Abstract
During apparent motion (AM), an object appears to move smoothly between several locations despite being physically present only intermittently. It has been shown that AM can modulate neural activity within regions of motion-sensitive cortex (hMT+) and primary visual cortex (V1) that represent non-stimulated regions of space along the AM path. However, it is not clear (1) where corresponding neural modulations originate: in early visual cortex due to feedback from higher visual areas or through long-range horizontal connections within early areas, and (2) what information is being represented within these regions: whether it is shape or identity information about the object itself or non-object-specific information such as position, velocity or even the focus of attention. We applied fMRI in human subjects to examine the representation of information within visual cortex along the AM path. Observers viewed AM displays in which either a circle or a star traversed back and forth between the upper-right and lower-left visual fields. Importantly, these stimuli were unlikely to trigger long-range, orientation-specific, horizontal connections within early visual cortex. In a partial replication of previous findings, we show that when a subtle occlusion cue is present, there is an increase in activity corresponding to the AM path in early visual cortex (V1, V2, and V3). However, this activity does not encode enough information to dissociate circles from stars. In contrast, patterns of activity along the AM path in higher-order topographic regions and in LOC and hMT+ are able to dissociate circles from stars even when the occlusion cue is absent. Our results suggest that object-specific information along the AM path is represented within relatively high levels of visual cortex and not the earliest stages of cortical processing. Moreover, when occlusion cues are present, non-identity-specific positional information likely arising due to feedback is represented within early visual cortex.
Meeting abstract presented at VSS 2016