'Serial-Like' Sampling of Visual Objects During Sustained Attention

Jianrong Jia,Fang,Huan Luo
DOI: https://doi.org/10.1167/17.10.1306
2017-01-01
Journal of Vision
Abstract:In a crowded visual scene, attention must be efficiently and flexibly distributed over time and space to accommodate various contexts. It is well established that selective attention enhances the corresponding neural responses, presumably implying that during sustained attention (e.g., focusing on one over others), attention would persistently dwell on the task-relevant item. Meanwhile, recent studies propose that attention, rather than being stationary, is essentially a dynamic process that organizes copious information temporally. However, how the brain dynamically coordinates attention among multiple visual objects during sustained attention remains largely unknown. To address the issue, here we employed electroencephalography and a temporal response function (TRF) approach to extract object-specific neuronal impulse responses in sustained attentional tasks. First, we demonstrate that attentional process, characterized by inhibitory alpha (~10 Hz) activities in TRFs, switches between the attended and unattended locations, suggesting rhythmic sequential sampling even during sustained attention. Furthermore, the attentional spatiotemporal pattern is modulated by task contexts such that the alpha-mediated switching becomes increasingly prominent as the task requires more uniform attentional distribution. Finally, this attentional spatiotemporal profile maintains in a multiple-object-tracking (MOT) paradigm, indicating a general temporal organization mechanism for object-based attention. Our results suggest that sustained attention, in addition to the classically posited attentional 'focus', contains a dynamic mechanism for monitoring all objects outside of the focus. Our findings also advocate that attention is intrinsically dynamic, acting as a series of concatenating attentional chunks that operate on one object at one time. Meeting abstract presented at VSS 2017
What problem does this paper attempt to address?