Inferring mental overload based on postural behavior and gestures

Humans rely on body gestures and posture when communicating. This topic has been covered in great detail by researchers from various fields. Within this work, approaches to transfer findings in psychology and behavioral studies regarding the relation between gestures and emotions to machine learning methods, will be investigated. Knowledge about the users emotional state is important to achieve human like, natural HCI in modern technical systems. The main focus lies on discriminating between mental overload and mental underload, when completing a given task, which for instance can be useful in an e-tutorial system. Mental underload is a new term used to describe the state a person is in when completing a dull or boring task. A suggestion how the affective states of overload and underload can be expressed using the established notation in the Valence, Arousal and Dominance (V,A,D) space will be given. In a further step it will be shown how to select suitable features, such as gestures, movement and postural behavior patterns. Based on the features selected, a classifier is designed and trained capable of deciding whether a person is experiencing mental overload or underload.