Purchase this article with an account.
Ricardo Max, Yehoshua Tsal, Marisa Carrasco; Flanking Distractors are Recognized and Suppressed Before the Target is Identified. Journal of Vision 2018;18(10):725. doi: 10.1167/18.10.725.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Introduction. In typical flanker tasks, responses to a target (with two alternative identities) that is flanked by incongruent distractors (bearing the alternative target identity) are slower compared to neutral distractors. We separately assessed the processing timecourses of incongruent and neutral distractors by conducting speed-accuracy trade-off (SAT) analyses on six experiments employing the mutations protocol (Max & Tsal, 2015). Methods. While the target remained unchanged until response, distractors mutated once, at a random time within 100 ms following stimulus onset. There were three trial types: incongruent distractors mutated to neutral; vice versa; or neutral to neutral (control trials). Comparisons among trial types within each mutation time, revealed the time-window of information extraction from distractors. Mutations effects. Incongruent distractors were suppressed 29 ms later than neutral distractors (69 vs 40 ms after onset). Targets in incongruent displays were identified 41 ms later than in neutral displays (104 vs 63 ms), which delayed responses to incongruent displays by 22 ms, compared to neutral displays (429 vs 407 ms). SAT analyses. For all trials on which distractors mutated during the initial 33 ms after onset, responses executed slower than 220 ms resulted in performance above chance level (50%). For trials mutated later than 33 ms, responses executed faster than 140 ms remained at chance level. All responses slower than 140 ms on neutral-to-neutral and neutral-to-incongruent trials rendered above chance performance. Yet, surprisingly, on incongruent-to-neutral trials, responses executed between 140 and 220 ms plunged significantly below chance (32%, p=.003). In sum, when incongruent distractors were presented during at least the initial 33 ms, fast responses reflected the distractors' identity rather than the target's. Conclusions. Challenging conventional assumptions, these results suggest that immediately following stimulus onset, most resources are initially invested into recognizing and suppressing distractors. As distractors become suppressed, resources become increasingly available for target identification.
Meeting abstract presented at VSS 2018
This PDF is available to Subscribers Only