Abstract
Numerous studies have demonstrated a link between spatial attention and alpha (8-13 Hz) activity measured with electroencephalography (EEG). In spatial cueing studies, a posterior alpha desynchronization is seen contralateral to the cued visual hemifield relative to the ipsilateral side (e.g., Thut, Nietzel, Brandt, Pascual-Leone, 2006). In light of recent work showing that the pattern of alpha activity across the scalp tracks the content of visual working memory (Anderson, Serences, Vogel, & Awh, 2014), we examined whether the topography of alpha activity provides precise information about the locus of spatial attention. We recorded EEG while participants performed a spatial cueing task. A central cue (87.5% valid) directed participants to one of eight placeholders arranged in a circle around fixation. After 1250 ms a search array was presented and participants were asked to identify the digit among letters. A robust spatial cueing effect confirmed that participants attended the cued location. We examined whether the distribution of EEG activity across the scalp carried information about the attended location during the cue-target interval. Using a linear classifier, we identified the frequencies between 4 and 30 Hz that allowed for above-chance decoding of the locus of attention. Activity in the alpha frequency band carried information about the cued location, beginning ~500 ms after cue onset and sustaining until onset of the search array. Using a forward encoding model of location selectivity, we were able to reconstruct location-based channel tuning function (CTFs) from the topography of alpha activity that tracked the position of attention. Consistent with the classification results, the CTF emerged ~500 ms after cue onset. These results support the hypothesis that alpha activity plays a role in the deployment of spatial attention, and suggest that CTFs provide a time-resolved measure for tracking the deployment of covert spatial attention.
Meeting abstract presented at VSS 2015