Cite as:
Pollick F E, Kay J, Heim K, Stringer R, 2002, "A review of gender recognition from gait" Perception 31 ECVP Abstract Supplement

A review of gender recognition from gait

F E Pollick, J Kay, K Heim, R Stringer

It is often claimed that point-light displays of gait provide sufficient information to recognise properties such as gender. We have examined several aspects of the recognition of gender from point-light displays including: (i) a meta-analysis of the numerous experimental studies of gender recognition; (ii) examination of the predictions of the centre-of-moment calculation of Cutting and colleagues based on human hip and shoulder dimensions provided by anthropometric databases; (iii) examination of the recognition of gender from point-light arm movements. Results from the meta-analysis indicate performance generally equivalent to the low levels of performance originally reported by Kozlowski and Cutting (1977 Perception & Psychophysics 21 575 - 580), for gait-only displays, and that recognition increases when gait is coupled with other activities. Results from the centre-of-moment calculation indicate that centre-of-moment information is sufficient for recognising gender, but owing to overlapping distributions performance is not perfect. Results from the human perception of arm movements indicate that arm movements alone are sometimes sufficient for recognising gender. Taken together, these results indicate that gender recognition is typically a difficult task that is potentially aided by incorporating information from the entire body.

[Supported by the Nuffield Foundation, the Wellcome Trust.]

These web-based abstracts are provided for ease of seaching and access, but certain aspects (such as as mathematics) may not appear in their optimum form. For the final published version of this abstract, please see
ECVP 2002 Abstract Supplement (complete) size: 1753 Kb