Skip to main content
NCBI home page
PsyCh Journal logoLink to PsyCh Journal
. 2024 Mar 26;13(5):717–725. doi: 10.1002/pchj.748

Working memory components modulation of attentional disengagement from evaluative distractor

Minmin Yan 1, Yanying Tian 2, Min Hai 2, Bohua Zhang 1, Antao Chen 3,✉
PMCID: PMC11444721  PMID: 38530871

Abstract

It is important for people to disengage attention from a distraction, which can help them complete the task at hand as quickly as possible. Recent studies have shown that people's attention stays longer on reward‐distractors than on loss‐distractors, and a delay in attentional disengagement is noted when reward‐distractors are present. However, few studies have examined whether attentional disengagement from an evaluative distractor relies upon working memory (WM) components. In the present study, we used an attentional disengagement paradigm in which reward‐ or loss‐distractors were presented at a central location and the target was presented at a peripheral location, in combination with different WM tasks. The results from Experiment 1 showed that participants were slower to disengage their attention from a central reward‐distractor than a loss‐distractor regardless of cognitive load when the phonological loop component of WM was involved. The results from Experiment 2 revealed that people had difficulty in shifting their attention away from a reward‐distractor in comparison to a loss‐distractor when spatial WM was low, whereas no such difference was observed when spatial WM was high. We conclude that WM components differently modulate attentional disengagement from evaluative distractors. That is, the processing of evaluative (reward and loss) distractors may rely on the same cognitive resources as the spatial WM component, but not the phonological loop component.

Keywords: attentional disengagement, reward, value salience, working memory components

INTRODUCTION

Bright headlights on the road, red flowers in the green grass—these physical saliences can quickly capture an individual's attention, yet they are not meaningful to people, who can then swiftly shift their focus away from these stimuli (Awh et al., 2012; Gaspelin & Luck, 2018). However, researchers propose that value salience is quite distinct from physical salience (i.e., stimulus‐driven) or goal salience (i.e., goal‐driven) (Anderson et al., 2011; Rusz et al., 2020; Theeuwes, 2019), and is mainly based on prior experience. Accordingly, it is rather difficult for people to disengage attention from these value salience stimuli (Watson et al., 2020).

A recent study showed that value salience plays an important role in modulation of attentional disengagement (Yan et al., 2022, 2023). In that study, a distractor was presented at a central location of display after a central fixation disappeared, and participants had to disengage their attention from the central‐distractor as soon as possible to search for the target at a peripheral location within a limited time. Specifically, a reward‐ /loss‐distractor predicted that if participants found the target quickly, they would receive reward/avoidance of loss, and the monetary quantity of gaining reward and avoiding loss was equal. Participants have difficulty disengaging from a reward‐distractor relative to a loss‐distractor, even if they understand that staying longer on reward will result in loss of reward. In brief, evaluative distraction has a reliable impact on attentional disengagement.

Notably, completing an attentional disengagement task would deplete working memory (WM) resources, which is similar to attentional selection tasks (additional singleton task). Initially, it is essential to maintain different characteristics of distractors and the target in visual WM. Second, in the additional singleton task participants had to inhibit the interference of distractors and to locate the target, which requires spatial WM. Many studies have shown that performance on attentional selection tasks was impaired when visual, spatial, or central executive components were heavily occupied by other tasks (Burnham et al., 2014; Lavie & De Fockert, 2005). Moreover, these different components of WM can affect attentional selection. For example, the phonological loop component of WM was occupied by a secondary task that did not impair performance on an additional singleton task, whereas visual or spatial WM being occupied would impair this performance (Burnham et al., 2014; Woodman & Luck, 2004). Based on this, we hypothesized that evaluative distractor modulation of attentional disengagement is mainly dependent on visual or spatial components, but not phonological loop components of WM. That is, WM components may play a key role in attentional disengagement of evaluative distractor.

We designed two experiments to investigate whether WM components affect attentional disengagement from evaluative distractors. Experiment 1 involved the phonological loop component of WM, while Experiment 2 involved spatial WM. We manipulated low and high WM load that were combined with the attentional disengagement paradigm developed by Watson et al. (2020). That is, before the attentional disengagement task, participants were instructed to retain either a low or high WM load, and they were asked to recall the content of the WM task after completing the attentional disengagement task. In the attentional disengagement task, a distractor associated with a value (reward or loss) was presented at the central location of the display, while the target was presented at a peripheral location. As attention was focused on the central location after the central fixation disappeared, participants had to disengage their attention from the central distractor in order to search for the target. Previous studies have suggested that, although participants acknowledge that both reward‐associated and loss‐associated stimuli offer equal economic benefits, they still exhibit delayed disengagement from a reward distractor compared to a loss distractor (Yan et al., 2022). Thus, we predicted that attentional disengagement from evaluative distractors would be present when the phonological loop component was occupied, regardless of whether the WM load was high or low. In other words, the results from Experiment 1 would show that the reward distractor would hold attention longer than the loss distractor, which was equally strong for both high and low WM loads. We predicted that the difference in attentional disengagement between the reward and loss conditions would be reduced at a high‐spatial‐WM load compared to a low/no‐spatial‐WM load in Experiment 2.

