Abstract
We recently employed the N2pc component as a real-time electrophysiological marker of attentional selectivity to demonstrate the existence of a fast mechanism for the independent parallel selection of two colour-defined target objects at different locations. When these objects were presented in rapid succession in different displays (with SOAs between 10 and 100 ms), both triggered N2pc components of similar size and with onset latencies that matched the SOA between the two displays, indicating two parallel attentional selection processes with independent time courses (Eimer & Grubert, 2014; Grubert & Eimer, 2015, 2016). Here, we investigated whether these parallel processes can operate for more than two targets simultaneously, given that multiple-object tracking studies suggested that up to four attentional foci can be maintained in parallel (Cavanagh & Alvarez, 2005). In Experiment 1, four horizontal colour targets could appear in four successive displays, each separated by a 10 ms SOA. Trials where one of the four targets appeared on the vertical midline were also included. These were subtracted from ERPs on all-horizontal target trials, to extract N2pcs separately for each of the four targets. All targets triggered reliable N2pcs, with onset latencies matching the SOA between displays, suggesting that four objects can be selected independently and in parallel. In Experiment 2, the same procedures were used, but now with eight targets in eight consecutive displays. All eight targets elicited N2pc components that approximately matched their temporal position, but N2pc amplitudes were reduced by 50% relative to Experiment 1. This difference suggests that feature-based attentional selection mechanisms can operate in parallel for up to four items, although a lower capacity limit remains possible. We discuss the implications of these results for models of attentional control.
Meeting abstract presented at VSS 2018