Witchel, Harry J, Westling, Carina E I, Tee, Julian, Needham, Rob, Healy, Aoife and Chockalingam, Nachiappan (2014) A time series feature of variability to detect two types of boredom from motion capture of the head and shoulders. In: ECCE 2014 European Conference on Cognitive Ergonomics, 1-3 September 2014, Vienna.

Preview

PDF - Accepted Version Download (1MB) | Preview

Abstract

Boredom and disengagement metrics are crucial to the correctly timed implementation of adaptive interventions in
interactive systems. psychological research suggests that
boredom (which other HCI teams have been able to partially quantify with pressure-sensing chair mats) is actually a composite: lethargy and restlessness. Here we present an innovative approach to the measurement and recognition of these two kinds of boredom, based on motion capture and video analysis of changes in head and shoulder positions. Discrete, three-minute, computer-presented stimuli (games, quizzes, films and music) covering a spectrum from engaging to boring/disengaging were used to elicit changes in cognitive/emotional states in seated, healthy volunteers. Interaction with the stimuli occurred with a handheld trackball instead of a mouse, so movements were assumed to be non-instrumental. Our results include a feature (standard deviation of windowed ranges) that may be more specific to boredom than mean speed of head movement, and that could be implemented in computer vision algorithms for disengagement detection.

Item Type:	Conference or Workshop Item (Paper)
Schools and Departments:	Brighton and Sussex Medical School > Brighton and Sussex Medical School Brighton and Sussex Medical School > Neuroscience School of Media, Film and Music > Media and Film
Subjects:	Q Science > QA Mathematics > QA0299 Analysis. Including analytical methods connected with physical problems T Technology > T Technology (General)
Depositing User:	Harry Witchel
Date Deposited:	27 Jun 2014 05:32
Last Modified:	27 Jul 2017 17:07
URI:	http://srodev.sussex.ac.uk/id/eprint/49089

View download statistics for this item

📧 Request an update