Abstract
When participants see an exogenous cue in a region in space, they respond more quickly to discrimination targets that subsequently appear in that region. This effect on reaction time is standardly attributed to spatial visual attention being drawn to the initial cue. We set out to determine whether this type of attention could be adapted to a spatial offset. Participants performed a simple cued discrimination task with the caveat that cues and targets were never spatially congruent. Rather, targets appeared in adjacent locations to cues. Furthermore, targets were more likely to appear (e.g. 85%) to one side of the cue than the other side. Over 384 trials of training, participants demonstrated a robust facilitation when targets appeared in the more probable location. Interestingly, the majority of participants reported no explicit knowledge of the cue-target contingencies. Following training, participants completed a generalization phase, wherein non-predictive cues appeared in novel locations. In the first 100 trail block, participants responded more quickly to targets appearing to the side of the cue that had been more probable in the training phase of the task. This effect was transient, fading in subsequent blocks. Thus, training transiently generalized to novel cue locations. Finally, in a separate experiment, participants underwent a modified training procedure wherein the cue was visually masked. Participants did not report awareness of the cue, but there was a clear attention effect for targets appearing at cued locations as opposed to adjacent locations. However, participants did not adapt to cue-target contingencies under these conditions, showing no preference for more probable adjacent locations. In summary, attention to an exogenous cue can be adapted to a spatial offset, the adaptation occurs without explicit knowledge, it generalizes to untrained spatial locations, and it is dependent on visual awareness of the cue.
Meeting abstract presented at VSS 2018