Université libre de Bruxelles Research inventory
Site en français
CLEEREMANS Axel



Units

Research center Cognition and Neurosciences

Person in charge of the Unit : Oui

The research unit Centre for Cognition and Neurosciences of the Faculty of Psychological Sciences and Education

Consciousness, Cognition & Computation Group

Person in charge of the Unit : Oui

The goal of the Co3 is to contribute to our understanding of the elementary cognitive processes involved in learning, cognitive development, and automaticity. One of the central research topics of the Co3 concerns the role of consciousness in these elementary processes, and in particular the issue of determining which can occur without awareness. The research projects of the Co3 generally combine behavioral methods (experimentation with healthy participants in a lifelong perspective and with brain-damaged patients), modelling (neural networks), and neuroimaging methods (PET, MEG et fMRI). Recently, a Babylab has been installed, in other words a laboratory dedicated to the study of the development of the infants and toddlers' cognitive and social abilities. http://babylab.ulb.ac.be/Bienvenue.html

Projetcs

Somatic markers and implicit learning

On the basis of Damasio et al.'s work (Bechara, Damasio, Tranel, et Damasio, 1997) and in collaboration with Dick Bierman (Universiteit Amsterdam), we have conducted an experiment during which participants must categorize exemplars from a finite state grammar in a context similar to that of the Iowa Gambling Task, that is, with reinforcement. Our results indicate a dissociation between verbal and electro-dermal responses, which suggests that Damasio's paradigm may be extended to more complex situations in which the structure of the stimuli can be learned.

Dissociation between conscious expectancy and behaviour in conditioning and simple reaction time paradigms

The biological reaction to an unconditioned stimulus can be transferred to a neutral stimulus through a conditioning procedure. A recurrent debate concerns the question of whether pavlovian conditioning in humans requires that the subject be conscious of the relationships between conditioned and unconditioned stimuli. In this context, we explore the contrast between two conditioning situations : Delay conditioning, in which US and CS overlap temporally, and Trace conditioning, in which US and CS are separated by a delay. These two preparations appear to engage consciousness in different ways, and the contrast between the two situations is similar to the temporal effects we have observed with sequence learning.

Modelling of the mechanisms involved in sequence learning

This project aims at developing a model of sequence learning that is able to (1) simulate the time course of information processing during serial reaction time tasks, and (2) to make it possible to account for the differences between incidental anticipation (upon which implicit learning is based in these situations) and conscious prediction. To do so, we use neurally inspired models, such as Elman's Simple Recurrent Network.

Addiction and gambling: A study of decision-making deficits in uncertain situations

Individuals with pathological gambling (DSM-IV-TR) keep gambling despite of the occurrence of negative consequences (e.g., financial ruin). The continuation of gambling practice may reflect poor decision making in the sense that immediate gratification or relief from an uncomfortable state is more influential that delayed negative consequences. Based on the neuroeconomic framework, situations in which decisions are being made may range from their level of uncertainty. Situations with 'ambiguity' defines situations characterized by an unknown outcome probabilities distribution.  Situation with 'risk' defines the class of situations with explicit distribution of outcome probabilities (e.g., 50/50 chance to win/lose).  	Recent neurocognitive studies reported that pathological gamblers (PG) are impaired in task that involved decision making under ambiguity (e.g., Goudriaan et al., 2005) or under risk (e.g., Brand, et al., 2007). However, little is known about the relationship between decision making under ambiguity and under risk in PG as well as the relative contribution of response inhibition, working memory and speed processing to these impairments. Furthermore, decision making deficits have only been investigated among PG willing to change something in their gambling habits (i.e., under treatment). This project aims at investigating separately decision making under ambiguity and under risk and their cognitive underlying processes in individuals with pathological gambling being or not under treatment. This project is to investigate separately decision making under ambiguity and under risk and their cognitive underlying processes in individuals with pathological gambling being or not under treatment.  

Neural correlates of the dynamics of learning

This project aims at identifying the cerebral regions involved in learning with and without consciousness through imaging methods (PET, fMRI and MEG). We explore both sequence learning and conditioning situations. A second line of research specifically concerns the time course of information processing. We explore what the neural correlates of response preparation are in conditions that either promote implicit or explicit learning. 

