Affective computing in the modern workplace

Abstract

Affective computing refers to a class of computer-based solutions that analyse data about human behaviour, such as facial expressions, gestures, and language, for its emotional information. The term was first coined 25 years ago when the ability for computers to perform basic sensorimotor tasks such as object detection in images was in its infancy. This article revisits the subject and considers how it is being applied in real-world applications today. We look at research from cognitive science informing our understanding of emotions and how computing capabilities have advanced in recent years to produce cognitive algorithms capable of detecting human attention, emotion and health. Affective computing offers the potential to revolutionise how we incorporate emotion as information in communications and decision systems. However, much of the underlying research that forms the foundations for emotion detection is being challenged, raising concerns about the ethics, trustworthiness and viability of such platforms. This article presents a critique of the technologies and appraises their suitability for real-world applications as part of a modern workplace.