Hans Liljenström (ed.)Advances in Cognitive Neurodynamics (IV)Advances in Cognitive Neurodynamics10.1007/978-94-017-9548-7_2

Computational Emotions

Mauricio Iza Miqueleiz¹ and Jesús Ezquerro Martínez²

(1)

Department Basic Psychology, University of Malaga, Malaga, Spain

(2)

Department of Logic and Philosophy Science, University of Basque Country, Donostia, Spain

Mauricio Iza Miqueleiz

Email: iza@uma.es

Abstract

Research on the interaction between emotion and cognition has become particularly active in the last years, and many computational models of emotion have been developed. However, none of these models is able to address satisfactorily the integration of emotion generation and its effects in the context of cognitive processes. This work tries to unify several models of computational emotions for embodied agents with the work done in cognitive architectures, based on psychological theories and applications.

Keywords

Emotion processesComputational modelsEmbodied agentsCommunication

1 Introduction: Computational Models

Computational models take different frameworks in research and applications [1]. On the one hand, psychological models emphasize on fidelity with respect to human emotion processes. On the other hand, AI models evaluate how the modeling of emotion impacts reasoning processes or improves the fitness between agent and its environment. That is to say, the model improves and makes more effective the human-computer interaction.

Several models have been proposed and developed. However, some fundamental differences arise from their underlying emotional constructs. For instance, as we will see below, some discussions on whether emotion precedes or follows cognition disappears if one adopts a dynamic system perspective. Here, we will discuss three main approaches.

2 Discrete Approach

This approach assumes appraisal theories of emotion [2]. In appraisal theories emotions are connected with how organisms sense events, relate them to internal needs, characterize appropriate responses and recruit cognitive, physical and social resources to act in adaptive ways.

Some models focus on appraisal as the core process to be modeled. In this sense, emotion is not completely elaborated. Mechanisms for deriving appraisal variables, via if-then rules, model specific emotion label. Emphasizing on a cognitive model of the situation, many models assume that (i) specific appraisal patterns are needed for emotion arising; and (ii) cognitive responses are determined by these appraisals.

This perspective is mainly concentrated on the cognitive structure of emotions, but doesn’t account for the overall emotion process. As we will see below, the resulting computational models reflect this limitation [3]. Embodied emotion is considered as a dynamic and situated process, adjusting to the changing demands of the environment, rather than an appraisal of cognitive representations [4].

3 Dimensional Approach

Dimensional theories argue that emotions are not discrete entities. Rather, it is a continuous dimensional space [5]. These theories conceptualize emotion as a cognitive label attributed to a perceived body state, mood or core affect [6]. An agent is considered in an affective state at a given moment and the space of possible states within broad, continuous dimensions.

Only gold members can continue reading. Log In or Register to continue