An exploration of representation and maintenance in visuo-spatial working memory for simultaneously and sequentially presented information
Multiple lines of evidence have indicated that the system responsible for the short term representation and maintenance of visually presented, non-verbal information - Visuo-Spatial Working Memory - is not unitary. One of the most influential dissociations observed is that between performance on a task requiring memory for patterns within a matrix (visual patterns task) and memory for sequences of locations (Corsi blocks task). A growing body of research suggests the difference of importance between these two tasks is manner of presentation - whether information is simultaneously or sequentially presented. It has been proposed that visuo-spatial working memory can be fractionated into two components - a visual and a spatial subsystem, which are considered to support memory for visual matrix patterns and sequences of locations respectively. The extent to which these components are thought to interact varies between models, and recent questions have been raised over differential use of executive resources. The current thesis investigated whether the same visual information was represented by separate subsystems on the basis of the format in which it was presented. Experiments 1-3 used interference methodology, employed during the retention interval of a recognition task which required the locations of three dots to be remembered. Presentation of primary task stimuli was of either a simultaneous or a sequential format and interference stimuli were designed to target possible visual or spatial subsystems. Passive viewing of simultaneous or sequential interference stimuli was not found to produce a reliable disruption to memory performance. A typical spatial interference task and a novel visual task requiring detection of images were found to disrupt memory for both simultaneously and sequentially presented information. There was, however, no evidence of selective interference. Experiment 4 employed memory for the appearance of a series of items to provide further evidence that the novel visual paradigm affects visuo-spatial memory rather than an imagery strategy. Experiments 5-8 investigated memory for static arrays and sequences consisting of a greater number of locations. It was found that emphasis on retaining order in sequence recall did not exacerbate differences in performance level between memory for arrays and sequences. An individual differences approach including seven other measures revealed recall of simultaneously presented information and recall of sequentially presented information were best predicted by each other, and then by different predictors. The results from the above studies suggest common resources support memory for static arrays and sequences of locations as well as indicating use of different resources. That memory for simultaneously and sequentially presented information was equally disrupted by a primarily visual and a primarily spatial interference task suggests maintenance of information may employ overlapping processes regardless of presentation format. Models of visuo-spatial working memory which advocate sequences of locations are maintained in memory by processes operating separately from those maintaining static arrays of information are hard to reconcile with the current data; it is suggested that a model which proposes separable yet interdependent subsystems provides the best account of both the results obtained and the literature reviewed.