Purchase this article with an account.
M. Dishon-Berkovits, A. Treisman; Feature binding: Competing needs in working memory and long term associative learning?. Journal of Vision 2001;1(3):214. doi: https://doi.org/10.1167/1.3.214.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The aim of the current experiments was to elucidate the representation in visual working memory (VWM) of purely visual object information free of confounds of spatial cues, using a change detection paradigm. We explored two questions: 1. The relation between VWM and long term memory (LTM). Are the two forms of storage independent? Does the need to clear working memory interfere with acquisition of information about long term contingencies? If not, can such knowledge be used to help increase the capacity of working memory? 2. Are features on different dimensions automatically stored in working memory as aspects of an integrated object? We used an indirect measure of memory for irrelevant features and for their binding to relevant ones. Participants made a same-different judgment of one relevant dimension, shape or color, and we tested whether their performance is affected by a matching or conflicting judgment on the other (irrelevant) dimension. Additionally, for one group of participants, we introduced a correlation between the relevant and irrelevant dimensions. For the other group of participants, the pairings were all randomly selected. The question of interest was whether participants in the correlated group would pick up the redundancies and use them to facilitate the working memory tests of recognition, and whether there would also be a long term learning of the contingencies. We found evidence for working memory storage of bound features within trials, explicit learning of long term contingencies between features, but no use of that long term learning to enhance working memory performance. The results are discussed within a larger framework of the relations between VWM and LTM.
This PDF is available to Subscribers Only