Plasticity in body and peripersonal space representations

A successful interaction with objects in the environment requires integrating information concerning object-location with the shape, dimension and position of body parts in space. The former information is coded in a multisensory representation of the space around the body, i.e. peripersonal space (PPS), whereas the latter is enabled by an online, constantly updated, action-orientated multisensory representation of the body (BR) that is critical for action. One of the critical features of these representations is that both PPS and BR are not fixed, but they dynamically change depending on different types of experience. In a series of experiment, I studied plastic properties of PPS and BR in humans. I have developed a series of methods to measure the boundaries of PPS representation (Chapter 4), to study its neural correlates (Chapter 3) and to assess BRs. These tasks have been used to study changes in PPS and BR following tool-use (Chapter 5), multisensory stimulation (Chapter 6), amputation and prosthesis implantation (Chapter 7) or social interaction (Chapter 8). I found that changes in the function (tool-use) and the structure (amputation and prosthesis implantation) of the physical body elongate or shrink both PPS and BR. Social context and social interaction also shape PPS representation. Such high degree of plasticity suggests that our sense of body in space is not given at once, but it is constantly constructed and adapted through experience.

Behavioral and neural correlates of perceived and imagined musical timbre

The generality of findings implicating secondary auditory areas in auditory imagery was tested by using a timbre imagery task with fMRI. Another aim was to test whether activity in supplementary motor area (SMA) seen in prior studies might have been related to subvocalization. Participants with moderate musical background were scanned while making similarity judgments about the timbre of heard or imagined musical instrument sounds. The critical control condition was a visual imagery task. The pattern of judgments in perceived and imagined conditions was similar, suggesting that perception and imagery access similar cognitive representations of timbre. As expected, judgments of heard timbres, relative to the visual imagery control, activated primary and secondary auditory areas with some right-sided asymmetry. Timbre imagery also activated secondary auditory areas relative to the visual imagery control, although less strongly, in accord with previous data. Significant overlap was observed in these regions between perceptual and imagery conditions. Because the visual control task resulted in deactivation of auditory areas relative to a silent baseline, we interpret the timbre imagery effect as a reversal of that deactivation. Despite the lack of an obvious subvocalization component to timbre imagery, some activity in SMA was observed, suggesting that SMA may have a more general role in imagery beyond any motor component.

Brain responses to auditory and visual stimulus offset: Shared representations of temporal edges

Edges are crucial for the formation of coherent objects from sequential sensory inputs within a single modality. Moreover, temporally coincident boundaries of perceptual objects across different sensory modalities facilitate crossmodal integration. Here, we used functional magnetic resonance imaging in order to examine the neural basis of temporal edge detection across modalities. Onsets of sensory inputs are not only related to the detection of an edge but also to the processing of novel sensory inputs. Thus, we used transitions from input to rest (offsets) as convenient stimuli for studying the neural underpinnings of visual and acoustic edge detection per se. We found, besides modality-specific patterns, shared visual and auditory offset-related activity in the superior temporal sulcus and insula of the right hemisphere. Our data suggest that right hemispheric regions known to be involved in multisensory processing are crucial for detection of edges in the temporal domain across both visual and auditory modalities. This operation is likely to facilitate cross-modal object feature binding based on temporal coincidence. Hum Brain Mapp, 2008. (c) 2008 Wiley-Liss, Inc.

Exploring the emotional brain: Neural correlates of homeostatic and sensory-evoked emotions and their interaction in emotional rivalry

