Electrophysiological hallmarks of location-based and object-based visual multiple objects tracking

Abstract

Converging evidence shows that our visual system can track multiple visual, independently moving items over time. This is accomplished location-based by maintaining the individual spatial information of each target item or object-based by constructing an abstract object-based representation out of the tracked items. Previous work showed specific behavioural, electrophysiological and haemodynamic markers for location-based or object-based representations of the relevant targets by probing the encoded information subsequently after tracking. However, domain-specific differences of representational correlates during visual tracking itself have not been reported yet. The current study aims to identify spectral properties of the electrophysiological signal during tracking that might indicate location-based versus object-based maintenance of visual information. Subjects had to covertly track four out of eight visually identical items for several seconds while electrophysiological signals were recorded. Subsequently, a probe consisting of four items appeared and the subjects had to indicate with a button press whether the probe matched all targets or not. Subjects employing an object-based strategy showed an enhanced gamma response during the presentation of the target items at the beginning of the trial. On the other hand, subjects using a location-based strategy showed enhanced gamma synchronization throughout the tracking itself. Both the object- and location-based gamma responses yielded identical spatial topographical field distributions. These results indicate that object-based tracking is supported by enhanced encoding during the initial presentation of the targets to be tracked. Location-based tracking is characterized by the sustained maintenance of the individual targets during the entire tracking period in that same processing network.

Keywords: gamma oscillations; multiple object tracking; object-based attention; principal component analysis.