EXPERIMENT 1

Methods

Participants

G*Power software (Faul et al., 2007) indicated that a sample of 24 participants would allow for the detection of a main effect of WM load and value‐distractor type, as well as an interaction between them, with a medium effect size (f = 0.25) and a power of (1‐β) = 0.80 (α = .05). A total of 25 college students, ranging in age from 18 to 25 years, participated in the experiment (20 females; age: M = 19.88 years, SD = 1.45). All of the students reported that they were in good physical and psychological health, with no history of psychiatric or neurological disorders. The ethics committee of the School of Psychology at Southwest University in China approved the experimental protocol.

Stimuli and design

Attention disengagement task

Subjects were instructed to perform experiments in a separate soundproof room. The experimental stimuli were presented on a black background, controlled by E‐prime Version 2.0. The computer screen was an 18.5‐inch display with a resolution of 1366 × 768 pixels and a refresh rate of 60 Hz. We used the attentional disengagement task developed by Watson et al. (2020), and we added a WM load based on their task. In the disengagement task, a fixation was presented at the center of the screen for 500–700 ms. Then, a search display was presented for 2000 ms after a 150‐ms blank screen disappeared. The search display consisted of a set of seven shapes, in which only one was a diamond (the target) and the other six were circles. Specifically, one of the circles was a colored distractor located at the center of the search display (central distractor). The diamond target and five circles were equally distributed, forming a circular ring with a diameter of 10 degrees of visual angle (d.v.a.). The location of the diamond was randomly chosen from one of six peripheral locations in each trial. In each of the six circles, there was a line segment oriented at either 45 or 135°, and within half of the circles, the line segment was oriented at 45°. The orientation of the line segment in the target was either 0 or 90° (horizontally or vertically).

The distractors were imbued with blue, orange, or green and were considered salient trials, each with a different value. The non‐salient trial was the no‐distractor condition, where all shapes were gray. Notably, the value of a non‐salient trial was neutral, which was identical to the value of a green‐salient distractor. In the search display, a blue circle was associated with reward, while an orange circle was associated with loss, and a green circle was associated with neutrality. Participants needed to make a response in the search display, using the “F” key on a keyboard to represent a horizontal line segment in the target and the “J” key to represent a vertical line segment. The content of the feedback screen differed according to the participants' responses. In the reward condition, the content of the feedback screen was “Congratulations, +10 points” for 2000 ms if the participants made a correct response; otherwise, it was “+1 point.” In the loss condition, the content of the feedback screen was “–1 point” if the participants made a correct response; otherwise, it was “Sorry, –10 points.” In the neutral and no‐distractor conditions, irrespective of the correct or incorrect response, the content of the feedback was “0 points.” Meanwhile, the cumulative scores of the search task were presented.

There were a total of eight formal blocks. There were 48 trials in each formal block, one‐third of which contained distractors associated with reward value (16 trials), one‐third of which contained distractors associated with loss value (16 trials), and one‐third of which contained neutral distractors and no distractors associated with neutral value (eight trials in each condition, respectively). The type of value distractor was randomly selected and presented at the center. Only after completing two practice blocks could participants continue to the formal blocks.

WM task

The manipulation of WM load was similar to that of Watson et al. (2019). For high WM load, each trial began with a 1000‐ms display of a set of six randomly ordered numbers (1, 2, 3, 8, 9, and 0). After a 500‐ms blank screen, participants performed an attentional disengagement task, as previously mentioned. After the task ended, participants were asked to make a keypress response to the number covered by the question mask. They also needed to recall the number that randomly appeared at one of the six locations. For low WM load, only one digit was presented at the beginning of each trial. After completing the disengagement task, participants made keypress responses to the digit covered by the question mark. After each block, the feedback screen displayed the mean accuracy of the memory task. There were eight blocks with visual search and WM load tasks combined, alternating between blocks with low and high WM load.

Procedure

Figure 1 illustrates the experimental procedure. The participants were instructed that monetary compensation was dependent on their performance in the digit WM task and the attentional disengagement task. They were required to complete practice trials before the formal experiment, and were allowed to memorize the numbers by silently reading or rehearsing them. Half of the participants were informed that the blue circle, which was presented at the central location of the search display, was associated with reward (while the orange circle was associated with loss). The other half were informed that the blue circle was associated with loss. All participants were required to search for the target at the peripheral location. They were informed: “If you can disengage attention from the reward‐distractor within 2 s and your response is correct, you will win 10 points. Otherwise, you will receive 1 point. If you can disengage attention from the loss‐distractor within 2 s and your response is correct, you will lose 1 point. Otherwise, you will lose 10 points. If you can disengage attention from the neutral‐distractor and the no‐distractor conditions within the limited time, regardless of whether your response is correct or not, you will receive 0 points.” The participants were allowed to take a break after finishing one block.

