Driss Boussaoud

Abstract: Functional magnetic resonance imaging (fMRI) exploits blood-oxygen-level-dependent (BOLD) contrasts to map neural activity associated with a variety of brain functions including sensory processing, motor control, and cognitive and emotional functions. The general linear model (GLM) approach is used to reveal task-related brain areas by searching for linear correlations between the fMRI time course and a reference model. One of the limitations of the GLM approach is the assumption that the covariance across neighbouring voxels is not informative about the cognitive function under examination. Multivoxel pattern analysis (MVPA) represents a promising technique that is currently exploited to investigate the information contained in distributed patterns of neural activity to infer the functional role of brain areas and networks. MVPA is considered as a supervised classification problem where a classifier attempts to capture the relationships between spatial pattern of fMRI activity and experimental conditions. In this paper , we review MVPA and describe the mathematical basis of the classification algorithms used for decoding fMRI signals, such as support vector machines (SVMs). In addition, we describe the workflow of processing steps required for MVPA such as feature selection, dimensionality reduction, cross-validation, and classifier performance estimation based on receiver operating characteristic (ROC) curves.

Abstract: Lack of insight into illness is a prevalent and distinguishing feature of schizophrenia, which has a complex history and has been given a variety of definitions. Currently, insight is measured and treated as a multidimensional phenomenon, because it is believed to result from psychological, neuropsychological and organic factors. Thus, schizophrenia patients may display dramatic disorders including demoralization, depression and a higher risk of suicide, all of which are directly or indirectly related to a lack of insight into their illness, and make the treatment difficult. To improve the treatment of people with schizophrenia, it is thus crucial to advance research on insight into their illness. Insight is studied in a variety of ways. Studies may focus on the relationship between insight and psychopathology, may view behavioral outcomes or look discretely at the cognitive dysfunction versus anatomy level of insight. All have merit but they are dispersed across a wide body of literature and rarely are the findings integrated and synthesized in a meaningful way. The aim of this study was to synthesize findings across the large body of literature dealing with insight, to highlight its multidimensional nature, measurement, neuropsychology and social impact in schizophrenia. The extensive literature on the cognitive consequences of lack of insight and the contribution of neuroimaging techniques to elucidating neurological etiology of insight deficits, is also reviewed.

Abstract: Much theoretical attention is currently devoted to social learning. Yet, empirical studies formally comparing its effectiveness relative to individual learning are rare. Here, we focus on free choice, which is at the heart of individual reward-based learning, but absent in social learning. Choosing among two equally valued options is known to create a preference for the selected option in both humans and monkeys. We thus surmised that social learning should be more helpful when choice-induced preferences retard individual learning than when they optimize it. To test this prediction, the same task requiring to find which among two items concealed a reward was applied to rhesus macaques and humans. The initial trial was individual or social, rewarded or unrewarded. Learning was assessed on the second trial. Choice-induced preference strongly affected individual learning. Monkeys and humans performed much more poorly after an initial negative choice than after an initial positive choice. Comparison with social learning verified our prediction. For negative outcome, social learning surpassed or at least equaled individual learning in all subjects. For positive outcome, the predicted superiority of individual learning did occur in a majority of subjects (5/6 monkeys and 6/12 humans). A minority kept learning better socially though, perhaps due to a more dominant/aggressive attitude toward peers. Poor learning from errors due to over-valuation of personal choices is among the decision-making biases shared by humans and animals. The present study suggests that choice-immune social learning may help curbing this potentially harmful tendency. Learning from successes is an easier path. The present data suggest that whether one tends to walk it alone or with a peer's help might depend on the social dynamics within the actor/observer dyad.