R.O.G.E.R. project : Realistic Observation in Game and Experience in Rehabilitation

Current tools for the assessment of cognitive abilities are based on neuropsychological tests designed to control the metric and to fragment cognitive functions. They are incomplete because they are too analytical and do not give an embedded profile. In this context, the project aims to develop simulated realistic environments to dive people into cognitive impairment in an immersive, captivating and realistic experience. Within the created virtual environment, patients will play scenarios close to daily activities. In neuropsychological terms, these activities are referred to as ecological situations. In virtual reality, the person feels less ''assessed'', which will increase both the quality of his experience and his involvement, thus rendering the collected data more reliable.

Measuring Consciousness: Bridging the Mind-Brain Gap

MindBridge is a collaboration between HNRU, ULB, UKE Germany (Andreas Engel), UCL UK (Geraint Rees), ULg Belgium (Steven Laureys), and PUC Germany (Kai Vogeley). The general aim of this EU Nest program project is to bridge the gap between the inner subjective (first-person)experience of consciousness, and the objective (third-person) observation of neural phenomena. Specifically, the aim of this work will be to develop experimental methodologies that allow for correlating of subjective and objective measures of awareness. In other words, the idea is to design tasks in which relative contributions of conscious and unconscious processing can be distinguished, so that once we have people do these tasks in an fMRI scanner, the observed neural activity may tell us something about the differences or similarities between conscious and unconscious processing.

Metarepresentation and consciousness

Certain theories of consciousness (« Higher-Order Thought theory », David Rosenthal) assume that conscious representations are representations that we are conscious of ; that is, that what makes knowledge conscious is the fact that I am aware that I possess this knowledge. We explore the implications of this view by attempting to simulate the emergence of systems of meta-representations (i.e., representations of representations) in neural networks.

Computational modeling of the learning of disjoint contingencies

This project aims at exploring the mechanisms that drive learning of disjoint contingencies (such as, for instance, involved in learning about number agreement between the head and the verb of a sentence, where the head and the verb are separated by several other words) in sequence learning and artificial grammar studies. 

Differences between continuous and discrete sequence learning

Just as discrete sequence learning, the learning of continuous mouvements requires a sensitivity to the temporal context in which actions are executed. However, despite the fact that an abundant literature has documented implicit learning when responses are discrete, this is not the case for the execution of continuous mouvements. This project aims at elucidating the differences between the two situations.

How memory structures information : behavioural approach and brain imaging investigation of a learning process, a ''change over time'' perspective

Le processus de consolidation est fondamental dans les processus mnésiques. Si l'on a très tôt accepté l'idée que ce processus renvoyait tant à une activité physiologique de réverbération au niveau neuronal qu'à une activité psychologique donnant lieu à un processus continu d'organisation de l'information, notre connaissance des processus cognitifs élémentaires et des mécanismes cérébraux impliqués dans l'apprentissage et la consolidation de l'information en mémoire reste encore imparfaite. Dans cette perspective, il apparaît que la stabilisation de l'information en mémoire pourrait dépendre du phénomène de chunking. En particulier, si de nombreux travaux ont clairement identifié la création de chunks (regroupement ou unité de sens) comme indispensable à toute stabilisation, le processus à l'origine de l'émergence de ces chunks apparaît encore bien mal documentée tant au niveau comportemental qu'au niveau neurophysiologique. Dans ce contexte, notre projet de recherche aura pour objectif de mieux comprendre comment s'organise l'information en MLT ainsi que les substrats neurophysiologiques de cette stabilisation. Nous aborderons cette question à travers deux tâches bien distinctes, à savoir une tâche de mémorisation visuo-spatiale et une tâche de mémorisation de séquence motrice. Pour chacune de ces tâches, nous couplerons une analyse comportementale de l'évolution du système au cours de l'apprentissage et une étude des modifications d'activité régionale cérébrale, l'objectif étant de mettre à jour les déterminants comportementaux et neurologiques du processus de chunking.

