کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
378304 | 659013 | 2014 | 28 صفحه PDF | دانلود رایگان |

In spite of the progress in emotion research over the past 20 years, emotions remain an elusive phenomenon. While some underlying circuitry has been identified for some aspects of affective processing (e.g., amygdala-mediated processing of threatening stimuli, the role of orbitofrontal cortex in emotion regulation), much remains unknown about the mechanisms of emotions. Computational models of cognitive and affective processes provide a unique and powerful means of refining psychological theories, and can help elucidate the mechanisms that mediate affective phenomena. This paper outlines a number of open questions and challenges associated with developing computational models of emotion, and with their integration within biologically-inspired cognitive architectures. These include the following: the extent to which mechanisms in biological affective agents should be simulated or emulated in affective BICAs; importance of more precise, design-based terminology; identification of fundamental affective processes, and the computational tasks necessary for their implementation; improved understanding of affective dynamics and development of more accurate models of these phenomena; and understanding the alternative means of integrating emotions within agent architectures. The challenges associated with data availability and model validation are also discussed.
Journal: Biologically Inspired Cognitive Architectures - Volume 7, January 2014, Pages 98–125