Abstract: The present behavioral study re-addresses the question of habit learning in Parkinson's disease (PD). Patients were early onset, non-demented, dopa-responsive, candidates for surgical treatment, similar to those we found earlier as suffering greater dopamine depletion in the putamen than in the caudate nucleus. The task was the same conditional associative learning task as that used previously in monkeys and healthy humans to unveil the striatum involvement in habit learning. Sixteen patients and 20 age- and education-matched healthy control subjects learned sets of 3 visuo-motor associations between complex patterns and joystick displacements during two testing sessions separated by a few hours. We distinguished errors preceding vs. following the first correct response to compare patients' performance during the earliest phase of learning dominated by goal-directed actions with that observed later on, when responses start to become habitual. The disease significantly retarded both learning phases, especially in patients under 60 years of age. However, only the late phase deficit was disease severity-dependent and persisted on the second testing session. These findings provide the first corroboration in Parkinson patients of two ideas well-established in the animal literature. The first is the idea that associating visual stimuli to motor acts is a form of habit learning that engages the striatum. It is confirmed here by the global impairment in visuo-motor learning induced by PD. The second idea is that goal-directed behaviors are predominantly caudate-dependent whereas habitual responses are primarily putamen-dependent. At the advanced PD stages tested here, dopamine depletion is greater in the putamen than in the caudate nucleus. Accordingly, the late phase of learning corresponding to the emergence of habitual responses was more vulnerable to the disease than the early phase dominated by goal-directed actions.

Abstract: Behavioral studies have shown that hand position influences saccade characteristics. This study examined the neuronal changes that could underlie this behavioral observation. Single neurons were recorded in the frontal eye field (FEF) of 2 monkeys as they executed a visually guided saccade task, while holding their hand at given locations on a touch screen. The task was performed with the hand either visible or invisible, in order to assess the relative contribution of visual and proprioceptive information on hand position. Among the 224 neurons tested, the visual, saccadic and/or preparatory activity of more than half of them was modulated by hand position, whether the hand was visible or invisible. Comparison of lower (hand's workspace) and upper (out of reach) visual targets showed that hand modulation was predominant in the hand's workspace. Finally, some cells preferred congruency of hand and target in space, others preferred incongruency. Interestingly, hand modulation of saccadic activity correlated with hand position effects on saccade reaction times. We conclude that visual and proprioceptive signals derived from the hand are integrated by FEF neurons. These signals can modulate target selection through attention and allow the oculomotor system to use hand-related somatosensory signals for the initiation of visually guided saccades.

Abstract: The dorsal striatum is crucial for the acquisition and consolidation of instrumental behaviour, but the underlying computations and internal dynamics remain elusive. To address this issue, we combined a model of key computations supporting decision-making during instrumental learning with human behavioural and functional magnetic resonance imaging (fMRI) data. The results showed that the associative and sensorimotor dorsal striatum host complementary computations that, we suggest, may differentially support goal-directed and habitual processes. The anterior caudate nucleus integrates information about performance and cognitive control demands, whereas the putamen tracks how likely the conditioning stimuli lead to correct response. Contrary to current models, the putamen is recruited during initial acquisition. As the exploratory phase proceeds, the relative contribution of the caudate nucleus becomes dominant over the putamen. During early consolidation, caudate nucleus and putamen settle to asymptotic values and share control. We then investigated how dorsal striatal computations may affect decision-making. We found that portion of reaction times' variance parallels the combined cost associated with the dorsal striatal computations. Overall, our findings provide a deeper insight into the functional heterogeneity within the dorsal striatum and suggest that the dynamic interplay between caudate nucleus and putamen, rather than their serial recruitment, underlies the acquisition and early consolidation of instrumental behaviours.

Abstract: Associative theory postulates that learning the consequences of our actions in a given context is represented in the brain as stimulus-response-outcome associations that evolve according to prediction-error signals (the discrepancy between the observed and predicted outcome). We tested the theory on brain functional magnetic resonance imaging data acquired from human participants learning arbitrary visuomotor associations. We developed a novel task that systematically manipulated learning and induced highly reproducible performances. This granted the validation of the model-based results and an in-depth analysis of the brain signals in representative single trials. Consistent with the Rescorla-Wagner model, prediction-error signals are computed in the human brain and selectively engage the ventral striatum. In addition, we found evidence of computations not formally predicted by the Rescorla-Wagner model. The dorsal fronto-parietal network, the dorsal striatum, and the ventrolateral prefrontal cortex are activated both on the incorrect and first correct trials and may reflect the processing of relevant visuomotor mappings during the early phases of learning. The left dorsolateral prefrontal cortex is selectively activated on the first correct outcome. The results provide quantitative evidence of the neural computations mediating arbitrary visuomotor learning and suggest new directions for future computational models.