Human emotions are essential for survival. They are vital for the satisfaction of basic needs, the regulation of personal life and successful integration into social structures. Depending on which aspect of an emotion is used in its definition, many different theories offer possible answers to the questions of what emotions are and how they can be distinguished. The systematic investigation of emotions in cognitive neuroscience is relatively new, and neuroimaging studies specifically focussing on the neural correlates of different categories of emotions are still lacking. Therefore, the current thesis aimed at investigating the behavioural and neurophysiological correlates of different human emotional levels and their interaction in healthy subjects. We differentiated between emotions according to their cerebral entry site and neural processing pathways: homeostatic emotions, which are elicited by metabolic changes and processed by the interoceptive system (such as thirst, hunger, and need for air), and sensory-evoked emotions, which are evoked by external inputs via the eyes, ears or nose, or their corresponding mental representations and processed in the brain as sensory perception (e.g. fear, disgust, or pride). Using functional magnetic resonance imaging (fMRI) and behavioural parameters, we examined both the specific neural underpinnings of a homeostatic emotion (thirst) and a sensory-evoked emotion (disgust), and their interaction in a situation of emotional rivalry when both emotions were perceived simultaneously. This thesis comprises three research articles reporting the results of this research. The first paper presents disgust-related brain imaging data in a thirsty and a satiated condition. We found that disgust mainly activated the anterior insular cortex. In the thirsty condition, however, we observed an interaction effect between disgust and thirst: when thirsty, the subjects rated the disgusting stimulus as less repulsive. On the neurobiological level, this reduction of subjective disgust was accompanied by significantly reduced neural activity in the insular cortex. These results provide new neurophysiological evidence for a hierarchical organization among homeostatic and sensory-evoked emotions, revealing that in a situation of emotional conflict, homeostatic emotions are prioritized over sensory-evoked emotions. In the second paper, findings on brain perfusion over four different thirst stages are reported, with a special focus on the parametric progression of thirst. Cerebral perfusion differences over all thirst stages were found in the posterior insular cortex. Taking this result together with the findings of the first paper, the insular cortex seems to be a key player in human emotional processing, since it comprises specific representations of homeostatic and sensory-evoked emotions and also represents the site of cortical interaction between the two levels of emotions. Finally, although this thesis focussed on the homeostatic modulation of disgust, we were also interested in whether dehydration modulates taste perception. The results of this behavioural experiment are described in the third paper, where we show that dehydration alters the perception of neutral taste stimuli.

Learning -dependent plasticity of movement representations in the primate primary motor cortex

Primary motor cortex (M1) is involved in the production of voluntary movement and contains a complete functional representation, or map, of the skeletal musculature. This functional map can be altered by pathological experiences, such as peripheral nerve injury or stroke, by pharmacological manipulation, and by behavioral experience. The process by which experience-dependent alterations of cortical function occur is termed plasticity. In this thesis, plasticity of M1 functional organization as a consequence of behavioral experience was examined in adult primates (squirrel monkeys). Maps of movement representations were derived under anesthesia using intracortical microstimulation, whereby a microelectrode was inserted into the cortex to electrically stimulate corticospinal neurons at low current levels and evoke movements of the forelimb, principally of the hand. Movement representations were examined before and at several times after training on behavioral tasks that emphasized use of the fingers. Two behavioral tasks were utilized that dissociated the repetition of motor activity from the acquisition of motor skills. One task was easy to perform, and as such promoted repetitive motor activity without learning. The other task was more difficult, requiring the acquisition of motor skills for successful performance. Kinematic analysis indicated that monkeys used a consistent set of forelimb movements during pellet extractions. Functional mapping revealed that repetitive motor activity during the easier task did not produce plastic changes in movement representations. Instead, map plasticity, in the form of selective expansions of task-related movement representations, was only produced following skill acquisition on the difficult task. Additional studies revealed that, in general, map plasticity persisted without further training for up to three months, in parallel with the retention of task-related motor skills. Also, extensive additional training on the small well task produced further improvements in performance, and further changes in movement maps. In sum, these experiments support the following three conclusions regarding the role of M1 in motor learning. First, behaviorally-driven plasticity is learning-dependent, not activity-dependent. Second, plastic changes in M1 functional representations represent a neural correlate of acquired motor skills. Third, the persistence of map plasticity suggests that M1 is part of the neural substrate for the memory of motor skills. ^

Transforming neural computations and representing time

Motifs of neural circuitry seem surprisingly conserved over different areas of neocortex or of paleocortex, while performing quite different sensory processing tasks. This apparent paradox may be resolved by the fact that seemingly different problems in sensory information processing are related by transformations (changes of variables) that convert one problem into another. The same basic algorithm that is appropriate to the recognition of a known odor quality, independent of the strength of the odor, can be used to recognize a vocalization (e.g., a spoken syllable), independent of whether it is spoken quickly or slowly. To convert one problem into the other, a new representation of time sequences is needed. The time that has elapsed since a recent event must be represented in neural activity. The electrophysiological hallmarks of cells that are involved in generating such a representation of time are discussed. The anatomical relationships between olfactory and auditory pathways suggest relevant experiments. The neurophysiological mechanism for the psychophysical logarithmic encoding of time duration would be of direct use for interconverting olfactory and auditory processing problems. Such reuse of old algorithms in new settings and representations is related to the way that evolution develops new biochemistry.

