Document Type


Publication details

MacHatch, C 2103, 'Orientation invariant sex cells! Possible neural mechanisms for discriminating sex from movement', PhD thesis, Southern Cross University, Lismore, NSW.

Copyright C MacHatch 2013


The overarching aim of this research was to investigate the human ability to discriminate another’s’ sex based on patterns of bodily movement or biological motion. Initial experiments explored the effects of changes in stimulus orientation on sex discrimination, with particular focus on the so-called “inversion effect”. This work has revealed that, unlike in the case of other perceptual biological motion tasks that are detrimentally affected by changes in stimulus orientation, sex discrimination remains unaffected. As such the data raised the interesting possibility that the neural encoding of sex is orientation invariant. To test that possibility, a second suite of experiments was conducted, this time using the adaptation paradigm.

Recent work by two different groups of researchers (Jordan, Fallah and Stoner, 2006 and Troje, Sadr, Geyer and Nakayama, 2006) proposed the existence of sex-encoding neurons. In a second phase of this study it was aimed to explore the possibility that the neurons encoding sex are not orientation tuned by testing the existence of repulsive adaptation after-effects when the adapting and test stimuli were in either congruent or incongruent orientations. It was shown that irrespective of the congruence in the orientation of the adapting and the test walkers, repulsive adaptation after-effects can be obtained: prolonged viewing of the gait of one sex shifted the judgments of subsequent walkers towards the opposite sex. Taken as a whole, data patterns support the existence of sex-tuned cells and additionally suggest that the encoding process associated with those cells is orientation invariant.

In the remaining studies, the integration of sex cues across the visual and auditory modalities was explored. It was revealed that when visually presented sex ambiguous walkers were paired with female footsteps, they were judged to be female. Furthermore, adaptation to pairings of visual and auditory stimuli that made an ambiguous walker look more female gave rise to an after-effect such that the test walker was judged more male. Taken together, the results of this research provide evidence that the processing of sex integrates information from different sensory inputs, encodes it by a special form of central-tendency coding - namely opponent coding - and is orientation invariant.