Current neurobiological models of adolescent decision-making suggest that heightened risk taking during adolescence is a result of the asynchronous development of neural regions underlying cognitive control and reward processing, particularly during periods of heightened social and affective arousal (e.g., Casey, Getz, & Galván, 2008; Steinberg, 2008). Despite the emphasis on the interplay of cognitive and emotional processes during adolescence, the developmental literature has largely overlooked the potential importance of maturational changes in the anterior insular cortex (AIC), a region known for its role as a cognitive-emotional hub. In a recent review we proposed a theory of adolescent risk-taking in which development of the AIC, and its connectivity to other regions, biases adolescents towards engagement in risky behaviors (Smith, Steinberg, & Chein, 2014b). The current studies provide a test of the proposed model through an examination of specific aspects of AIC development and functioning, including the trajectory of structural development within the AIC, the role of AIC engagement in adolescents' risky decision-making, and the impacts of affective arousal on AIC recruitment. Results from Study 1 suggest that the AIC exhibits continued developmental changes during adolescence that likely affect its involvement in cognitive processes. Using a risk-taking task, Study 2 demonstrates the flexible role of the AIC during adolescent decision-making and explores how affective arousal biases the AIC towards engagement in risky behaviors. Implications for both the proposed model and the developmental literature are discussed. / Psychology

Orientadora: Profª Drª Raquel Vecchio Fornari / Dissertação (mestrado) - Universidade Federal do ABC, Programa de Pós-Graduação em Neurociência e Cognição, 2015. / O cortex insular (CI), ou insula, conquistou nos ultimos anos um lugar de destaque na area cientifica por seu suposto envolvimento em processos emocionais e cognitivos. No rato, como nos seres humanos, o CI pode ser dividido em duas sub-regioes funcionalmente heterogeneas. Embora a maioria dos estudos realizados em animais tenha mostrado um envolvimento da regiao mais rostral (CI anterior) na memoria gustativa, outros sugerem um papel mais amplo, abrangendo desde o reconhecimento de objetos ate o processamento de memorias espaciais e aversivas. A regiao mais caudal (CI posterior), por sua vez, recebe aferencias multissensoriais e supoe-se que esteja envolvida em processamento multissensorial e nociceptivo. Entretanto, pouquissimos trabalhos avaliaram a participacao dessa sub-regiao posterior em tarefas de memoria, com resultados inconclusivos. Nao havia, ate o momento, nenhum trabalho que tivesse investigado isoladamente o papel de ambas as sub-regioes do CI na consolidacao da memoria emocional. Portanto, o objetivo deste estudo foi investigar os efeitos da inativacao temporaria do CI anterior e posterior sobre a consolidacao da memoria de medo de tarefas de condicionamento de medo ao contexto e ao som. Ratos Wistar de 3 meses de idade passaram por cirurgia estereotaxica para implante de canulas-guia bilaterais no CI anterior ou posterior. Os animais tiveram pelo menos 7 dias de recuperacao e foram manipulados por 3 dias antes do inicio do procedimento comportamental. Para o treino de condicionamento de medo ao contexto e ao som, os ratos foram colocados individualmente em um caixa de condicionamento. Apos 120 segundos de livre exploracao, um som (90 decibeis, 2 kHz) foi emitido por 30 segundos, co-terminando com um choque nas patas (0,7 mA, 1s). Imediatamente apos, cada rato recebeu uma microinfusao bilateral de muscimol (agonista gabaergico, 0,5¿Êg/0,5¿ÊL por hemisferio) ou salina (grupo controle). O teste de condicionamento de medo ao contexto (CMC) ocorreu 48 horas apos o treino e consistiu na re-exposicao a caixa de condicionamento por 5 minutos, sem apresentacao de som ou choque. 