Es considera que la cognició biològica fa servir mecanismes com la predicció, l'anticipació i l'atenció per resoldre tasques complexes. S'ha suggerit que aquests mecanismes es materialitzen en els mamífers a través d'interaccions entre les capes corticals, mentre que la seva manifestació en cervells relativament més simples, como el dels invertebrats, és encara poc clara. En la cognició artificial, la naturalesa i la interacció dels mecanismes mencionats roman, en gran mesura, no quantificada. Aquí proposem un enfoc filogènic i basat en models per descobrir com interactuen aquests mecanismes cognitius. Comencem amb el model simple del cervell d'un insecte i demostrem la necessitat dels anomenats forward models per explicar el comportament d'un insecte a escenaris dinàmics. Llavors proposem el marc PASAR per integrar i quantificar la interacció dels mencionats components de la cognició. Validem el PASAR en tasques robòtiques i en un experiment psicofísic humà, demostrant que el PASAR és una eina valuosa per modelar i avaluar la cognició biològica i per construir sistemes cognitius artificials. / Biological cognition is thought to employ mechanisms like prediction, anticipation and attention for solving complex tasks. These mechanisms are suggested to be materialized through inter-layer cortical interactions in mammals, whereas their manifestation in relatively simpler brains, like the invertebrate brain, remains unclear. In artificial cognition, the nature and interplay of the above mechanisms remain largely unquantified. Here we propose a phylogenic, model-based approach to answer how these cognitive mechanisms interplay. We start with a simple model of the insect brain and demonstrate the necessity of the so-called forward models to account for insect behavior in dynamic scenarios. We then propose the PASAR framework to integrate and quantify the interplay of the above components of cognition. We validate PASAR in robotic tasks and in a human psychophysical experiment, proving PASAR as a valuable tool to model and evaluate biological cognition and to construct artificial cognitive systems.

Dissertation (Ph.D. in Software Engineering)--Naval Postgraduate School, March 2007.D / Dissertation supervisor: Berzins, Valdis. "March 2007." Description based on title screen as viewed on April 14, 2010. DTIC Descriptors: Software Engineering, Cognition, Learning Machines, Artificial Intelligence, Computer Programming, Simulation, Military Training, Vision, User Needs, Patterns, Employment, Command And Control Systems, Humans. DTIC Identifier(s): Software Engineering, Cognitive Systems, Machine Learning, Simulation, System Deployment, Artificial Intelligence, System Evaluation, Software Evaluation. Author(s) subject terms: Software Engineering, Cognitive Systems, Machine Learning, Simulation, System Deployment, Artificial Intelligence, System Evaluation, Software Evaluation. Includes bibliographical references (p. 395-399). Also available in print.

The Vessel Traffic Service (VTS) is a complex system tasked with ensuring the safety of navigation within specified areas known as VTS areas. Earlier research in the domain has often focused on the decision support systems and other tools employed by the VTS operators to provide the vessels in the area with VTS services. Consequently, less effort has gone into looking at the system itself and the human factors aspects of the system. This study uses the Functional Resonance Analysis Method (FRAM) to create a functional model of the VTS. It looks at how a VTS works, what the different components are and how these components are related. The main purpose of the FRAM model is to serve as a basis for future application by identifying the functions that constitute the system, and to illuminate the potential variability therein. To demonstrate how it might be used, an instantiation of an observed scenario will be presented. A structural description of the VTS is also presented, which aims to serve as an introduction to the domain for readers who are previously unfamiliar with it. The functional model shows that a lot of the potential variability seems to lie in the functions that rely heavily on human interaction, which is to be expected, as human performance is highly variable. It also shows that the availability and reliability of relevant information is crucial in order to be able to provide the VTS services, and if the information for some reason is unavailable or insufficient it seems likely to cause variability.

