The responsiveness of biological motion processing areas to selective attention towards goals

Abstract

A growing literature indicates that visual cortex areas viewed as primarily responsive to exogenous stimuli are susceptible to top-down modulation by selective attention. The present study examines whether brain areas involved in biological motion perception are among these areas-particularly with respect to selective attention towards human movement goals. Fifteen participants completed a point-light biological motion study following a two-by-two factorial design, with one factor representing an exogenous manipulation of human movement goals (goal-directed versus random movement), and the other an endogenous manipulation (a goal identification task versus an ancillary color-change task). Both manipulations yielded increased activation in the human homologue of motion-sensitive area MT+ (hMT+) as well as the extrastriate body area (EBA). The endogenous manipulation was associated with increased right posterior superior temporal sulcus (STS) activation, whereas the exogenous manipulation was associated with increased activation in left posterior STS. Selective attention towards goals activated a portion of left hMT+/EBA only during the perception of purposeful movement-consistent with emerging theories associating this area with the matching of visual motion input to known goal-directed actions. The overall pattern of results indicates that attention towards the goals of human movement activates biological motion areas. Ultimately, selective attention may explain why some studies examining biological motion show activation in hMT+ and EBA, even when using control stimuli with comparable motion properties.

Figure 1

Task Design. This figure contains snapshots of the point-light videos used in the experiment. The paradigm followed a two-by-two design with main effects of Question (goal-focused versus color change-focused) and Stimulus Type (human movement with a clear goal versus plausible but random human movement). The top row depicts a trial that includes a goal-directed animation (saluting), whereas the bottom row shows a randomly moving animation. Goal and color questions were counterbalanced between trial types. This figure is a schematic of the actual stimuli used (the dots here are increased in size, and the figures cropped differently).

Figure 2

Main effects of Question and Stimulus. Areas responsive to Question (Goal > Color) are in red. Areas responsive to Stimuli (Goal > Random) are in blue. Areas where these overlapped are in gold. The top left and right panels show hMT+/EBA activation projected onto the outer surface of the brain in the left and right hemispheres (respectively). The bottom left and right panels cut into the temporal lobes to illustrate protrusions of these clusters into posterior STS. All clusters are significant at a family-wise error corrected threshold of p < .05.

Figure 3

Simple effect of Question (Goal > Color) for Random stimuli. Panel a: clusters in posterior (circled in green) and anterior (circled in yellow) STS. Panel b: coronal view of posterior STS cluster (circled in green). All clusters were significant at a family-wise error-corrected threshold of p < .05. Boxplots below each picture show averaged standardized parameter estimates (z-values) associated with each level of each contrast (compared to fixation).