24 horas depois, os mesmos animais foram submetidos ao teste de condicionamento de medo ao som (CMS), o qual ocorreu em uma caixa modificada, com duracao de 5 minutos. Ao final do segundo e terceiro minutos, o mesmo estimulo sonoro apresentado no treino foi emitido por 30 segundos. O tempo de congelamento e o comportamento motor foram utilizados como medidas de condicionamento. No CMS, os ratos que receberam a microinfusao de muscimol no CI anterior e posterior apresentaram uma media de tempo de congelamento menor durante o periodo pos-som. Entretanto, no CMC nao houve diferencas entre grupos para ambas as subregioes do CI. Portanto, os resultados deste estudo indicam que a inativacao pos-treino do CI como um todo prejudica exclusivamente o CMS. Entretanto, o prejuizo deste tipo de memoria, provocado pela inativacao do CI posterior, foi maior, evidenciando, portanto, que esta subregiao esta mais importantemente envolvida na circuitaria neural responsavel pela consolidacao do medo condicionado a um estimulo sonoro discreto. / The insular cortex (IC), or insula, has achieved over the last years an eminent position in the scientific literature due to its involvement in emotional and cognitive processes. In the rat, as in humans, the IC can be divided into two functionally heterogeneous sub-regions. Although most animal studies have shown an involvement of the rostral subregion (anterior IC) in gustatory memory, others suggest a broader role in memory, ranging from object recognition to the processing of spatial and aversive memories. In addition, even though the most caudal area (posterior IC) seems to be involved in multisensory and nociceptive processing, very few studies have evaluated its role in mnemonic processes and the results so far are unclear. Nevertheless, no work, to the best of our knowledge, had investigated the specific role of both sub-regions of the IC on consolidation of fear conditioning tasks. Thus, the aim of the present study was to investigate the effects of temporary inactivation of the anterior and posterior IC on memory consolidation of contextual and tone fear conditioning tasks. 3-month-old male Wistar rats underwent stereotaxic surgery for implantation of bilateral guide cannulae aimed directly above the anterior or posterior IC. The animals were allowed at least 7 days of recovery and were handled once a day for 3 days prior to behavioral sessions. For the contextual and tone fear conditioning training session, the rats were individually placed in the conditioning box. After 120 seconds of free exploration, a tone (90 decibels, 2 kHz) was delivered for 30 seconds, coterminating with a footshock (0.7 mA, 1 s). Immediately after, each rat received a bilateral microinjection of muscimol (GABAergic agonist, 0.5 ìg/0.5ìL by hemisphere) or saline (control group) into the intended IC subregion. The contextual fear conditioning test (CFC) was performed 48 hours after training and consisted in the re-exposure to the conditioning box for 5 minutes, without delivery of tone and shock. After 24 hours, the same animals were submitted to tone fear conditioning test (TFC), which occurred in a modified chamber, for 5 minutes. At the end of the second and third minutes, the same tone stimulus presented in the training session was delivered for 30 seconds. Freezing time and motion behavior were used as measures of conditioning. In TFC, the rats that had received muscimol microinfusion into the anterior and posterior IC displayed a lower freezing time during the post-tone period. However, for both IC subregions, there were no differences between groups in the CFC. Thus, our findings indicate that the posttraining inactivation of both IC subregions impaired the TFC. However, the impairment in this kind of memory, caused by the the inactivation of the posterior IC, was higher, thus, highlighting that this subregion is more importantly involved in the neural circuitry related to the consolidation of the discrete tone conditioned fear.

