Our perception of gender from biological movement: It's adaptation suggests brain neurons selective for this task.
You have probably had the experience of sitting in a railway car and watching a train slowly moving past you for a period. After it has passed the stationary platform opposite you seems to be moving in the opposite direction. This is an example of motion adaptation that reflects changes in simple feature detecting neurons very early in our visual pathway. What about a higher order feature like gender discrimination from body movement? We can readily identify the gender of a human walk, even when the only visible information comes from lights attached to the major joints: Female (left), Neutral (middle), and Male (right) of the body (the point light walker, PLWer). Jordan et al have shown that viewing a female PLWer for a period of time increases the probability of viewing a subsequent ambiguous PLWer as male, and vice versa. As no single light alone conveys sufficient information, this gender discrimination requires the integration of local information into a global percept. Because prolonged viewing of PLWers is found to produce systematic shifts in gender discrimination of subsequent ambiguous stimuli, this provides evidence of gender-specific adaptation that cannot be accounted for by adaptation of constitutive lower-order features like those in motion adaptation. Rather, it suggests the existence of neurons or groups of neurons selective for gender, as derived from biological motion. These neurons might be in cortical areas previously identified as gender-related.