The Influence Of Visual Working Memory Representations On Attentional Guidance

Visual search is a common daily cognitive task.Performing visual search task,people have to search task-related targets and inhibit task-irrelated distractors,which reflects the interaction of attention and visual working memory(VWM).Attention and VWM are two important aspects of cognitive ability.Attention reflects the ability to select information for further processing and VWM reflects the ability to store and process information over short intervals.These two cognitive processes are often been found interact with each other.Generally speaking,maintaining search targets in VWM can guide attention to the matched stimulus in visual field quickly,which promote visual search.However,whether maintaining distractors in VWM promote visual search is still controversal.Firstly,what decides whether VWM representations guide attention? Some researchers suggested that the state of VWM representations influenced whether VWM representations guide attention,only the VWM representation in active state can guide attention.While other researcheres suggested that the allocation of VWM resources influenced whether VWM representations guide attention,only the VWM representation received more resources can guide attention.Secondly,how do VWM representations guide attention,when there were more than one VWM representation? Some researchers suggested that only a single active VWM representation at a time can guide attention;when there were more active VWM representations,no representation can guide attention.While other researcheres suggested that when the two VWM representations receive an equal amount of resources,both two representation can guide attention.Thirdly,what's the manner of attentional guidance by VWM representaions? Some researchers suggested that VWM representations guided attention in a feature-based manner,because maintaining feature bindings in VWM cost more attention than maintaining features,which consumed the resource for visual search.While other researcheres suggested that VWM representations guided attention in a object-based manner,because maintaining feature bindings in VWM induce greater activation in occipital cortex than maintaining features,which promote the match between VWM representation and search items.This study try to answer these questions by manipulating the resource devoting to VWM processing.In theory,it contributes to explain the mechanism of VWM-based attentional guidance and further understand the relationship of attention and VWM.In practically,it contributes to many daily search,such as security check.This study explored the influence of VWM representation on attentional guidance though three parts: the precision of VWM representation,the number of VWM representations,and the manner of attentional guidance.In the first part,four experiments was designed to explore the underlying mechanism of attentional guidance for one single-feature VWM representation via contingent capture paradigm,by using behavioural experiment(Experiments 1 and 2),event-related potential(ERP)technique(Experiments 3),and event-related optical signal(EROS)technique(Experiment 4).In this part,participants had to perform a VWM task and a gap-location search task that was inserted into the maintaining interval of the VWM task.In the VWM task,participants had to retain one color in VWM to perform a change detection task.The precision of VWM representation was manipulted by changing the difference magnitude between the memory color and the probe color.Participants had to devote more resource to VWM processing under high precision condition than under low precision condition.The result showed that,greater attentional capture effect was observed in high precision condition than that in low precision condition(Experiment 1a and 1b),and the attentional capture effect in high precision condition was not affected by the state of VWM representation(Experiment 2).During the maintenance phase of VWM,items maintained in high precision elicited larger negative slow waves(NSW)and late positive complex(LPC)(Experiment 3),and greater activation in frontal cortex and occipital cortex(Experiment 4)than that maintained in low precision.During the search task,there was larger N2 to distractors,smaller N2 pc to targets in high precision condition(Experiment 3),greater activation in frontal cortex,and greater inhibition in occipital cortex(Experiment 4)than in low precision condition.These results indicated that when the VWM representation matched with distractors in the visual search task,the precision of VWM representation influences whether or not VWM captures attention and its underlying mechanism is that maintaining VWM representations under high precision requirement cost more resources,and therefore the matched distractors capture more attention and search targets acquire fewer resources.In the second part,four experiments was designed to explore the underlying mechanism of attentional guidance for two single-feature VWM representations via contingent capture paradigm and target search paradigm,by using behavioural experiment(Experiments 5-7)and ERP technique(Experiment 8).In Experiment 5,the tasks were same with the gap-loaction search task in Experiment1,except there were two colors had to be retained in VWM.In Experiments 6-8,the gap-loaction search task was changed to target search task.The result showed that,under contingent capture paradigm,two VWM representations simultaneously guided attention in high precision condition,while only one VWM representation guided attention in low precision condition(Experiment 5);under target search paradigm,two VWM representations simultaneously guided attention in both high and low precision condition(Experiment 6).The dual target cost was smaller in high precision condition than that in low precision condition(Experiment 6),while the dual target cost was not affected by extra VWM load(Experiment 7).During the maintenance phase of VWM,two targets in high precision condition elicited comparable NSW and LPC with one target,while two targets in low precision condition elicited larger NSW and LPC than one target(Experiment 8).During the search task,searching one target and searching two targets elicited comparable N2 pc.Under high precision condition,searching one target and searching two targets elicited comparable SPCN,while under low precision condition,searching two targets elicited smaller SPCN than searching one target(Experiment 8).These results indicated that the number of VWM representations that guide attention simultaneously was influenced by resources that devoted to VWM processing and suppression mechanism.The decline of VWM representations' quality or the increase of representation suppression could explain why the attentional guidance was worse in dual-target search than in single-taeget search.In the third part,four experiments was designed to explore the underlying mechanism of attentional guidance for two multi-feature VWM representations via contingent capture paradigm,by using behavioural experiment(Experiments 9-11)and ERP technique(Experiment 12).The results showed that,when distractor matched with VWM representation on one feature,memorizing one multi-feature VWM representation elicited greater attentional capture effect than memorizing one single-feature VWM representation(Experiment 9),and memorizing two features or the bingding of a two-feature VWM representation caused comparable attentional capture effect(Experiment 10).When participants were asked to memorize two features,distractors with incorrect bindings and distractors with correct bindings of VWM representations caused comparable attentional capture effect.When participants were asked to memorize bindings,distractors with correct bindings caused larger attentional capture effect than distractors with incorrect bindings(Experiment 11).During the maintenance phase of VWM,memorizing bindings elicited larger NSW and LPC than memorizing two features.During the search task,no matter memorizing bindings or two features,distractors with correct bindings or incorrect bindings elicited comparable N2.When participants were asked to memorize two features,targets elicited comparable N2 pc when distractors mateched with correct bindings or incorrect bindings.When participants were asked to memorize bindings,targets elicited smaller N2 pc when distractors mateched with correct bindings than distractors mateched with incorrect bindings(Experiment 12).These results indicated that when the VWM representation matched with distractors in the visual search task,the manner of attentional guidance was affected by resources that devoted to VWM processing and the matched magnitude between distractors and VWM representations.The decline of resource used to inhibit distractors and the greater matched magnitude both contributed to object-based attentional manner.The main conclusions are as follows:(1)Resource that devoted to VWM processing decides whether a VWM representation guides attention or not.(2)Resource that devoted to VWM processing influences the number of VWM representations that guide attention simultaneously.(3)Resource that devoted to VWM processing influences the manner that VWM representation guides attention.