Abstract: In daily life, activities requiring the hand and eye to work separately are as frequent as activities requiring tight eye-hand coordination, and we effortlessly switch from one type of activity to the other. Such flexibility is unlikely to be achieved without each effector "knowing" where the other one is at all times, even when it is static. Here, we provide behavioral evidence that the mere position of the static hand affects one eye movement parameter: saccadic reaction time. Two monkeys were trained and 11 humans instructed to perform nondelayed or delayed visually guided saccades to either a right or a left target while holding their hand at a location either near or far from the eye target. From trial to trial, target locations and hand positions varied pseudorandomly. Subjects were tested both when they could and when they could not see their hand. The main findings are 1) the presence of the static hand in the workspace did affect saccade initiation; 2) this interaction persisted when the hand was invisible; 3) it was strongly influenced by the delay duration: hand-target proximity retarded immediate saccades, whereas it could hasten delayed saccades; and 4) this held true both for humans and for each of the two monkeys. We propose that both visual and nonvisual hand position signals are used by the primates' oculomotor system for the planning and execution of saccades, and that this may result in a hand-eye competition for spatial attentional resources that explains the delay-dependent reversal observed.

Abstract: Observational learning allows individuals to acquire knowledge without incurring in the costs and risks of discovering and testing. The neural mechanisms mediating the retrieval of rules learned by observation are currently unknown. To explore this fundamental cognitive ability, we compared the brain responses when retrieving visuomotor associations learned either by observation or by individual learning. To do so, we asked eleven adults to learn two sets of arbitrary visuomotor associations: one set was learned through the observation of an expert actor while the other was learned by trial and error. During fMRI scanning, subjects were requested to retrieve the visuomotor associations previously learned under the two modalities. The conjunction analysis between the two learning conditions revealed a common brain network that included the ventral and dorsal lateral prefrontal cortices, the superior parietal lobe and the pre-SMA. This suggests the existence of a mirror-like system responsible for the storage of rules learned either by trial and error or by observation of others' actions. In addition, the pars triangularis in the right prefrontal cortex (BA45) was found to be selective for rules learned by observation. This suggests a preferential role of this area in the storage of rules learned in a social context.

Abstract: Successful adaptation relies on the ability to learn the consequence of our actions in different environments. However, understanding the neural bases of this ability still represents one of the great challenges of system neuroscience. In fact, the neuronal plasticity changes occurring during learning cannot be fully controlled experimentally and their evolution is hidden. Our approach is to provide hypotheses about the structure and dynamics of the hidden plasticity changes using behavioral learning theory. In fact, behavioral models of animal learning provide testable predictions about the hidden learning representations by formalizing their relation with the observables of the experiment (stimuli, actions and outcomes). Thus, we can understand whether and how the predicted learning processes are represented at the neural level by estimating their evolution and correlating them with neural data. Here, we present a bayesian model approach to estimate the evolution of the internal learning representations from the observations of the experiment (state estimation), and to identify the set of models' parameters (parameter estimation) and the class of behavioral model (model selection) that are most likely to have generated a given sequence of actions and outcomes. More precisely, we use Sequential Monte Carlo methods for state estimation and the maximum likelihood principle (MLP) for model selection and parameter estimation. We show that the method recovers simulated trajectories of learning sessions on a single-trial basis and provides predictions about the activity of different categories of neurons that should participate in the learning process. By correlating the estimated evolutions of the learning variables, we will be able to test the validity of different models of instrumental learning and possibly identify the neural bases of learning.