Elsevier

Cognition

Volume 82, Issue 2, December 2001, Pages B51-B61
Cognition

Brief article
Perceiving affect from arm movement

https://doi.org/10.1016/S0010-0277(01)00147-0Get rights and content

Abstract

We examined the visual perception of affect from point-light displays of arm movements. Two actors were instructed to perform drinking and knocking movements with ten different affects while the three-dimensional positions of their arms were recorded. Point-light animations of these natural movements and phase-scrambled, upside-down versions of the same knocking movements were shown to participants who were asked to categorize the affect of the display. In both cases the resulting confusion matrices were analyzed using multidimensional scaling. For the natural movements the resulting two-dimensional psychological space was similar to a circumplex with the first dimension appearing as activation and the second dimension as pleasantness. For the scrambled displays the first dimension was similar in structure to that obtained for the natural movements but the second dimension was not. With both natural and scrambled movements Dimension 1 of the psychological space was highly correlated to the kinematics of the movement. These results suggest that the corresponding activation of perceived affect is a formless cue that relates directly to the movement kinematics while the pleasantness of the movement appears to be carried in the phase relations between the different limb segments.

Introduction

Our day to day movements serve the primary function of achieving the variety of tasks which sustain our existence. While these movements are not necessarily communicative, common experience would serve to tell us that they often do carry a message. One way to investigate such movement in isolation of other visual cues is to present it as point-lights that show only the motion of the joints (Johansson, 1973). Although these displays are impoverished they are spontaneously organized into the percept of human movement and basic levels of competence have been demonstrated for the recognition of properties of the actor and action (Barclay et al., 1978, Cutting and Kozlowski, 1977, Hill and Pollick, 2000, Kozlowski and Cutting, 1977, Mather and Murdoch, 1994, Runeson and Frykholm, 1981, Runeson and Frykholm, 1983). In this research we examine a particular aspect of the interpretation of human movement – how affect is perceived from the arm movements of knocking and drinking actions.

The problem of recognizing emotion from human movement has been explored for the special case of the interpretation of stylized dance movements (Dittrich et al., 1996, Walk and Homan, 1984). Of the six emotions examined (surprise, fear, anger, disgust, grief/sadness, joy/happiness) it was found in both studies that anger was the most reliably identified emotion. Other differences among the identifiability of the different emotions were noted though they fell into no particular pattern between the two studies. The overall rate of recognition for the six emotions reported by Dittrich et al. (1996) was 63% for point-light displays and 88% for full-video displays. These studies provide good evidence that stylized movements can be seen as expressive, but do not address the more general case of movements which are not stylized.

With affect it is possible to discuss not only the accuracy with which an affect is recognized, but also to examine the apparent structure of the representation of affect. A number of descriptive models for the structure of experienced affect have been suggested (Larsen and Diener, 1992, Russell, 1980, Thayer, 1989, Watson and Tellegen, 1985). The essential properties of the various models describe a structure that is anchored by two bipolar but independent dimensions of experience (pleasantness and activation) with the different possible affects located on a circle centered at the origin (Yik, Russell, & Barrett, 1999). This circular formation has led to such models being termed ‘circumplex’ models of affect. Both dimensions of a circumplex model correspond to the conscious experience of affect rather than pleasantness corresponding to the affect itself and activation being modulated by simple physiological arousal.

One important issue to consider is that the circumplex model has been established as a model of one's own experiences of affect and thus would not necessarily apply to the perception of affect. However, there is evidence to support the prediction that an internal model of one's own experience could serve in the perception of the movement of others. For example, a number of investigations into the perception and production of arm movements indicate that visual and motor processing interact whilst an observer watches an action (Decety and Grezes, 1999, Rizzolatti et al., in press, Rizzolatti et al., 1996, Vogt, in press-a, Vogt, in press-b). Moreover, evidence from the study of biological motion (Shiffrar and Freyd, 1990, Shiffrar and Freyd, 1993, Thornton et al., 1998) indicates that the perception of biological motion relies on more than just low-level motion detectors specialized for human movement (Mather et al., 1992, Neri et al., 1998). Thus, it appears reasonable to conjecture that an internal model of affect would be part of the processes involved in organizing the perception and categorization of affect.

Section snippets

Experiments 1 and 2

Two experiments were performed to investigate how affect is perceived from point-light displays of human movement. In Experiment 1 we presented knocking and drinking movements with ten different affects and measured the ability of participants to categorize affect. In Experiment 2 we performed the same categorization task with the same knocking movements, however the relationship among individual points was distorted by displaying the motions upside-down and with scrambled phase (Bertenthal &

Discussion

The results of Experiment 1 showed that the psychological space resulting from the perceived affect of arm movements conformed to a circumplex structure with one dimension corresponding to an activation axis and the other axis corresponding to a pleasantness axis. Moreover, the activation axis of the circumplex was correlated to physical characteristics of the movement in a consistent manner such that greater activation was related to greater magnitudes of velocity, acceleration and jerk of the

Acknowledgements

We would like to thank Mitsuo Kawato, Chris Atkeson and Andy Calder for their help on this project, and the reviewers for their helpful suggestions. In addition, we acknowledge the support of EPSRC grant (GR/M36052) to the first author and an ESRC Research Studentship to the second author.

References (34)

  • J Decety et al.

    Neural mechanisms subserving the perception of human actions

    Trends in Cognitive Sciences

    (1999)
  • G Rizzolatti et al.

    Premotor cortex and the recognition of motor actions

    Cognitive Brain Research

    (1996)
  • B.J Scholl et al.

    Perceptual causality and animacy

    Trends in Cognitive Sciences

    (2000)
  • K Amaya et al.

    Emotion from motion

  • C.D Barclay et al.

    Temporal and spatial factors in gait perception that influence gender recognition

    Perception and Psychophysics

    (1978)
  • J.N Bassili

    Emotion recognition: the role of facial movement and the relative importance of upper and lower areas of the face

    Journal of Personality and Social Psychology

    (1979)
  • B.I Bertenthal et al.

    Global processing of biological motions

    Psychological Science

    (1994)
  • J.E Cutting

    Generation of synthetic male and female walkers through manipulation of a biomechanical invariant

    Perception

    (1978)
  • J.E Cutting et al.

    Recognizing friends by their walk: gait perception without familiarity cues

    Bulletin of the Psychonomic Society

    (1977)
  • W.H Dittrich et al.

    Perception of emotion from dynamic point-light displays represented in dance

    Perception

    (1996)
  • T Flash et al.

    The coordination of arm movements: an experimentally confirmed mathematical model

    Journal of Neuroscience

    (1985)
  • H Hill et al.

    Exaggerating temporal differences enhances recognition of individuals from point light displays

    Psychological Science

    (2000)
  • G Johansson

    Visual perception of biological motion and a model for its analysis

    Perception and Psychophysics

    (1973)
  • L.T Kozlowski et al.

    Recognizing the sex of a walker from a dynamic point-light display

    Perception and Psychophysics

    (1977)
  • J.B Kruskal et al.

    Multidimensional scaling. Sage University paper series on quantitative applications in the social sciences, 07–011

    (1978)
  • R.J Larsen et al.

    Promises and problems with the circumplex model of emotion

  • B Mather et al.

    Gender discrimination in biological motion displays based on dynamic cues

    Proceedings of the Royal Society of London B

    (1994)
  • Cited by (481)

    View all citing articles on Scopus
    View full text