Purchase this article with an account.
Alejandro Lleras, Ronald A. Rensink, James T. Enns; Rapid Resumption is modulated by high-level strategies.. Journal of Vision 2004;4(8):180. doi: 10.1167/4.8.180.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
If a display is interrupted while participants search for a target (a T among Ls) they are able to resume search rapidly when the display reappears (Enns, Rensink, Vandenbeld & Lleras, 2003). This has been called Rapid Resumption (RR) because whereas correct responses to an initial display do not begin before 500 ms have elapsed, many responses following its reappearance are initiated within 100–400 ms. In this study we tested for the influence of high-level strategies on RR, by varying the likelihood that the display would reappear. On repeated-look trials, each presentation was 100 ms, separated by intervals of 900 ms. On single-look trials, only one 100 ms presentation occurred. Experiment 1 included a random mix of 80% repeated looks and 20% single looks; Experiment 2 contained an equal mix of repeated and single looks (50%). If RR depends on participant's strategy then it should be reduced in Experiment 2, where repeated looks were much less likely to occur. Also, when uncertain about the target identity, participants may withhold responses while waiting for confirmation if they know reappearance is likely. Consistent with a strategic component, the results for repeated-look trials showed much stronger evidence for RR in Experiment 1 than in Experiment 2. The results of the single look trials also supported this interpretation: participants were more likely to find the target in a single look when these trials were more frequent. However, neither experiment provided evidence that participants were withholding responses following a single-look: responses made once participants realized the display would not reappear contained an equal number of correct and incorrect responses. These findings show that (1) RR can be influenced by strategy, but (2) these do not include the strategic withholding of responses. Instead, RR is interpreted as an index of the interactions that normally occur between sensory input and perceptual hypotheses in everyday vision.
Grant from Nissan Motor Company Ltd. to Ronald Rensink and James Enns and NSF postdoctoral fellowship to Alejandro Lleras.
This PDF is available to Subscribers Only