FIGURE 1.

FIGURE 1

Open in a new tab

Illustration of the experimental procedure in Experiment 1. (A) In the low memory load, each trial started with one digit for 1000 ms, and participants needed to remember it. This was followed by the attentional disengagement task, after which participants were required to recall the number by pressing the keyboard. (B) In the high load, each trial started with a memory set of six numbers for 1000 ms. After the attentional disengagement task finished, participants were required to recall one of the numbers. A question mask was presented randomly in the one of six locations.

Data analysis

To indicate that the value‐modulation effect was not overshadowed by the physical saliency effect, we used two different contrast analyses that previous studies had used (Watson et al., 2019). By comparing performance on trials with a reward distractor to those with a loss distractor or a neutral distractor, we were able to isolate the effect of value on attention disengagement, as the only difference between them was the size of value signaled by the distractor. (In the current experimental design, different colored distractors were, respectively, associated with reward, loss, and neutral value. Thus, the difference in physical saliency between the reward, loss, and neutral conditions was identical). These contrasts were compared for low and high load condition using a 2 (load: low and high) × 3 (value: reward, loss and neutral) analysis of variance (ANOVA).

By comparing the neutral‐salient distractor with non‐salient (no distractor) trials, the effect of physical salience on attention could be isolated, since these trial types had the same value (0 points) but differed in whether the search display contained a color‐singleton distractor. These contrasts were compared for low and high load conditions using a 2 (load: low and high) × 2 (salient type: salient distractor and non‐salient distractor) ANOVA.

The Bayes factor (BF) was calculated using Bayesian hypothesis testing in JASP (JASP Team, 2018) to assess the evidence for the alternative hypothesis (H1) compared to the null hypothesis (H0) when the results of traditional null hypothesis testing were not significant (van Doorn et al., 2021).

Results

Reward effect on attentional disengagement

A two‐way (2 × 3) ANOVA with within‐subject factors of the WM load (load: high and low) and distractor type (value: reward, loss, and neutral) was conducted on the response times (RTs) and accuracies (see Figure 2). The results showed a main effect for the WM load (F(1,24) = 6.16, p = .02, η 2 = .20), with faster RTs in the low‐load (M = 1108 ms) than in the high‐load condition (M = 1162 ms). A main effect for distractor type was also significant (F(2,48) = 4.70, p = .014, η 2 = .16); post hoc tests revealed that the RTs in the central‐reward distractor (p = .016, M = 1150 ms) were larger than central‐loss (M = 1125 ms), and RTs in the central‐reward distractor were not significantly larger than the central‐neutral distractor (p = .08, M = 1130 ms). There was no RT difference between the central loss and neutral distractor condition (p = .98, all p values corrected by Bonferroni method). Critically, the interaction between these two factors was not significant (F(2,48) < 0.01, p = .99, η 2 < .01, BF01 = 8.83). The Bayes factor provides moderate to strong evidence in favor of the H0. This indicates that the reward‐distractor held attention longer than the loss‐distractor, which was equally strong for the high‐ and low‐load condition. In other words, regardless of memory load, participants were slower to disengage attention from a central‐reward distractor than from a central‐loss distractor.

FIGURE 2.

FIGURE 2

Open in a new tab

Results from Experiment 1. The mean response times (left panel) and mean accuracies (right panel) between the different distractor conditions under low and high load. Individual data points are superimposed. Error bars denote ±1 standard error of the mean.

The accuracy (ACC) results revealed that accuracy in the low load was lower compared to (M = 0.96) the high load (M = 0.98, F(1,24) = 5.05, p = .03, η 2 = .17). The above analyses showed that faster responding to the low load was accompanied by lower accuracy. To test for any potential speed‐accuracy trade‐off on load, we calculated the inverse efficiency (IE) score. This is an effective indicator for weighting response times by accuracy (Bruyer & Brysbaert, 2011). Analysis of high/low load trials showed that the IE score did not differ significantly between the low load (M = 1157 ms) and the high load (M = 1185 ms, t (24) = −.94, p = .35), indicating that there was no speed‐accuracy trade‐off on load. There was no main effect for the distractor type (F(2,48) = 0.09, p = .90, η 2 = .004). Importantly, the interaction between the WM load and distractor type was also not significant (F(2,48) = 1.73, p = .18, η 2 = .06), suggesting that there was no evidence for speed‐accuracy trade‐off.

