Purchase this article with an account.
Guy Snir, Yaffa Yeshurun; Temporal grouping enables selection of multiple targets in rapid streams of visual information. Journal of Vision 2017;17(10):1190. doi: 10.1167/17.10.1190.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
When two targets are embedded in a rapid serial visual presentation (RSVP) the ability to report the second target (T2) is impaired if it is temporally close to the first target (T1): The attentional blink. This deficit diminishes with consecutive targets, even if three targets are employed. However, T1 identification deteriorates (a cost) when trailing targets are spared from the blink. The present study evaluated a theory that addresses both sparing and cost effects. The theory suggests that sequential stimuli are represented together in volatile "snapshots" formed during an initial processing stage. Within a snapshot, the representational quality of stimuli decays as they grow "older". A snapshot may be selected and sent to a limited-capacity processing stage where stimuli are differentiated and consolidated. With multiple-target RSVP, snapshot selection is subject to strategic constraints: Early selection might prevent identification of later targets, while late selection might prevent individuation of early targets. These principles suggest that T1 identification would deteriorate with successful identification of consecutive targets, and that snapshot selection timing would vary with the temporal distribution of targets. Experiment 1 tested these predictions using streams with different target distributions (T1T2T3, T1T2DT3, T1DT2T3; D=Distractor). As predicted, regardless of stream type, T1 identification was better when earlier snapshots were selected. Additionally, different distribution of targets led to different patterns of snapshot selection. The theory suggests that early targets are missed because their representational quality is low when selection occurs, whereas late targets are missed primarily because they appear after selection. Thus, manipulating the targets' representational quality should have the strongest effect on T1 identification and the weakest effect on T3 identification. This prediction was confirmed using manipulations of target-distractor discriminability and the targets' presentation time in Experiments 2 and 3, respectively. Overall, the results support a temporal grouping account of selection in the attentional blink procedure.
Meeting abstract presented at VSS 2017
This PDF is available to Subscribers Only