Purchase this article with an account.
Rosanne L. Rademaker, Ilona M. Bloem, Peter De Weerd, Alexander T. Sack; Properties of high-fidelity visual working memory representations for orientation. Journal of Vision 2013;13(9):804. doi: 10.1167/13.9.804.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Visual working memory serves as a highly efficient buffer for maintaining information that can no longer be accessed directly. While most investigations have emphasized the capacity limitations of working memory, here we investigated the fidelity of remembered visual representations when memory load is kept well within capacity confines. First, we examined memory fidelity for the orientation of a grating, and whether it changes over time. Participants were briefly shown a randomly oriented grating that was remembered for 1-12 seconds, after which they reported the orientation in memory by method-of-adjustment. Memory for a single orientation proved highly stable, with a <2º increase of response variability when comparing the shortest and longest intervals. No evidence was found to imply forgetting of orientation information, even at very long retention intervals. Next, we introduced a second, to-be-ignored grating, presented midway through a fixed retention interval. We parametrically varied the orientation of this task-irrelevant grating relative to the orientation in memory. We found a reliable (2-3º) shift in participant’s response distributions, towards the orientation of the irrelevant grating. Thus, despite explicit instructions to ignore the irrelevant second grating, orientation information present in this stimulus exerted a systematic bias on participant’s responses. We investigated the role of attention in another experiment by making the second grating task-relevant – it was probed with equal probability as the first grating. This almost completely abolished the bias exerted by the second grating on responses to the first. Interestingly, when the second grating was probed, a stable shift towards the orientation of the first grating was still present. Our results suggest that visual working memory for orientation, while robust, displays a small but systematic bias towards task-irrelevant information that conflicts with same-feature information being held in memory.
Meeting abstract presented at VSS 2013
This PDF is available to Subscribers Only