Neural mechanisms for visual memory and their role in attention

Recent studies show that neuronal mechanisms for learning and memory both dynamically modulate and permanently alter the representations of visual stimuli in the adult monkey cortex. Three commonly observed neuronal effects in memory-demanding tasks are repetition suppression, enhancement, and delay activity. In repetition suppression, repeated experience with the same visual stimulus leads to both short- and long-term suppression of neuronal responses in subpopulations of visual neurons. Enhancement works in an opposite fashion, in that neuronal responses are enhanced for objects with learned behavioral relevance. Delay activity is found in tasks in which animals are required to actively hold specific information “on-line” for short periods. Repetition suppression appears to be an intrinsic property of visual cortical areas such as inferior temporal cortex and is thought to be important for perceptual learning and priming. By contrast, enhancement and delay activity may depend on feedback to temporal cortex from prefrontal cortex and are thought to be important for working memory. All of these mnemonic effects on neuronal responses bias the competitive interactions that take place between stimulus representations in the cortex when there is more than one stimulus in the visual field. As a result, memory will often determine the winner of these competitions and, thus, will determine which stimulus is attended.

Neural codes: Firing rates and beyond

Computational neuroscience has contributed significantly to our understanding of higher brain function by combining experimental neurobiology, psychophysics, modeling, and mathematical analysis. This article reviews recent advances in a key area: neural coding and information processing. It is shown that synapses are capable of supporting computations based on highly structured temporal codes. Such codes could provide a substrate for unambiguous representations of complex stimuli and be used to solve difficult cognitive tasks, such as the binding problem. Unsupervised learning rules could generate the circuitry required for precise temporal codes. Together, these results indicate that neural systems perform a rich repertoire of computations based on action potential timing.

Learning representations for Information Retrieval

La recherche d'informations s'intéresse, entre autres, à répondre à des questions comme: est-ce qu'un document est pertinent à une requête ? Est-ce que deux requêtes ou deux documents sont similaires ? Comment la similarité entre deux requêtes ou documents peut être utilisée pour améliorer l'estimation de la pertinence ? Pour donner réponse à ces questions, il est nécessaire d'associer chaque document et requête à des représentations interprétables par ordinateur. Une fois ces représentations estimées, la similarité peut correspondre, par exemple, à une distance ou une divergence qui opère dans l'espace de représentation. On admet généralement que la qualité d'une représentation a un impact direct sur l'erreur d'estimation par rapport à la vraie pertinence, jugée par un humain. Estimer de bonnes représentations des documents et des requêtes a longtemps été un problème central de la recherche d'informations. Le but de cette thèse est de proposer des nouvelles méthodes pour estimer les représentations des documents et des requêtes, la relation de pertinence entre eux et ainsi modestement avancer l'état de l'art du domaine. Nous présentons quatre articles publiés dans des conférences internationales et un article publié dans un forum d'évaluation. Les deux premiers articles concernent des méthodes qui créent l'espace de représentation selon une connaissance à priori sur les caractéristiques qui sont importantes pour la tâche à accomplir. Ceux-ci nous amènent à présenter un nouveau modèle de recherche d'informations qui diffère des modèles existants sur le plan théorique et de l'efficacité expérimentale. Les deux derniers articles marquent un changement fondamental dans l'approche de construction des représentations. Ils bénéficient notamment de l'intérêt de recherche dont les techniques d'apprentissage profond par réseaux de neurones, ou deep learning, ont fait récemment l'objet. Ces modèles d'apprentissage élicitent automatiquement les caractéristiques importantes pour la tâche demandée à partir d'une quantité importante de données. Nous nous intéressons à la modélisation des relations sémantiques entre documents et requêtes ainsi qu'entre deux ou plusieurs requêtes. Ces derniers articles marquent les premières applications de l'apprentissage de représentations par réseaux de neurones à la recherche d'informations. Les modèles proposés ont aussi produit une performance améliorée sur des collections de test standard. Nos travaux nous mènent à la conclusion générale suivante: la performance en recherche d'informations pourrait drastiquement être améliorée en se basant sur les approches d'apprentissage de représentations.