Effect of physical salience

A two‐way (2 × 2) ANOVA with within‐subject factors of the WM load (load: high and low) and salient type (salient: salient and non‐salient distractor) was conducted on the RTs and accuracies (see Figure 2). The results showed no main effect for the WM load (F(1,24) = 3.38, p = .07, η 2 = .12). A main effect for distractor type was significant (F(1,24) = 12.78, p = .002, η 2 = .34), and the RTs of the distractor with physical saliency was larger than that of the no‐distractor condition. The interaction between these two factors was not significant (F(1,24) = .43, p = .51, η 2 = .01). These results may suggest that the physical saliency effect existed regardless of the high and low load.

The ACC results revealed that there was no significant difference between the low spatial load and high spatial load (F(1,24) = 1.53, p = .22, η 2 = .06). There was no significant accuracy difference between the neutral and no‐distractor type (F(1,24) = 0.004, p = .95, η 2 < .001). The interaction between the WM load and distractor type was also not significant (F(1,24) = 0.69, p = .41, η 2 = .03).

Discussion

In Experiment 1, we investigated whether phonological WM components affect attentional disengagement from an evaluative distractor. Obviously, participants' RTs were significantly slower in the high‐load than in the low‐load condition, which indicated that our manipulation of WM load was effective. The RTs of the distractor with physical saliency were larger than those for the no‐distractor condition even in the high load, which is consistent with previous studies (Burnham, 2010). Importantly, the results showed that the central reward‐distractor held attention longer than the loss‐distractor irrespective of WM load. These results are in accordance with our hypothesis. In the current study, we used a strings of numbers as the high WM load. Participants memorized these numbers by silently reading and rehearsing them, which undoubtedly used phonological loop components by processing of verbal and acoustic information (Burnham et al., 2014; Lavie et al., 2004). These results showed that evaluative distractor modulation of attentional disengagement was not dependent on the phonological loop WM component in Experiment 1.

According to the WM model, central executive‐control is a resource‐constrained system that includes visual, spatial and phonological loop components (Baddeley, 1992, 1998, 2012). We needed to further examine whether this disengagement effect by evaluative distractors was dependent on spatial memory, which was not involved in Experiment 1. We investigated this in Experiment 2.

EXPERIMENT 2

In the Experiment 2, we manipulated WM load by using spatial memory task based on Gao & Theeuwes (2020). In the high spatial memory load, participants were asked to memorize the exact location of two small diamonds that were sequentially presented. Participants had to remember different locations of two diamonds and put them together as a whole configuration representation in their visuospatial WM (Gao & Theeuwes, 2020). After completing the disengagement task, they were asked to judge whether the locations of two squares were the same as the previous locations. In the no‐spatial‐memory load, participants were instructed to view only one location of the small diamond, and they were asked to press the “space bar” instead of making a judgement. Initially, participants were asked to complete a spatial WM task, which would occupy the spatial WM component that would result in fewer spatial resources being available to process attentional disengagement from evaluative distractors. We investigated whether this performance of attentional disengagement from the evaluative distractor would be impaired under the high‐spatial‐WM load compared to the no‐spatial‐WM load.

Methods

Participants

Participants were 24 individuals ranging in age from 19 to 26 years (19 females; age: M = 21.17 years, SD = 1.92) who completed a spatial WM task and an attentional disengagement task. All participants reported being in good physical and psychological health, with no history of psychiatric or neurological disorders. The ethics committee of the School of Psychology at Southwest University in China approved the experimental protocol.

Stimuli and design

Attentional disengagement task

Attentional disengagement task of Experiment 2 was identical to that of Experiment 1. There were also four conditions: the reward, loss, neutral, and no‐distractor conditions. In the reward condition, the feedback would present “+10 points” for a correct response and “+1 point” for an incorrect response. In the loss condition, the feedback would present “–1 point” for a correct response and “–10 points” for an incorrect response. In the neutral and no‐distractor trials, irrespective of correct or incorrect response, the feedback would present “0 points.” In each feedback screen, it would display the current total scores of the search task.

There were a total of eight blocks. Each block consisted of 36 trials. One‐third of the trials contained distractors that were associated with reward value (12 trials), one‐third contained distractors associated with loss value (12 trials), and one‐third contained neutral distractors and no distractors associated with neutral value (six trials in each condition). Participants were required to complete two practice blocks before proceeding to the eight blocks.

Spatial WM task

We employed a modified spatial WM task, which was based on Woodman and Luck (2004). In the high‐spatial‐WM load, each trial began with a fixation point for 500 ms, then the first white square (0.3° × 0.3°) was presented for 500 ms. After that, the second white square was also presented for 500 ms, and the two white squares were separated by a 500‐ms interval. The white square was randomly presented at one of the quadrants. In each quadrant (four quadrants), there were 3 × 3 locations, for a total of 36 locations. After finishing the search task, participants were asked to remember the position of the two sequential squares in the high load. In the no‐spatial‐WM load, the procedure was identical to the high load, except there was only one white square randomly presented in one of the quadrants.