Relationships between consciousness and learning

Can we learn without consciousness ? On the other hand, what is the effect of learning on consciousness? Addressing this controversial issue forms the core of the ensemble of our projects. In this context, we explore three central hypotheses. (1) conscious knowledge implies higher-quality representations, as compared to implicit knowledge, (2) Representations in memory are consolidated by experience and constantly bias information processing, regardless of their quality, and (3) It takes time for high-quality representations to emerge. To explore these ideas, our research combines behavioral methods, computational modelling, and brain imaging methods.

Behavioral and brain dynamics of consciousness

Authors differ on (1) whether consciousness is a graded or a dichotomous phenomenon, and on (2) the cortical distribution of the brain regions involved in conscious perception. To contribute to resolving this controversy, we propose to examine the influence of the complexity of stimuli on subjective reportability and on objective performance, while both obtaining behavioral and neurophysiological measures. A second line of research is rooted in the fact that theory development is complicated by the question of how to measure subjective experience adequately. Therefore, two frequently used scales, which have so far led to incompatible results, will be directly compared in several experimental designs, to assess whether results obtained with the scales genuinely reflect subjective reports, or merely biases. Finally, we want to explore the features of the representations on which such subjective reports are based. To this end, computational models will be implemented, also allowing to empirically test recently outlined representational theories on consciousness.

Theory of Mind and Self-Awareness

What is the nature of the link between the theory of mind and the self-awareness ? Study of the development of these two abilities in infants and in bonobos (Pan Paniscus) will allow to answer more clearly to this question.

Temporal effects in information processing

This project's main goal is to explore the general hypothesis that consciousness takes time, at different scales : During a single trial (milliseconds-seconds), during learning of a task (minutes-hours), during cognitive development (years). We test the hypothesis by manipulating the temporal delays that separate successive trials ; by varying the duration of training ; by comparing information processing at different points in development. 

Comparing direct and indirect measures of learning

This study aims at exploring the role of consciousness in sequence learning, by means of adapting the two most fruitful methods of comparing direct and indirect measures of acquired knowledge, e.g., Merikle et Reingold's approach (1991) and Jacoby's Process Dissociation Procedure (Jacoby, 1991; Jacoby, Toth, et Yonelinas, 1993).

The mechanisms of serial action : Formation of cognitive units

A central hypothesis concerning the changes induced by experience is that people form « cognitive units » or chunks. In this project, we seek to document as precisely as possible the emergence of such chunks by analyzing performance in a 1023-choice reaction time task.  (Seibel, 1963). 

Neural bases of learning and long term consolidation of sequence regularities.

Au cours de ces dernières années, un débat grandissant s'est fait jour quant aux processus mis en jeu au cours de l'acquisition de régularités séquentielles complexes ainsi que par rapport au rôle joué par le sommeil pour leur consolidation en mémoire. Notre connaissance des processus cognitifs élémentaires et des mécanismes cérébraux impliqués dans l'apprentissage et la consolidation de séquences motrices reste encore imparfaite. Dans cette perspective, un paramètre crucial pourrait être le caractère discret vs. continu lors de l'acquisition d'une régularité séquentielle. Une question fondamentale reste de savoir dans quelle mesure des systèmes neuronaux distincts ou similaires sont activés lors de l'acquisition de séquences motrices discrètes et continues. De plus, nous souhaitons observer si la privation de sommeil gêne le processus de consolidation d'un apprentissage implicite de séquences complexes en empêchant une restructuration des circuits cérébraux sous-tendant la performance dans différentes tâches visuomotrices (tâches de Temps de Réaction Sériel --TRS - et tâche de poursuite de cibles). L'ensemble de ces recherches sera réalisé au moyen de techniques d'imagerie cérébrales (IRMf anatomique et structurelle) 

Hypnosis, non hypnotic suggestion and placebo effect: Looking for the link!

The main goal of this study is to explore what are the cognitive and neural mechanisms underlying hypnosis, suggestion and the placebo effect. We hypothesize that these three phenomena described so far as being underpinned by different mechanisms are in fact controlled by a set of common processes. We assume, moreover, that this process triggers the induction of states of consciousness in which mental representations of high level (e.g., beliefs) are able to modulate the processing of information in an unconscious and unintentional manner.