Int J Soc Robot (2013) 5:35–51 DOI 10.1007/s12369-012-0169-4 Perception and Generation of Affective Hand Movements Ali-Akbar Samadani · Eric Kubica · Rob Gorbet · Dana Kuli´ c Accepted: 14 September 2012 / Published online: 10 October 2012 © Springer Science+Business Media Dordrecht 2012 Abstract Perception and generation of affective move- ments are essential for achieving the expressivity required for a fully engaging human-machine interaction. This pa- per develops a computational model for recognizing and generating affective hand movements for display on an- thropomorphic and non-anthropomorphic structures. First, time-series features of these movements are aligned and converted to fixed-length vectors using piece-wise linear re-sampling. Next, a feature transformation best capable of discriminating between the affective movements is obtained using functional principal component analysis (FPCA). The resulting low-dimensional feature transformation is used for classification and regeneration. A dataset consisting of one movement type, closing and opening the hand, is considered for this study. Three different expressions, sadness, happi- ness and anger, were conveyed by a demonstrator through the same general movement. The performance of the de- A. Samadani () · D. Kuli´ c Department of Electrical and Computer Engineering, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, N2L 3G1, Canada e-mail: asamadan@ecemail.uwaterloo.ca D. Kuli´ c e-mail: dkulic@ece.uwaterloo.ca E. Kubica Department of Systems Design Engineering, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, N2L 3G1, Canada e-mail: eric.kubica@uwaterloo.ca R. Gorbet Center for Knowledge Integration, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, N2L 3G1, Canada e-mail: rbgorbet@uwaterloo.ca veloped model is evaluated objectively using leave-one-out cross validation and subjectively through a user study, where participants evaluated the regenerated affective movements as well as the original affective movements reproduced both on a human-like model and a non-anthropomorphic struc- ture. The proposed approach achieves zero leave-one-out cross validation errors, on both the training and testing sets. No significant difference is observed between participants’ evaluation of the regenerated movements as compared to the original movement, which confirms successful regeneration of the affective movement. Furthermore, a significant effect of structure on the perception of affective movements is ob- served. Keywords Affective hand movements · Affective computing · Functional data analysis · Feature extraction · Human affect perception · Affective movement generation 1 Introduction Humans are adept at perceiving and estimating the affective states of others [77]. One important channel of affect com- munication is motor behavior (e.g., body and hand move- ment). Studies in psychology suggest that body movements and posture are important observable features of underly- ing affective states [13, 14, 78]. Humans associate different body movements and postures with distinct affective expres- sions (e.g., anger is associated with frequent tempo changes) [2, 16, 30, 52, 82], and are able to identify the feeling en- coded in a displayed movement even when the demonstrator tries to conceal its expression (e.g., negative body language) [8, 22, 55]. Moreover, the psychology literature reports on the human tendency to ascribe human-like social and affec- tive attributes to non-anthropomorphic structures such as ab-