If the positions of two squares were the same as the previous positions of two squares, participant were asked to press the “F.” Otherwise, they were asked to press “J.” The feedback display was presented as either correct or incorrect, depending on the participants' memory performance. It was displayed for 500 ms. In the no‐spatial‐memory load, participants were required to press “space” to continue when a square appeared. There were eight blocks in total, with four low‐load blocks and four high‐load blocks presented alternately during the formal experiment.

Procedure

Figure 3 illustrates the procedure of Experiment 2. Participants were informed that they initially needed to memorize two locations of white squares for the high‐spatial‐WM load. After finishing the disengagement task, they were then asked to judge whether the locations of the two squares were the same as the previous locations. For the no‐spatial‐WM load, participants were instructed to view one location of a white square and only press the space bar instead of making a judgement. The instruction content of the visual search task (attentional disengagement task) and feedback for Experiment 2 were the same as for Experiment 1.

FIGURE 3.

FIGURE 3

Open in a new tab

Illustration of the experimental procedure in Experiment 2. (A) The spatial memory task of high load. Participants were required to remember two locations of white squares that were sequentially presented in the high spatial load. (B) In the low‐spatial‐memory load, participants simply viewed a location of the white square. This was followed by an attentional disengagement task, after which a new white square was presented and participants were required to press the “space” key to continue. (C) In the high load, after finishing the attentional disengagement task, participants were required to judge whether the current two squares appeared in the same locations as the previous two squares.

Results

Reward effect on attentional disengagement

A two‐way (2 × 3) ANOVA with within‐subject factors—the spatial WM load (load: high and low) and distractor type (value: reward, loss, and neutral)—was conducted on the RTs and accuracies (see Figure 4). The results showed a main effect for the spatial memory load (F(1,23) = 11.93, p = .002, η 2 = .34), with faster RTs in the low‐load (M = 978 ms) than in the high‐load condition (M = 1019 ms). A main effect for distractor type was marginally significant (F(2,46) = 3.08, p = .056, η 2 = .12). Importantly, there was an interaction effect between spatial WM load and distractor type (F(2,46) = 5.18, p = .01, η 2  = .18, BF01 = 0.84/BF10 = 1.2). Bayes factor indicated that the interaction was anecdotal, but it also provided evidence in favor of attentional disengagement effect being affected by WM load. Post‐hoc tests revealed that, for low‐spatial‐WM load, RTs were significantly slower when the central location was a reward‐distractor (M = 990 ms) than when it was a loss‐distractor (M = 960 ms, p < .001), but not slower than when it was a neutral‐distractor (M = 984 ms, p = .97). There was no significant difference in RTs between the loss‐distractor and neutral‐distractor condition (p = .16, all p values corrected by Bonferroni method). For the high‐spatial‐WM load, the RTs difference between central reward‐distractor (M = 1031 ms) and loss‐distractor (M = 1025 ms) was not significant (p = .99), and there is a difference between reward‐ and neutral‐distractor (M = 1001 ms, p = .05). The difference between loss‐distractor and neutral‐distractor condition was not significant (p = .41, all p values corrected by Bonferroni method). The results may suggest that participants had difficulty in discriminating value information at high‐spatial‐WM load compared to low spatial load.

FIGURE 4.

FIGURE 4

Open in a new tab

Results from Experiment 2. The mean response times (left panel) and the mean accuracies (right panel) between the different distractor conditions under the low and high load. Individual data points are superimposed. Error bars denote ±1 standard error of the mean (**p < .01, n.s > .05).

The ACC results revealed that there was no significant difference between low spatial load and high spatial load (F(1,23) = 3.95, p = .06, η 2 = .14). There was no main effect for the distractor type (F (2,46) = .58, p = .56, η 2 = .02). The interaction between the WM load and distractor type was also not significant (F(2,46) = 1.63, p = .21, η 2 = .06).

Effect of physical salience

A two‐way (2 × 2) ANOVA with within‐subject factors of spatial WM load (load: high and no) and salient type (salient: salient and non‐salient distractor) was conducted on the RTs and accuracies (see Figure 4). The results showed no main effect for the WM load (F(1,23) = 3.5, p = .07, η 2 = .13). A main effect for distractor type was significant (F(1,23) = 8.96, p = .006, η 2 = .28), and the RTs of distractor with physical saliency were larger than those for the no‐distractor condition. The interaction between these two factors was not significant (F(1,23) = 0.48, p = .49, η 2 = .02; p = .008; p = .135). These results may suggest physical saliency effect existed regardless of high and no spatial load.

The ACC results revealed that there was a significant difference between no spatial load (M = 0.95) and high spatial load (M = 0.97, F(1,23) = 18.30, p < .001, η 2 = .43). There was no significant accuracy difference between neutral and no distractor type (F(1,23) = 1.71, p = .20, η 2 = .06). Importantly, the interaction between the WM load and distractor type was also not significant (F(1,23) = 0.33, p = .56, η 2 = .01).

