A Review of Affective Computing Research Based on Function-Component-Representation Framework

Inferring Causal Factors of Core Affect Dynamics on Social Participation through the Lens of the Observer

“…More abstractly, valence and arousal together form a unified, continuous state-space. Emotions and core affect are thus defined at different levels, and by no means can the former be conceived as a discretised version of the latter, or vice versa the latter as a continuous representation of emotions, each emotion being a “point” in the 2D continuous space of valence and arousal—an unjustified statement which, unfortunately, is by and large assumed as a working hypothesis in the affective computing practice [ 1 ]. Cogently, the categorical nature of emotions provides the bridge between the individual’s, perceiver-independent biology of the brain and body and socially real categories that in turn allow for the sharing of emotions among individuals, i.e., the understanding by agreement of everyday concepts such as “fear” and “happiness”.…”

“…Albeit simple, such mapping suffices to conceptually capture the mapping process that, on the one hand, is assumed at the labelling level; on the other hand, it lies at the heart of the majority of affective methods. In the latter case, and typically in current end-to-end models [ 1 ], the might represent features extracted from multimodal data, and the mapping function (regression/classification) is shaped in the form of some complex architecture, e.g., a deep neural network (which generalises Equation ( 1 ) to a nonlinear mapping ).…”

“…The problem: context and importance. Consider the scenario in Figure 1 , a canonical one in affective computing [ 1 , 2 , 3 ]. Participants are involved in an interaction task (here, we will consider a dyadic one), while a third person (the rater or “labeller”) observes their behaviour and rates both the evolution of their affective experience and subsequently its social behaviour.…”

Wuthering heights: gauging fear at altitude in virtual reality

“…Affective computing is a vibrant area of interdisciplinary research (see, e.g. [29,46,59,65] for up-to-date overviews). Unfortunately, it is often the case that basic concepts and working assumptions are loosely defined; also, the same terms are occasionally adopted to denote different classes of phenomena.…”

“…-as typically derived from Basic Emotion Theories (BET, e.g., Ekman's Neurocultural Theory [30,31]) with Russell's dimensional representation of affect (over the valence/arousal dimensions, [56]). The first one is usually referred to as a discrete representation of emotions; the second, as a continuous representation of emotions [46]. This clearly is, at best, an incorrect statement: Russell's psychological construction view of emotions posits emotion and affect as different phenomena.…”