Catálogo de publicaciones - libros

Research on anticipatory behavior in adaptive learning systems continues to gain more recognition and appreciation in various research disciplines. This book provides an overarching view on anticipatory mechanisms in cognition, learning, and behavior. It connects the knowledge from cognitive psychology, neuroscience, and linguistics with that of artificial intelligence, machine learning, cognitive robotics, and others. This introduction offers an overview over the contributions in this volume highlighting their interconnections and interrelations from an anticipatory behavior perspective. We first clarify the main foci of anticipatory behavior research. Next, we present a taxonomy of how anticipatory mechanisms may be beneficially applied in cognitive systems. With relation to the taxonomy, we then give an overview over the book contributions. The first chapters provide surveys on currently known anticipatory brain mechanisms, anticipatory mechanisms in increasingly complex natural languages, and an intriguing challenge for artificial cognitive systems. Next, conceptualizations of anticipatory processes inspired by cognitive mechanisms are provided. The conceptualizations lead to individual, predictive challenges in vision and processing of event correlations over time. Next, anticipatory mechanisms in individual decision making and behavioral execution are studied. Finally, the book offers systems and conceptualizations of anticipatory processes related to social interaction.

Animals anticipate the future in a variety of ways. For instance: (a) they make motor actions that are timed to a reference stimulus and motor actions that anticipate future movement dynamics; (b) they learn to make choices that will maximize reward they receive in the future; and (c) they form memories of behavioral episodes such that the animal’s future actions can be predicted by current neural activity associated with those memories. Although these effects are clearly observable at the behavioral level, research into the mechanisms of such anticipatory learning are still largely in the early stages. This review, intended for those who have a computational background and are less familiar with neuroscience, addresses neural mechanisms found in the mammalian cerebellum, basal ganglia, and the hippocampus that give rise to such adaptive anticipatory behavior.

We review some of the main theories about how language emerged. We suggest that including the study of the emergence of artificial languages, in simulation settings, allows us to ask a more general question, namely, what are the minimal initial conditions for the emergence of language ? This is a very important question from a technological viewpoint, because it is very closely tied to questions of intelligence and autonomy. We identify anticipation as being a key underlying computational principle in the emergence of language. We suggest that this is in fact present implicitly in many of the theories in contention today. Focused simulations that address precise questions are necessary to isolate the roles of the minimal initial conditions for the emergence of language.

It seems characteristic for humans to detect structural patterns in the world to anticipate future states. Therefore, scientific and common sense cognition could be described as information processing which infers rule-like laws from patterns in data-sets. Since information processing is the domain of computers, artificial cognitive systems are generally designed as pattern discoverers. This paper questions the validity of the information processing paradigm as an explanation for human cognition and a design principle for artificial cognitive systems. Firstly, it is known from the literature that people suffer from conditions such as information overload, superstition, and mental disorders. Secondly, cognitive limitations such as a small short-term memory, the set-effect, the illusion of explanatory depth, etc. raise doubts as to whether human information processing is able to cope with the enormous complexity of an infinitely rich (amorphous) world. It is suggested that, under normal conditions, humans construct information rather than process it. The constructed information contains anticipations which need to be met. This can be hardly called information processing, since patterns from the “outside” are not used to produce action but rather to either justify anticipations or restructure the cognitive apparatus. When it fails, cognition switches to pattern processing, which, given the amorphous nature of the experiential world, is a lost cause if these patterns and inferred rules do not lead to a (partial) reorganisation of internal structures such that constructed anticipations can be met again. In this scenario, superstition and mental disorders are the result of a profound and/or random restructuring of already existing cognitive components (e.g., action sequences). This means that whenever a genuinely cognitive system is exposed to pattern processing it may start to behave superstitiously. The closer we get to autonomous self-motivated artificial cognitive systems, the bigger the danger becomes of superstitious information processing machines that “blow up” rather than behave usefully and effectively. Therefore, to avoid superstition in cognitive systems they should be designed as information constructing entities.

How can goals be represented in natural and artificial systems? How can they be learned? How can they trigger actions? This paper describes, analyses and compares two of the most influential models of goal-oriented behavior: the ideomotor principle (IMP), which was introduced in the psychological literature, and the “test, operate, test, exit” model (TOTE), proposed in the field of cybernetics. This analysis indicates that the IMP and the TOTE highlight complementary aspects of goal-orientedness. In order to illustrate this point, the paper reviews three computational architectures that implement various aspects of the IMP and the TOTE, discusses their main peculiarities and limitations, and suggests how some of their features can be translated into specific mechanisms in order to implement them in artificial intelligent systems.

The paper proposes the framework for an animat control system (the Animat Brain) that is based on the Petr K. Anokhin’s theory of functional systems. We propose the animat control system that consists of a set of functional systems (FSs) and enables predictive and purposeful behavior. Each FS consists of two neural networks: the actor and the predictor. The actors are intended to form chains of actions and the predictors are intended to make prognoses of future events. There are primary and secondary repertoires of behavior: the primary repertoire is formed by evolution; the secondary repertoire is formed by means of learning. This paper describes both principles of the Animat Brain operation and the particular model of predictive behavior in a cellular landmark environment.

This paper describes the integration of several cognitively inspired anticipation and anticipatory learning mechanisms in an autonomous agent architecture, the Learning Intelligent Distribution Agent (LIDA) system. We provide computational mechanisms and experimental simulations for variants of payoff, state, and sensorial anticipatory mechanisms. The payoff anticipatory mechanism in LIDA is implicitly realized by the action selection dynamics of LIDA’s decision making component, and is enhanced by importance and discrimination factors. A description of a non-routine problem solving algorithm is presented as a form of state anticipatory mechanism. A technique for action driven sensational and attentional biasing similar to a preafferent signal and preparatory attention is offered as a viable sensorial anticipatory mechanism. We also present an automatization mechanism coupled with an associated deautomatization procedure, and an instructionalist based procedural learning algorithm as forms of implicit and explicit anticipatory learning mechanisms.

We present a theoretical analysis of schema-based design (SBD) , a methodology for designing autonomous agent architectures. We also provide an overview of the AKIRA Schema Language (AKSL) , which permits to design schema-based architectures for anticipatory behavior experiments and simulations. Several simulations using AKSL are reviewed, highlighting the relations between pragmatic and epistemic aspects of behavior. Anticipation is crucial in realizing several functionalities with AKSL, such as selecting actions, orienting attention, categorizing and grounding declarative knowledge.

Visual forward models predict future visual data from the previous visual sensory state and a motor command. The adaptive acquisition of visual forward models in robotic applications is plagued by the high dimensionality of visual data which is not handled well by most machine learning and neural network algorithms. Moreover, the forward model has to learn which parts of the visual output are really predictable and which are not because they lack any corresponding part in the visual input. In the present study, a learning algorithm is proposed which solves both problems. It relies on predicting the mapping between pixel positions in the visual input and output instead of directly forecasting visual data. The mapping is learned by matching corresponding regions in the visual input and output while exploring different visual surroundings. Unpredictable regions are detected by the lack of any clear correspondence. The proposed algorithm is applied successfully to a robot camera head under additional distortion of the camera images by a retinal mapping. Two future applications of the final visual forward model are proposed, saccade learning and a task from the domain of eye-hand coordination.

Some visual search tasks require the memorization of the location of stimuli that have been previously focused. Considerations about the eye movements raise the question of how we are able to maintain a coherent memory, despite the frequent drastic changes in the perception. In this article, we present a computational model that is able to anticipate the consequences of eye movements on visual perception in order to update a spatial working memory.