Discussion

In Experiment 2, we investigated whether spatial WM components affect attentional disengagement from evaluative distractors. First, participants' RTs were significantly slower in the high‐load condition compared to the no‐load condition, indicating the effectiveness of our manipulation of spatial WM load. Second, the physical salience of reward, loss, and neutral trials was equal, ruling out the possibility that the disengagement effect of evaluative distractors was caused by physical salience. More importantly, these results were in line with our hypothesis. The difference in RTs was not significant between reward and loss trials at high‐spatial‐WM load compared to no‐spatial‐WM load. This means that in the no‐spatial‐load condition, participants' RTs were slower when disengaging attention from reward than from loss distractors. However, at high‐spatial‐WM load, there was no significant difference in attentional disengagement between the two types of distractors. This may be due to the fact that spatial WM resources were occupied by the secondary task, resulting in an insufficiency of spatial WM resources that impaired the performance of attentional disengagement from evaluative distractors. Unlike in Experiment 1, the results from Experiment 2 may suggest that attentional disengagement from evaluative stimuli shares the same component with spatial WM (Feng et al., 2012).

GENERAL DISCUSSION

The current study investigated the influence of WM components on attentional disengagement from evaluative distractors. Consistent with prior studies, participants took longer to disengage attention from reward‐related distractors compared to loss‐related distractors (Yan et al., 2022, 2023). This may be due to the fact that reward outcomes are related to the dopaminergic system, and loss outcomes are implicated in the serotonergic pathway. These different neuromodulatory systems are underlying mechanisms for the differences between reward and loss conditions (Gueguen et al., 2021). It should be noted that there was no significant difference between neutral and loss‐related distractors. One potential explanation for this finding is that individuals may not perceive the reward distraction as a conflict or a distraction, while they perceive other distractors, such as loss and neutral distractors, as conflicts or distractions.

In Experiment 1, we allowed participants to memorize a string of numbers by reading silently and rehearsing it, which may indicate that the phonological loop was occupied. The results showed that there was a difference in the effect of evaluative distractors on attention disengagement when the phonological loop component was occupied by a secondary task. Specifically, the central reward‐distractor held attention longer than the loss‐distractor, regardless of the load on the phonological WM. This may be due to the fact that the phonological loop uses multiple channels, such as the verbal and acoustic channels, which may rarely depend on the central executive control. These results are inconsistent with previous studies, which have shown that the interference of the reward‐distractor is larger than that of the loss‐distractor under high load, but not under low load (Gupta et al., 2016). However, it should be noted that in their studies, Gupta et al. (2016) used perceptual load, which is different from WM load (Gupta et al., 2016; Lavie et al., 2004).

In Experiment 2, the results showed that participants' RTs were slower when disengaging attention from reward distractors than from loss distractors at no spatial load. However, the difference in RTs between reward and loss distractors was not significant at high spatial WM load. This suggests that cognitive resources were largely occupied by the spatial memory task at high load, leading to participants being unable to process the difference between reward and loss. The main difference between Experiment 1 and Experiment 2 was the WM component (phonological and spatial WM). The most likely explanation for these results is that attentional disengagement from evaluative (reward and loss) distractors relies on the same cognitive resources as the spatial WM component (Feng et al., 2012), but not the phonological loop component. However, it is difficult to determine whether the effect of high‐spatial‐WM load affects the process of attentional disengagement from distractors or the learning of color‐reward/loss associations. Attentional disengagement is considered to involve top‐down controlled mechanisms that may occur in a proactive, endogenous manner in anticipation of a target (Cohen et al., 1994; Posner et al., 1984). Electrophysiological evidence suggests that the top‐down mechanism of attentional disengagement consumes large amounts of cognitive resources, which is indicated by an increased P3 amplitude (Zhang et al., 2023). Notably, maintaining a high‐spatial‐WM load also leads to an increased P3 amplitude (Shucard et al., 2009). Due to restricted cognitive resources, when a large amount of cognitive resource is occupied by high‐spatial‐WM load, top‐down control of attentional disengagement may be interrupted or impeded. Another possibility is that maintaining the presentation of the reward/loss‐distractor or the learning of the color‐reward/loss association consumes the same cognitive resources as spatial WM. If cognitive resources were occupied by a high spatial load under limited resources, there may be insufficient resources to strengthen the learning of the color‐reward/loss association, leading to the elimination of the difference in attentional disengagement between reward and loss distractors. While both perspectives offer potential explanations, limited evidence currently supports the latter. Future research should incorporate event‐related potential techniques to elucidate this mechanism.

This study has some limitations. The current study only investigated the phonological loop and spatial WM component on disengagement effects, and did not investigate other WM components, such as visual WM component (Hollingworth & Maxcey‐Richard, 2013; Van Moorselaar et al., 2014), which could be considered in future studies.