Learning representations for Information Retrieval

La recherche d'informations s'intéresse, entre autres, à répondre à des questions comme: est-ce qu'un document est pertinent à une requête ? Est-ce que deux requêtes ou deux documents sont similaires ? Comment la similarité entre deux requêtes ou documents peut être utilisée pour améliorer l'estimation de la pertinence ? Pour donner réponse à ces questions, il est nécessaire d'associer chaque document et requête à des représentations interprétables par ordinateur. Une fois ces représentations estimées, la similarité peut correspondre, par exemple, à une distance ou une divergence qui opère dans l'espace de représentation. On admet généralement que la qualité d'une représentation a un impact direct sur l'erreur d'estimation par rapport à la vraie pertinence, jugée par un humain. Estimer de bonnes représentations des documents et des requêtes a longtemps été un problème central de la recherche d'informations. Le but de cette thèse est de proposer des nouvelles méthodes pour estimer les représentations des documents et des requêtes, la relation de pertinence entre eux et ainsi modestement avancer l'état de l'art du domaine. Nous présentons quatre articles publiés dans des conférences internationales et un article publié dans un forum d'évaluation. Les deux premiers articles concernent des méthodes qui créent l'espace de représentation selon une connaissance à priori sur les caractéristiques qui sont importantes pour la tâche à accomplir. Ceux-ci nous amènent à présenter un nouveau modèle de recherche d'informations qui diffère des modèles existants sur le plan théorique et de l'efficacité expérimentale. Les deux derniers articles marquent un changement fondamental dans l'approche de construction des représentations. Ils bénéficient notamment de l'intérêt de recherche dont les techniques d'apprentissage profond par réseaux de neurones, ou deep learning, ont fait récemment l'objet. Ces modèles d'apprentissage élicitent automatiquement les caractéristiques importantes pour la tâche demandée à partir d'une quantité importante de données. Nous nous intéressons à la modélisation des relations sémantiques entre documents et requêtes ainsi qu'entre deux ou plusieurs requêtes. Ces derniers articles marquent les premières applications de l'apprentissage de représentations par réseaux de neurones à la recherche d'informations. Les modèles proposés ont aussi produit une performance améliorée sur des collections de test standard. Nos travaux nous mènent à la conclusion générale suivante: la performance en recherche d'informations pourrait drastiquement être améliorée en se basant sur les approches d'apprentissage de représentations.

Correcting errors in optical data transmission using neural networks

Optical data communication systems are prone to a variety of processes that modify the transmitted signal, and contribute errors in the determination of 1s from 0s. This is a difficult, and commercially important, problem to solve. Errors must be detected and corrected at high speed, and the classifier must be very accurate; ideally it should also be tunable to the characteristics of individual communication links. We show that simple single layer neural networks may be used to address these problems, and examine how different input representations affect the accuracy of bit error correction. Our results lead us to conclude that a system based on these principles can perform at least as well as an existing non-trainable error correction system, whilst being tunable to suit the individual characteristics of different communication links.

Correcting errors in optical data transmission using neural networks

Optical data communication systems are prone to a variety of processes that modify the transmitted signal, and contribute errors in the determination of 1s from 0s. This is a difficult, and commercially important, problem to solve. Errors must be detected and corrected at high speed, and the classifier must be very accurate; ideally it should also be tunable to the characteristics of individual communication links. We show that simple single layer neural networks may be used to address these problems, and examine how different input representations affect the accuracy of bit error correction. Our results lead us to conclude that a system based on these principles can perform at least as well as an existing non-trainable error correction system, whilst being tunable to suit the individual characteristics of different communication links.

Searching for Nearest Strings with Neural-Like String Embedding

We analyze an approach to a similarity preserving coding of symbol sequences based on neural distributed representations and show that it can be viewed as a metric embedding process.

How different are the visual representations used for object recognition in middle childhood and adulthood?

Recent experimental studies have shown that development towards adult performance levels in configural processing in object recognition is delayed through middle childhood. Whilst partchanges to animal and artefact stimuli are processed with similar to adult levels of accuracy from 7 years of age, relative size changes to stimuli result in a significant decrease in relative performance for participants aged between 7 and 10. Two sets of computational experiments were run using the JIM3 artificial neural network with adult and 'immature' versions to simulate these results. One set progressively decreased the number of neurons involved in the representation of view-independent metric relations within multi-geon objects. A second set of computational experiments involved decreasing the number of neurons that represent view-dependent (nonrelational) object attributes in JIM3's Surface Map. The simulation results which show the best qualitative match to empirical data occurred when artificial neurons representing metric-precision relations were entirely eliminated. These results therefore provide further evidence for the late development of relational processing in object recognition and suggest that children in middle childhood may recognise objects without forming structural description representations.

To move or not to move, that is the question! Body schema and non-action oriented body representations:an fMRI meta-analytic study

Many studies have attempted to identify the different cognitive components of body representation (BR). Due to methodological issues, the data reported in these studies are often confusing. Here we summarize the fMRI data from previous studies and explore the possibility of a neural segregation between BR supporting actions (body-schema, BS) or not (non-oriented-to-action-body-representation, NA). We performed a general activation likelihood estimation meta-analysis of 59 fMRI experiments and two individual meta-analyses to identify the neural substrates of different BR. Body processing involves a wide network of areas in occipital, parietal, frontal and temporal lobes. NA selectively activates the somatosensory primary cortex and the supramarginal gyrus. BS involves the primary motor area and the right extrastriate body area. Our data suggest that motor information and recognition of body parts are fundamental to build BS. Instead, sensory information and processing of the egocentric perspective are more important for NA. In conclusion, our results strongly support the idea that different and segregated neural substrates are involved in body representations orient or not to actions.