction between 'misrepresentative' HOTs that merely misrepresent their targets, and 'targetless' HOTs that occur in the absence of their targets. I argue that the objection is most forcefully presented in terms of targetless HOTs, where these are understood as HOTs whose intentional content fails to correspond to any existing mental state tokens. There are two main responses to this form of the objection, I argue. One of these is defended by David Rosenthal, who claims that the mental state token corresponding to the intentional content of the targetless HOT is conscious, despite failing to exist. I argue that this response contradicts both a popular conception of state consciousness, and an implicit principle of HOT theory. A more successful response, I argue, is to say that no mental state token is conscious in virtue of a targetless HOT. This response is often briskly dismissed as inconsistent or strongly counter-intuitive. I show, however, that this response can be consistently developed, and that some realistic examples (Anton's syndrome, peripheral vision) increase its intuitive plausibility. Moreover, I argue, those consequences of this response which seem most counter-intuitive are also consequences of numerous commonly accepted views. For example, my response entails that one can seem to be in a conscious state without actually being in one. But this is in any case a consequence of various philosophical accounts of self-knowledge, or so I argue. We thus have a promising defence of HOT theory against the objection from targetless HOTs.

Moving Beyond the Common Bias That Consciousness is an Evolutionary Adaptation

Juliane Wilcke, University of Canterbury, New Zealand

Saturday, June 6th, 17:30–19:30: Poster No. 93

Is consciousness an evolutionary adaptation? Only few have claimed that it is not, amongst them Thomas Huxley, Stephen Jay Gould, David Rosenthal (about state consciousness) and Colin Blakemore (about phenomenal consciousness). Yet despite the general consensus that consciousness is an adaptation, most of the arguments for this position have failed. Because it is important that we understand why, I discuss problems with five arguments, namely, the arguments from the evolution of consciousness, from its subjective centrality, from the standard view, from the absence of opposing evidence and from the cost of consciousness. An important factor in the arguments' failures seems to be an underlying bias that consciousness is an evolutionary adaptation. This is directly visible in two of the arguments and suggested more generally both by the nature of the mistakes in the arguments and a lack of attempts at providing support. The adaptation bias is widespread, as evinced by the small number of opponents and the repeated claim that the default position is that consciousness is an adaptation. Consciousness researchers should be concerned about this bias because it impedes progress in increasing our knowledge about consciousness. Fortunately, there is a simple remedy for the adaptation bias: Mere awareness of the bias can help to neutralise it by highlighting the need for good scientific evidence for-or against-consciousness being an adaptation. And there are methods for obtaining such evidence which are more promising than the arguments above. For example, a more credible preliminary conclusion about the adaptation status of consciousness could come from a combination of evidence for special design, species comparisons and the exclusion of plausible alternative hypotheses. Furthermore, instead of limiting progress by asking exclusively whether or not consciousness is an adaptation, we should examine its potential evolutionary functions, consider different stages of its evolutionary development and take into account contextual factors such as selection pressures. To increase the amount and quality of research on the evolution and function of consciousness, we need to use good methods from disciplines such as evolutionary biology, evolutionary psychology and evolutionary cognitive neuroscience to obtain as much relevant evidence as possible.

Does the Brain's Clock Work Better Than the Mind's Clock?

Agnieszka Wykowska, Ludwig-Maximilians-Universität
Valtteri Arstila, University of Turku, Turku
Niko Busch, Universitè Paul Sabatier, Toulouse

Sunday, June 7th, 17:30–19:30: Poster No. 28

With the use of the ERP methodology, we investigated the question of whether the ''brain's clock'' has a better temporal resolution as compared to the ''mind«s clock''. That is, whether the brain can differentiate stimuli more accurately than we can consciously do. If so, we predicted to find two types of ERP effects: one that mirrors the physical (a)synchrony between stimuli and a second type that reflects the consciously perceived temporal pattern. Participants made simultaneity judgments concerning two types of visual stimuli that were presented either simultaneously, with 30 ms or with 60 ms stimulus onset asynchrony (SOA). As the 30 ms SOA condition was just at the threshold of perceived asynchrony (Pöppel, E., 1994), it yielded a rate of 50% perceived synchrony. ERP waveforms elicited by the second stimulus were compared between three conditions: perceived simultaneity with simultaneous stimuli, perceived simultaneity with 30 ms SOA and perceived asynchrony with 30 ms SOA. The results showed that the brain was able to differentiate between asynchronous and simultaneous stimuli with high temporal resolution at the sensory stages of processing. However, this distinction was not long-lasting and the 30-ms asynchronous stimuli (short SOA) were processed similarly to simultaneous stimuli at later stages. This probably yielded the effect of perceived simultaneity in 50% of trials of the short SOA condition. We conclude that early neural processing stages exhibit high temporal accuracy, but this precision is lost at later stages, which are presumably related to conscious temporal perception and decision making. Identification of these stages can provide insights into the neural underpinnings of conscious and unconscious temporal experience. Reference: Pöppel, E. (1994). A hierarchical model of temporal perception. Trends in Cognitive Sciences, 1(2), 56-61.

Is the Revival of Mind-body Identity Theory Possible?

Jack Yang, National Yang Ming University
Allen Houng, National Yang Ming University

Sunday, June 7th, 17:30–19:30: Poster No. 52

One of the most remarkable objections to the classic mind-brain identity theory, proposed by Place and Smart, is Saul Kripke's modal argument. Most materialists will reply by denying that the entailment from conceivability to metaphysical possibility is false. But this essay is an attempt to make a new way out for the identity theory from a different approach without discarding theory of rigid designator and ontological commitment on which Kripke is based. Thus, I will explain away conceivability of mind-body separability. First, I will reject Kripke's premise that the confusion between heat -reality- and sensation of heat –appearance- is analogically the only source of Cartesian intuition of separating mind-body. Instead, a new source of apparent conceivability, between pain and C-fiber firing, can be discovered and extinguished so that the identity theory will survive from Kripke's attack. I will propose that Cartesian intuition of mind-body separability comes from different identity (or boundary) conditions when we conceive, and thereby we have different limited conceptual frameworks to conceive properties of the referent. If we fix the identity condition, we can only conceive in the same limited conceptual framework and thereby explain Cartesian intuition away. For example: heat is ''occurrence of new degree of freedom'' in classic physics, but we can only conceive that heat is ''molecular motion'' after thermodynamic theories are provided and they fix the strict identity condition. Therefore, creating new theories and identity conditions will change our concept. That is why our Cartesian intuition can be extinguished after scientific theories are built strictly. Moreover, according to the causal theory of reference, after we baptize what pain is, we have learned to use it by the chain of communication. It means that if everyone designates the same thing in our world, usage of pain has to be accessed by others. Therefore, either phenomenology is not totally private, or our concept of ''pain'' cannot be referred to our phenomenological experience, but our functional and behavioral states. After all, based on strict theory and the identity condition, we eventually can reconsider the mind-body identification.