The Influence Of Location-based Statistical Regularity On Attention Selection

Generally,because we need to filter and attenuate part of the received information,the choice of attention is crucial to our life.Recently,some studies thought that in addition to top-down and bottom-up factors,previous experience(i.e.,selection history)plays an important role in the influence of attention selection and can last for a long time,that is,attention experience will cause attention bias,and one way is statistical learning(SL).Although the existence of the historical effect of statistical learning has been fully confirmed,it is still unclear how this factor affects visual search.As for whether the high and low probability of the position presented by manipulating the target can cause inhibition,whether the priority map formed by manipulating the target position probability can only be used to process the target dimension or only play a role in the attention based on the dimension or feature level,and thus can not be used to suppress the salient distractors alone.Based on previous studies,these issues are still unclear.Therefore,we propose the following three questions:The first question is,which stages(pre-search stage,search stage,reaction stage)of visual search are affected by attention bias/spatial enhancement effect caused by target location probability learning? The second question is,which stages(pre-search stage,search stage,reaction stage)of visual search are affected by the attention inhibition/spatial filtering effect caused by distractor location probability learning? The third question is,can spatial suppression due to statistical regularities be driven by target activation?This paper explores the above problems through a series of experiments,using the additional singleton paradigm and manipulating the statistical regularities of target location or distractor location.In the Exp1,32 college students were selected as subjects,and the probability distribution of the target’s position was manipulated by using the additional singleton paradigm that encouraged the subjects to apply the singleton detection strategy.The Exp2 also took 32 college students as subjects,and adopted the additional singleton paradigm.Different from the Exp1,we manipulated the probability distribution of distractor location in Exp2.The results showed that the SL effect of target location and distractor location was mainly reflected in the search phase(related to explicit attention guidance).To a small extent,promoting reaction selection may also contribute to the SL effect of the target location,but it does not contribute to the SL effect of distractor location.However,the main difference between the two forms of SL occurs in the early stage or the pre-search stage(involving implicit attention),which is reflected in that the SL of the distractor location affects the pre-search stage,while the SL of the target location does not affect the pre-search stage.The Exp3 recruited 64 college students as participants,designed as an inter-group design,manipulated the probability distribution of target location and used the variation of additional singleton paradigm.Different from Exp2,in Exp3,we manipulates the probability of the target appearing in different positions(including a low probability target location and other high probability target locations),while highlighting the equal probability of the distractor appearing in each position.In Exp3,the number of stimuli was set to four conditions(search array size was set to 4,6,10 and 12).64 participants were divided into four groups(16participants in each group)and took part in the experiment of four conditions of the number of stimuli.The results showed that in the larger search array(10 and 12)where both the target and the distractor were more significant,the distractor/spatial inhibition still appeared when there was only a deviation of the target location probability,but in the smaller search array(the number of stimuli was 4 and 6),the distractor/spatial inhibition did not appear when only the probabilities of target location were biased.Our conclusion is that search array size(the number of stimuli in the search display)may be a potential key factor or boundary/limiting condition,which determines whether the SL at the target location can inhibit or filter the salient distractors.Therefore,this study draws the following conclusions: the SL effect of target location on visual search are not the same as that of distractor location,in other words,the two forms of SL affect the performance of attention capture and visual search in partly shared and partly independent cognitive mechanisms(a hybrid scenario could also be considered);For the SL at the target location,an important factor or boundary/constraint condition that determines whether it can inhibit the salient distractor or determine whether it has different effects on the attention capture in visual search may be the search array size(the number of stimuli in the search display).