This thesis presents a study of the provision of emotions for artificial agents with the ultimate aim of enhancing their autonomy, i.e. making them more flexible, robust and self-sufficient. In recent years, the importance of emotions and their assistance to cognition has been increasingly acknowledged. Emotions are no longer considered undesirable or simply useless. Their role in various aspects of human and animal cognition like perception, attention, memory, decision-making and social interaction has been recognised as essential. The importance of emotions is much more evident insocial interaction and therefore much of the emotions research done in artificial systems focuses on the expression and recognition of emotions. However, recent neurophysiological research suggests that emotions also play a crucial part in cognition itself. This thesis investigates ways in which artificial emotions can improve autonomous behaviour in the domain of a simple, but complete, solitary learning agent. For this purpose, a non-symbolic emotion model was designed and implemented. It takes the form of a recurrent artificial neural network where emotions influence the perception

The adaptive value of emotions in nature indicates that they might also be useful in artificial creatures. Experiments were carried out to investigate this hypothesis in a simulated learning robot. For this purpose, a non-symbolic emotion model was developed that takes the form of a recurrent artificial neural network where emotions both depend on and influence the perception of the state of the world. This emotion

The fact that emotions are considered to be essential to human reasoning suggests that they might play an important role in autonomous robots as well. In particular, the decision of when to interrupt on-going behaviour is often associated with emotions in natural systems. The question under examination here is whether this role of emotions can be useful for a robot which adapts to its environment. For this purpose, an emotion model was developed and integrated in a reinforcement learning framework. Robot experiments were done to test an emotion-dependent mechanism for the automatic detection of the relevant events of a learning task, against more traditional approaches. Experimental results are presented that conrm that emotions can be useful in this role, specically by improving the efficiency of the learning algorithm.

The purpose of this article is to develop a preliminary comprehensive model of antisocial development based on dynamic systems principles. The model is built on the foundations of behavioral research on coercion theory. First, the authors focus on the principles of multistability, feedback, and nonlinear causality to reconceptualize real-time parent–child and peer processes. Second, they model the mechanisms by which these real-time processes give rise to negative developmental outcomes, which in turn feed back to determine real-time interactions. Third, they examine mechanisms of change and stability in early- and late-onset antisocial trajectories. Finally, novel clinical designs and predictions are introduced. The authors highlight new predictions and present studies that have tested aspects of the model. An enormous amount of high-quality research has focused on understanding the development of antisocial behavior. The study of aggressive and antisocial development encompasses a large variety of broad theoretical perspectives (e.g., behavioral, cognitive, social) and diverse disciplines (e.g., psychology, sociology, epidemiology). The mechanisms that are studied in relation to the

ith different meaning, dependingon their content andthe context in which they are used. People may then have nearly asmany selvesas they have significant interpersonal relationships(Sullivan, 1953; see also Kelly, 1955), providing for both contextual variability andthe longstandingrepresentationsasa chronic influence. We assess idiosyncratic knowledge representations in memory and track their influence on affect and mot ivat ion . We also examine how self-regulato ry p rocesses furt her modulat e t hese responses. Our conceptualization focuseson the ways the self is linked to other people who are (or hadbeen) significant, who have hadan impact on one s life, and in whom one is (or once was) emotionally invested. Because mental representat ions of significant others and t heir relat ional linkages to the self are central in t he model, t he emotional investmentsone has in significant othersplay a role in determiningone sresponses, includingone s self-regulato ry effo r ts. One has a rel

Viewing emotion from an evolutionary perspective, researchers have argued that simple responses to affective stimuli can be triggered without mediation of cognitive processes. Indeed, findings suggest that positively and negatively valenced stimuli trigger approach and avoidance movements automatically. However, affective stimulus response compatibility phenomena share so many central characteristics with nonaffective stimulus response compatibility phenomena that one may doubt whether the underlying mechanisms differ. We suggest an ‘‘affectively enriched’’ version of the theory of event coding (TEC) that is able to account for both affective and nonaffective compatibility, and that can account for the observation that both types of compatibility seem to be modulated by goals and intentions. Predictions from the model are tested in an experiment where participants carried out approach and avoidance responses to either the valence or the orientation of emotionally charged pictures. Under affective instruction the positive-approach/ negative-avoid mapping yielded faster responses than the positive-avoid/negativeapproach mapping, but no such effect was observed under spatial instruction. Conversely, spatial compatibility effects were obtained under spatial, but not under affective instruction. We conclude that affective and nonaffective compatibility effects reflect the same mechanism.

Abstract not found

The focus of this study is the possible influence analogy making may have upon emotions experienced in everyday life. More specifically we address the question of whether emotions inherently associated with one domain tend to be transferred to another domain via analogical mapping connections. In an experiment we demonstrate that emotional appraisal is indeed a subject to analogical transfer. We also find a substantive difference between the situation in which the two domains are meaningfully mapped and the situation in which these are simply associated because of being encountered simultaneously. We also demonstrate that at least in cases when there is no clear distinction between the two domains in terms of which domain is the base and which domain is the target there is a bidirectional emotional transfer which affects emotional attitudes towards objects from both the domains. Thus whenever a positively charged object is mapped onto a negatively charged object the positively charged object receives negative valence from the negatively charged object and vice versa.

1 Psychology is often described as the scientific study of behavior. In practice it studies many other things, including thoughts and feelings, and indeed by some measures the direct observation of behavior has been disappearing from many laboratories and journals (Baumeister, Vohs, & Funder, 2007). Yet in principle the study of thoughts, feelings, and other phenomena is justified partly on the basis that understanding these things will help illuminate behavior. This chapter focuses on the relationship between emotion and behavior. It will present two main theories about that relationship. They are not equals. One is widely accepted, is simple, and enjoys the benefits of tradition and parsimony. The other has none of those advantages. By rights, therefore, the one deserves to be given the benefit of the doubt, and the second theory should only be considered seriously if the first one is found to be seriously inadequate to account for the evidence. But I shall propose that it has finally been revealed by the gradual accumulation of evidence to be seriously inadequate if not downright wrong. Hence a new theory is needed — preferably one that can fit the observed facts, especially including the ones that have gradually discredited

Four studies demonstrated that affective preferences persevere even if the cognition that gave rise to the affect is invalidated. In Study 1, participants learned associations between Chinese ideographs and their English meanings that were either positively or negatively valenced; subsequently, they were informed that these meanings were incorrect. Despite this cognitive invalidation, the affect associated with the ideographs persevered. Studies 2 and 3 ruled out alternative explanations. Study 4 showed that invalidating information containing opposite affective valence succeeded in changing the affective perseverance, whereas the neutral cognitive invalidation of Studies 1 through 3 did not. The present research provides an empirical demonstration of affective perseverance, specifying the conditions where affective preferences do and do not persevere when confronted with invalidating information. Discussion centers on issues of underlying mechanism and implications for other psychological phenomena. A ffective dispositions to objects and persons are difficult to change (Zajonc, 1980). Once an individual forms a positive or negative feeling, to a certain city, a model of a car, or a type of food, then that feeling tends to persist. Others may assail the individual with facts and evidence to sway the feeling—the pros and cons of the given city, the fuel efficiency of the new car, the voluminous calories of the food—but once formed, the affect tends to persevere. As these examples suggest, the tendency of affect to persevere despite evidence to the contrary is a very common phenomenon covering a rich range of experiences. Feelings are often independent of facts and evidence and it seems to be the case that it is harder to change how a person feels than what a person believes to be true. Yet, although affective perseverance has been speculated about in studies of belief perseverance (e.g., Edwards &