Lateral Biases in Attention and Working Memory Systems
Neurologically healthy individuals misbisect their visual space by erring towards the left. This misreprentation has been attributed to the right hemisphere dominance in processing of spatial information. Lateral biases are thought to emerge as behavioural outcomes of cognitive processing, mainly attention. Recently, attention mechanisms have been reported to be closely inter-related to memory systems, where attention directs what will be remembered and memory impacts where attention is directed. Although spatial biases attributed to attention have been widely accepted, the claim that memory exhibits similar biases has been more controversial. Recent research shows that recall of representations is biased towards the left side of space, indicating that lateral asymmetries may not necessarily be limited to perceptual and attentional mechanisms, but may extend to memory mechanisms as well. The purpose of this work is to understand better the relationship between lateral biases within working memory and attention interactions. Two approaches were considered. First, working memory, as defined by the representations and operations related to manipulate the representations, used time delay and visual load. Second, backward masking was used to control the relative formation of the working memory trace, which strengthens with recurrence of the visual stimuli and is through to progress from attention to working memory. To explore these two theoretical avenues, a novel task was constructed. Two circular arrays were presented at the top and bottom of the computer screen. These arrays were composed of six individual discs of varying shade. Hence, the overall array represented a greyscale gradient, where discs on one lateral side were darker compared to the middle discs and the other lateral side. For example, if two darkest discs were presented on the left side, the lightest discs were presented on the right side. Such array was presented with its left/right mirror reversed image. In this example, the second array was with the lightest discs on the left side and growing progressively darker, with darkest discs on the right side. Such presentation requires the participants to integrate the array of individual discs into an overall representation to perform a brightness judgement and select the array seemingly darker. A total of six behavioural studies addressed the two theoretical approaches. The first approach, to determine the impact of inter-stimulus time interval and visual load on lateral asymmetries, was addressed in four experiments. The findings indicated that participants were able to integrate the discrete disks into an overall array. Participants exhibited an overall leftward bias similar to that obtained in attention tasks, where they selected an array to be overall darker when the darkest disks were presented on the left side of the array. Furthermore, these biases increased the most when the stimuli were presented in the lower half of the computer screen, consistent with the lower visual field. Conversely, stimuli presented to the upper half of the screen elicited a rightward bias, which is consistent with the upper visual field. Stronger biases were observed when the stimulus noise, in the form of black, white and grey pixels, was relatively low and weaker biases were attained with a relatively high noise levels. In the second study, the findings showed that the magnitude of upper and lower visual field biases shows dependence on the vertical and lateral stimulus manipulation within these fields. Upper-left, lower-right interactions indicate that biases may not simply rely on the horizontal and vertical dimensions, as previously thought, but also on the relative spatial distribution of stimuli within these dimensions. The third study, which used the standard rectangular greyscale stimuli, revealed that visual load does not impact the lateral biases, but shows to impact the upper and lower visual field processing. Further, time interval between stimulus presentation and response, extended past 1 second eliminated lateral and vertical biases. The remaining two studies investigated lateral asymmetries within working memory by selectively manipulating the formation of working memory trace using backward masks. The presence of a mask, following a stimulus, inhibits the memory trace formation for that stimulus. Conversely, if no mask is presented following the stimuli, the memory trace is permitted to form within working memory. Again, using the circular array task, participants were required to select the overall darker array while retaining either a shade of position information from the array within their working memory. Findings showed increased rightward biases when memory trace was permitted to form with longer inter-stimulus (3 sec) time interval, as compared to shorter (0 to 1 sec) time interval. In the last study, the participants were required to make brightness judgement while maintaining either a position or shade information within working memory to determine whether previously acquired information, which does not serve as a cue, impact the brightness judgement task. Rightward biases were evident when participants were required to maintain either a position or shade information relating to the array, but did not provide any cue-type of information, which could facilitate performance. Rightward biases were stronger while retaining position information and completing the brightness task, hence indicating a spatial nature of the bias. As well, stronger rightward biases were obtained when the to-be-remembered position information was allowed to create a memory trace. Furthermore, recall accuracy of the position information was increased when the memory trace was permitted to form, indicating involvement of working memory processes. Overall, the data attained in this set of experiments can be interpreted using the activation-orientation model presented by Reuter-Lorenz (1990) indicating that this model may also be valuable when integrating working memory in addition to attentional processes.
DegreeDoctor of Philosophy (Ph.D.)
CommitteeKalynchuk, Lisa; Gray, Jack; Borowsky, Ron
Copyright DateApril 2014