Un pilier d'une vie sociale fructueuse est la capacité de prédire correctement les actions des autres et de percevoir leurs états émotionnels. Des études d'interaction sociale chez les primates ont montré qu'ils sont capables de déduire ce que les autres peuvent entendre ou voir, et de prédire leurs émotions et intentions. Il a été montré qu'ils peuvent manifester différents degrés de comportements prosociaux, allant de la coopération jusqu'à des comportements altruistes et empathiques. Des études d'imageries fonctionnelles chez l'homme ont identifié l'insula antérieur (AI) comme une région cérébrale clé dans le traitement de l'empathie.Spécifiquement, cette région apparait comme l'aire intégratrice des activités liées à la douleur ressentie et observée, suggérant que l'empathie pourrait impliquer un modèle « miroir » des propriétés affectives et sensorielles de la douleur d'autrui. Cependant, les bases neuronales de ce processus n'ont pas encore été découvertes. Dans le but d'examiner le rôle de l'AI dans le traitement de l'empathie, nous avons enregistré l'activité des neurones dans l'AI de deux singes pendant qu'ils sont engagés dans une tâche sociale leur permettant de délivrer un stimulus aversif ou appétitif à leur partenaire, à lui-même ou à personne. Les résultats comportementaux ont montré que les singes prennent en compte le bien-être de leur partenaire. Les données neuronales rapportent différentes populations neuronales répondant aux stimuli aversif ou appétitif et ceux délivrés à soi ou à autrui. Notamment, la population neuronale répondant au stimulus aversif a montré trois profils d'activité : une représentation neuronale de l'expérience désagréable du partenaire, une représentation neuronale de sa propre sensation désagréable et une minorité de neurones montrant des propriétés miroirs entre soi et autrui. Nos résultats suggèrent un modèle neuronal de l'empathie représentant des propriétés distinctes entre l'expérience vécue et observée / A cornerstone of a successful social life is the ability to correctly predict others’ actions and empathically perceive their emotional states. Studies on primates’ social interaction have shown that thanks to their keen cognitive abilities monkeys are able to deduce what others can hear or see, and to predict others’ emotions and intentions. It has been shown that primates are able to display different degrees of prosocial behavior, from cooperation to even altruism and empathically driven behavior. Studies using fMRI techniques inhumans have identified the anterior insula (AI) as a key brain region in the processing of empathy. More precisely, this region emerged as the overlapping area activated for both experienced and observed pain,leading to the idea that empathy for pain may involve a mirror-matching model of the affective and sensory features of others' pain. However, the neuronal basis of this process has yet to be uncovered. In an attempt toextend and to investigate the role of the AI in the process of empathy we have recorded single cell activity inthe AI of two monkeys while they were engaged in a social task where based on the performed trials positiveor negative reinforcements could be delivered to self, another monkey, or nobody. Behavioral results showed that monkeys take into account the welfare of their partners even when this has no impact on their ownwelfare. Our neuronal findings report that distinct population of neurons respond differentially to outcomesfor self and other, and to appetitive and aversive outcomes. Interestingly the neuronal population responding to the aversive outcome showed mainly three profiles of activity: neuronal representation of conspecifics’unpleasant experience, neuronal representation of own unpleasant experience and a minority of neurons showing mirroring properties between self and other. Thus, our results suggest a neuronal model of empathy that accounts for the distinctive features between feeling and empathizing

Chaque jour, nous prenons des décisions impliquant de choisir les options qui nous semblent les plus avantageuses, en nous basant sur nos expériences passées. Toutefois, les mécanismes et les bases neurales de l’apprentissage par renforcement restent débattus. D’une part, certains travaux suggèrent l’existence de deux systèmes opposés impliquant des aires cérébrales corticales et sous-corticales distinctes lorsque l’on apprend par la carotte ou par le bâton. D’autres part, des études ont montré une ségrégation au sein même de ces régions cérébrales ou entre des neurones traitant l’apprentissage par récompenses et celui par évitement des punitions. Le but de cette thèse était d’étudier la dynamique cérébrale de l’apprentissage par renforcement chez l’homme. Pour ce faire, nous avons utilisé des enregistrements intracérébraux réalisés chez des patients épileptiques pharmaco-résistants pendant qu’ils réalisaient une tâche d’apprentissage