In conclusion, this study highlights that WM components play key roles in attentional disengagement from evaluative distractors. The phonological loop and spatial WM components show different patterns in their disengagement effects from evaluative distractors, depending on the level of cognitive load. We conclude that attentional disengagement from evaluative distractors and spatial WM components share the same resources of executive control, while the phonological loop can process information through multiple channels to alleviate the load on executive control.

ACKNOWLEDGENTS

This work was supported by the National Natural Science Foundation of China (32171040).

CONFLICT OF INTEREST STATEMENT

The authors declare no potential conflicts of interest.

ETHICS STATEMENT

The study was conducted in accordance with the Declaration of Helsinki, and approved by the internal ethics committee of the Faculty of Psychology in Southwest University (Protocol code H22049). All participants provided written informed consent before taking part in the experiments.

Yan, M. , Tian, Y. , Hai, M. , Zhang, B. , & Chen, A. (2024). Working memory components modulation of attentional disengagement from evaluative distractor. PsyCh Journal, 13(5), 717–725. 10.1002/pchj.748

REFERENCES

  1. Anderson, B. A. , Laurent, P. A. , & Yantis, S. (2011). Value‐driven attentional capture. Proceedings of the National Academy of Sciences, 108(25), 10367–10371. 10.1073/pnas.1104047108 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Awh, E. , Belopolsky, A. V. , & Theeuwes, J. (2012). Top‐down versus bottom‐up attentional control: A failed theoretical dichotomy. Trends in Cognitive Sciences, 16(8), 437–443. 10.1016/j.tics.2012.06.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baddeley, A. (1992). Working memory. Science, 255(5044), 556–559. 10.1126/science.1736359 [DOI] [PubMed] [Google Scholar]
  4. Baddeley, A. (1998). Working memory. Compets Rendus Biologies, 321(2–3), 167–173. 10.1016/s0764-4469(97)89817-4 [DOI] [PubMed] [Google Scholar]
  5. Baddeley, A. (2012). Working memory: Theories, models, and controversies. Annual Review of Psychology, 63, 1–29. 10.1146/annurev-psych-120710-100422 [DOI] [PubMed] [Google Scholar]
  6. Bruyer, R. , & Brysbaert, M. (2011). Combining speed and accuracy in cognitive psychology: Is the inverse efficiency score (IES) a better dependent variable than the mean reaction time (RT) and the percentage of errors (PE)? Psychologica Belgica, 51(1), 5–13. 10.5334/pb-51-1-5 [DOI] [Google Scholar]
  7. Burnham, B. R. (2010). Cognitive load modulates attentional capture by color singletons during effortful visual search. Acta Psychologica, 135(1), 50–58. 10.1016/j.actpsy.2010.05.003 [DOI] [PubMed] [Google Scholar]
  8. Burnham, B. R. , Sabia, M. , & Langan, C. (2014). Components of working memory and visual selective attention. Journal of Experimental Psychology: Human Perception and Performance, 40(1), 391–403. 10.1037/A0033753 [DOI] [PubMed] [Google Scholar]
  9. Cohen, J. D. , Romero, R. D. , Servan‐Schreiber, D. , & Farah, M. J. (1994). Mechanisms of spatial attention: The relation of macrostructure to microstructure in parietal neglect. Journal of Cognitive Neuroscience, 6(4), 377–387. 10.1162/jocn.1994.6.4.377 [DOI] [PubMed] [Google Scholar]
  10. Faul, F. , Erdfelder, E. , Lang, A. G. , & Buchner, A. (2007). G*power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. 10.3758/bf03193146 [DOI] [PubMed] [Google Scholar]
  11. Feng, J. , Pratt, J. , & Spence, I. (2012). Attention and visuospatial working memory share the same processing resources. Frontiers in Psychology, 3, 103. 10.3389/fpsyg.2012.00103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gao, Y. , & Theeuwes, J. (2020). Learning to suppress a distractor is not affected by working memory load. Psychonomic Bulletin and Review, 27(1), 96–104. 10.3758/S13423-019-01679-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gaspelin, N. , & Luck, S. J. (2018). The role of inhibition in avoiding distraction by salient stimuli. Trends in Cognitive Sciences, 22(1), 79–92. 10.1016/j.tics.2017.11.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gueguen, M. C. M. , Lopez‐Persem, A. , Billeke, P. , Lachaux, J. P. , Rheims, S. , Kahane, P. , Minotti, L. , David, O. , Pessiglione, M. , & Bastin, J. (2021). Anatomical dissociation of intracerebral signals for reward and punishment prediction errors in humans. Nature Communications, 12(1), 3344. 10.1038/s41467-021-23704-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gupta, R. , Hur, Y. J. , & Lavie, N. (2016). Distracted by pleasure: Effects of positive versus negative valence on emotional capture under load. Emotion, 16(3), 328–337. 10.1037/emo0000112 [DOI] [PubMed] [Google Scholar]