Neuro-computational models allow to study the brain mechanisms involved in intelligent behavior and extract essential computational principles which can be implemented in cognitive systems. They are a promising solution to achieve a brain-like artificial intelligence that can compete with natural intelligence on realistic behaviors. A crucial property of intelligent behavior is motivation, defined as the incentive to interact with the world in order to achieve specific goals, either extrinsic (obtaining rewards such as food or money, or avoiding pain) or intrinsic (satisfying one’s curiosity, fun). In the human brain, motivated or goal-directed behavior depends on a network of different structures, including the prefrontal cortex, the basal ganglia and the limbic system. Dopamine, a neurotransmitter associated with reward processing, plays a central role in coordinating the activity of this network. It structures processing in high-level cognitive areas along a limbic-associative-motor gradient and impacts the learning capabilities of the whole system. In this habilitation thesis, I present biologically-constrained neuro-computational models which investigate the role of dopamine in visual object categorization and memory retrieval (Vitay and Hamker, 2008), reinforcement learning and action selection (Vitay and Hamker, 2010), the updating, learning and maintenance of working memory (Schroll et al., 2012) and timing processes (Vitay and Hamker, 2014). These models outline the many mechanisms by which the dopaminergic system regulates cognitive and emotional behavior: bistable processing modes in the cerebral cortex, modulation of synaptic transmission and plasticity, allocation of cognitive resources and signaling of relevant events. Finally, I present a neural simulator able to simulate a variety of neuro-computational models efficiently on parallel architectures (Vitay et al., 2015). / Neuronale Modelle nach dem Vorbild des Gehirns bieten die Möglichkeit intelligente, kognitive Prozesse nicht nur besser zu verstehen, sondern sie stellen auch eine vielversprechende Lösung dar, um eine Gehirn-ähnliche künstliche Intelligenz für Wahrnehmung und Verhaltensweisen zu erreichen, die mit natürlicher Intelligenz konkurrieren kann. Eine entscheidende Eigenschaft von intelligentem Verhalten ist Motivation, definiert als der Anreiz mit der Welt zu interagieren, um bestimmte Ziele zu erreichen, sei es extrinsisch (Belohnungen wie Nahrung oder Geld zu erhalten oder die Vermeidung von Schmerzen) oder intrinsisch (die Neugier zu befriedigen, Spaß zu haben). Im menschlichen Gehirn basiert motiviertes oder zielgerichtetes Verhalten auf einem Netzwerk von verschiedenen Strukturen, einschließlich des präfrontalen Cortex, der Basalganglien und des limbischen Systems. Dopamin, ein Neurotransmitter, welcher der Belohnungsverarbeitung zugeordnet wird, spielt eine zentrale Rolle bei der Koordination der Aktivität in diesem Netzwerk. Es strukturiert die Verarbeitung in High-Level-kognitiven Bereichen entlang eines limbischen-assoziativ-motor Gradienten und beinflusst die Lernfähigkeit des gesamten Systems. In dieser Habilitation, präsentiere ich biologisch motivierte neuronale Modelle, die die Rolle von Dopamin in der visuellen Objektkategorisierung und Gedächtnisabruf (Vitay and Hamker, 2008), Reinforcement Lernen und Aktionsauswahl (Vitay and Hamker, 2010), Aktualisierung, Lernen und Aufrechterhaltung von Arbeitsgedächtnis (Schroll et al., 2012) und Timing Prozessen (Vitay and Hamker, 2014) untersuchen. Diese Modelle beschreiben Mechanismen, durch die das dopaminerge System kognitives und emotionales Verhalten reguliert: bistabile Verarbeitungsmodi in der Hirnrinde, Plastizität und Modulation der synaptischen Übertragung, Zuweisung von kognitiven Ressourcen und Signalisierung von relevanten Ereignissen. Schließlich beschreibe ich einen neuronalen Simulator, der in in der Lage ist, eine Vielzahl von neuronalen Modellen effizient auf parallelen Architekturen zu simulieren (Vitay et al., 2015).

No description available.

No description available.

No description available.

No description available.

No description available.

Dans ce manuscrit nous proposons une architecture neuronale d'inspiration corticale, capable de développer un traitement émergent de type auto-organisation. Afin d'implémenter cette architecture neuronale de manière distribuée, nous utilisons le modèle de champs neuronaux dynamiques, un formalisme mathématique générique conçu pour modéliser la compétition des activités neuronales au niveau cortical mésoscopique. Pour analyser en détail les propriétés dynamiques des modèles de référence de ce formalisme, nous proposons un critère formel et un instrument d'évaluation, capable d'examiner et de quantifier le comportement dynamique d'un champ neuronal quelconque dans différents contextes de stimulation. Si cet instrument nous permet de mettre en évidence les avantages pratiques de ces modèles, il nous révèle aussi l'incapacité de ces modèles à conduire l'implantation des processus d'auto-organisation (implémenté par l'architecture décrite) vers des résultats satisfaisants. Ces résultats nous amènent à proposer une alternative aux modèles classiques de champs, basée sur un mécanisme de rétro-inhibition, qui implémente un processus local de régulation neuronale. Grâce à ce mécanisme, le nouveau modèle de champ réussit à implémenter avec succès le processus d'auto-organisation décrit par l'architecture proposée d'inspiration corticale. De plus, une analyse détaillée confirme que ce formalisme garde les caractéristiques dynamiques exhibées par les modèles classiques de champs neuronaux. Ces résultats ouvrent la perspective de développement des architectures de calcul neuronal de traitement d'information pour la conception des solutions logicielles ou robotiques bio-inspirées / In this work we propose a cortically inspired neural architecture capable of developping an emergent process of self-organization. In order to implement this neural architecture in a distributed manner, we use the dynamic neural fields paradigm, a generic mathematical formalism aimed at modeling the competition between the neural activities at a mesoscopic level of the cortical structure. In order to examine in detail the dynamic properties of classical models, we design a formal criterion and an evaluation instrument, capable of analysing and quantifying the dynamic behavior of the any neural field, in specific contexts of stimulation. While this instrument highlights the practical advantages of the usage of such models, it also reveals the inability of these models to help implementing the self-organization process (implemented by the described architecture) with satisfactory results. These results lead us to suggest an alternative to the classical neural field models, based on a back-inhibition model which implements a local process of neural activity regulation. Thanks to this mechanism, the new neural field model is capable of achieving successful results in the implementation of the self-organization process described by our cortically inspired neural architecture. Moreover, a detailed analysis confirms that this new neural field maintains the features of the classical field models. The results described in this thesis open the perspectives for developping neuro-computational architectures for the design of software solutions or biologically-inspired robot applications