probabiliste. Dans les deux premières études, nous avons d’investigué la dynamique de l’encodage des signaux de renforcement, et en particulier à celui des erreurs de prédiction des récompenses et des punitions. L’enregistrement de potentiels de champs locaux dans le cortex a mis en évidence le rôle central de l’activité à haute-fréquence gamma (50-150Hz). Les résultats suggèrent que le cortex préfrontal ventro-médian est impliqué dans l’encodage des erreurs de prédiction des récompenses alors que pour l’insula antérieure, le cortex préfrontal dorsolatéral sont impliqués dans l’encodage des erreurs de prédiction des punitions. De plus, l’activité neurale de l’insula antérieure permet de prédire la performance des patients lors de l’apprentissage. Ces résultats sont cohérents avec l’existence d’une dissociation au niveau cortical pour le traitement des renforcements appétitifs et aversifs lors de la prise de décision. La seconde étude a permis d’étudier l’implication de deux noyaux limbiques du thalamus au cours du même protocole cognitif. L’enregistrement de potentiels de champs locaux a mis en évidence le rôle des activités basse fréquence thêta dans la détection des renforcements, en particulier dans leur dimension aversive. Dans une troisième étude, nous avons testé l’influence du risque sur l’apprentissage par renforcement. Nous rapportons une aversion spécifique au risque lors de l’apprentissage par évitement des punitions ainsi qu’une diminution du temps de réaction lors de choix risqués permettant l’obtention de récompenses. Cela laisse supposer un comportement global tendant vers une aversion au risque lors de l’apprentissage par évitement des punitions et au contraire une attirance pour le risque lors de l’apprentissage par récompenses, suggérant que les mécanismes d’encodage du risque et de la valence pourraient être indépendants. L’amélioration de la compréhension des mécanismes cérébraux sous-tendant la prise de décision est importante, à la fois pour mieux comprendre les déficits motivationnels caractérisant plusieurs pathologies neuropsychiatriques, mais aussi pour mieux comprendre les biais décisionnels que nous pouvons exhiber. / We make decisions every waking day of our life. Facing our options, we tend to pick the most likely to get our expected outcome. Taking into account our past experiences and their outcome is mandatory to identify the best option. This cognitive process is called reinforcement learning. To date, the underlying neural mechanisms are debated. Despite a consensus on the role of dopaminergic neurons in reward processing, several hypotheses on the neural bases of reinforcement learning coexist: either two distinct opposite systems covering cortical and subcortical areas, or a segregation of neurons within brain regions to process reward-based and punishment-avoidance learning.This PhD work aimed to identify the brain dynamics of human reinforcement learning. To unravel the neural mechanisms involved, we used intracerebral recordings in refractory epileptic patients during a probabilistic learning task. In the first study, we used a computational model to tackle the brain dynamics of reinforcement signal encoding, especially the encoding of reward and punishment prediction errors. Local field potentials exhibited the central role of high frequency gamma activity (50-150Hz) in these encodings. We report a role of the ventromedial prefrontal cortex in reward prediction error encoding while the anterior insula and the dorsolateral prefrontal cortex encoded punishment prediction errors. In addition, the magnitude of the neural response in the insula predicted behavioral learning and trial-to-trial behavioral adaptations. These results are consistent with the existence of two distinct opposite cortical systems processing reward and punishments during reinforcement learning. In a second study, we recorded the neural activity of the anterior and dorsomedial nuclei of the thalamus during the same cognitive task. Local field potentials recordings highlighted the role of low frequency theta activity in punishment processing, supporting an implication of these nuclei during punishment-avoidance learning. In a third behavioral study, we investigated the influence of risk on reinforcement learning. We observed a risk-aversion during punishment-avoidance, affecting the performance, as well as a risk-seeking behavior during reward-seeking, revealed by an increased reaction time towards appetitive risky choices. Taken together, these results suggest we are risk-seeking when we have something to gain and risk-averse when we have something to lose, in contrast to the prediction of the prospect theory.Improving our common knowledge of the brain dynamics of human reinforcement learning could improve the understanding of cognitive deficits of neurological patients, but also the decision bias all human beings can exhibit.

No description available.