Purchase this article with an account.
Yoko Higuchi, Yoshiyuki Ueda, Jun Saiki; Variability during learning facilitates generalization in contextual cueing. Journal of Vision 2015;15(12):963. doi: 10.1167/15.12.963.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Visual search performance is facilitated when fixed spatial configurations are presented repeatedly, an effect known as contextual cueing (Chun & Jiang, 1998). Although a previous study showed that contextual cueing occurred even when objects were jittered (Chun & Jiang, 1998, Experiment 6), it is not clear how variability influences contextual cueing. In the current study, we tested the hypothesis that variability during learning facilitates generalization in contextual cueing. Participants were asked to search for a rotated T target among L distractors, and to judge whether the target was rotated to the left or right. In Experiment 1, participants were assigned to either a jittered-pattern or a fixed-pattern group. During the learning phase, objects of repeated configuration were independently jittered trial by trial in the jittered-pattern group, while they were invariant in the fixed-pattern group. In a subsequent test, we presented “similar” patterns that were not presented in the learning phase. Both groups showed significant contextual cueing effects in the learning phase, but in the test, only the jittered-pattern group showed cueing effects. Thus, learning transferred to the “similar” patterns only in the jittered-pattern group, suggesting that variability is of sufficient importance to generalize contextual cueing. Experiment 2 expands these findings by using different jitter ranges. Participants were assigned to one of three groups: high, middle, or low variability group. During a learning phase, objects were jittered according to gauss distributions, and in a subsequent test, they were presented at high-probability or low-probability locations. The results showed that participants in the high and middle variability groups made equally fast responses to high-probability and low-probability location, while participants in the low variability group made faster responses to high-probability location than low-probability location. These results suggest that higher variability leads to flexible representation that can be generalized to similar patterns in contextual cueing.
Meeting abstract presented at VSS 2015
This PDF is available to Subscribers Only