  16. Hollingworth, A. , & Maxcey‐Richard, A. M. (2013). Selective maintenance in visual working memory does not require sustained visual attention. Journal of Experimental Psychology: Human Perception and Performance, 39(4), 1047–1058. 10.1037/a0030238 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Team Jasp. (2018). JASP (Version 0.9)[Computer software]. Retrieved from https://jasp-stats.org/2018/06/20/introducing-jasp-0-9/
  18. Lavie, N. , & De Fockert, J. (2005). The role of working memory in attentional capture. Psychonomic Bulletin & Review, 12(4), 669–674. 10.3758/bf03196756 [DOI] [PubMed] [Google Scholar]
  19. Lavie, N. , Hirst, A. , de Fockert, J. W. , & Viding, E. (2004). Load theory of selective attention and cognitive control. Journal of Experimental Psychology: General, 133(3), 339–354. 10.1037/0096-3445.133.3.339 [DOI] [PubMed] [Google Scholar]
  20. Posner, M. I. , Walker, J. A. , Friedrich, F. J. , & Rafal, R. D. (1984). Effects of parietal injury on covert orienting of attention. Journal of Neuroscience, 4(7), 1863–1874. 10.1523/JNEUROSCI.04-07-01863.1984 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Rusz, D. , Le Pelley, M. E. , Kompier, M. A. J. , Mait, L. , & Bijleveld, E. (2020). Reward‐driven distraction: A meta‐analysis. Psychological Bulletin, 146(10), 872–899. 10.1037/bul0000296 [DOI] [PubMed] [Google Scholar]
  22. Shucard, J. L. , Tekok‐Kilic, A. , Shiels, K. , & Shucard, D. W. (2009). Stage and load effects on ERP topography during verbal and spatial working memory. Brain Research, 1254, 49–62. 10.1016/j.brainres.2008.11.063 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Theeuwes, J. (2019). Goal‐driven, stimulus‐driven, and history‐driven selection. Current Opinion in Psychology, 29, 97–101. 10.1016/j.copsyc.2018.12.024 [DOI] [PubMed] [Google Scholar]
  24. van Doorn, J. , van den Bergh, D. , Böhm, U. , Dablander, F. , Derks, K. , Draws, T. , Etz, A. , Evans, N. J. , Gronau, Q. F. , Haaf, J. M. , Hinne, M. , Kucharský, Š. , Ly, A. , Marsman, M. , Matzke, D. , Gupta, A. R. K. N. , Sarafoglou, A. , Stefan, A. , Voelkel, J. G. , & Wagenmakers, E. J. (2021). The JASP guidelines for conducting and reporting a Bayesian analysis. Psychonomic Bulletin and Review, 28, 813–826. 10.3758/S13423-020-01798-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Van Moorselaar, D. , Theeuwes, J. , & Olivers, C. N. L. (2014). In competition for the attentional template: Can multiple items within visual working memory guide attention? Journal of Experimental Psychology: Human Perception and Performance, 40(4), 1450–1464. 10.1037/a0036229 [DOI] [PubMed] [Google Scholar]
  26. Watson, P. , Pearson, D. , Chow, M. , Theeuwes, J. , Wiers, R. W. , Most, S. B. , & Le Pelley, M. E. (2019). Capture and control: Working memory modulates attentional capture by reward‐related stimuli. Psychological Science, 30(8), 1174–1185. 10.1177/0956797619855964 [DOI] [PubMed] [Google Scholar]
  27. Watson, P. , Pearson, D. , Theeuwes, J. , Most, S. B. , & Le Pelley, M. E. (2020). Delayed disengagement of attention from distractors signalling reward. Cognition, 195, 104125. 10.1016/j.cognition.2019.104125 [DOI] [PubMed] [Google Scholar]
  28. Woodman, G. F. , & Luck, S. J. (2004). Visual search is slowed when visuospatial working memory is occupied. Psychonomic Bulletin & Review, 11(2), 269–274. 10.3758/bf03196569 [DOI] [PubMed] [Google Scholar]
  29. Yan, M. , Li, Q. , Long, Q. , Xu, L. , Hu, N. , & Chen, A. (2022). Evaluative distractors modulate attentional disengagement: People would rather stay longer on rewards. Journal of Vision, 22(8), 12. 10.1167/jov.22.8.12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Yan, M. , Meng, Z. , Hu, N. , & Chen, A. (2023). Rewarding outcomes enhance attentional capture and delay attentional disengagement. PeerJ, 11, e15868. 10.7717/peerj.15868 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Zhang, Y. , Ye, Q. , He, H. , Jin, R. , & Peng, W. (2023). Neurocognitive mechanisms underlying attention bias towards pain: Evidence from a drift‐diffusion model and event‐related potentials. Journal of Pain, 24(7), 1307–1320. 10.1016/j.jpain.2023.03.003 [DOI] [PubMed] [Google Scholar]

Articles from PsyCh Journal are provided here courtesy of Wiley

RESOURCES