Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Sep 17.
Published in final edited form as: Vis cogn. 2024 Aug 23;32(9-10):1027–1044. doi: 10.1080/13506285.2024.2347605

How Does Mind-Wandering Affect Distractor Suppression?

Han Zhang 1, Kevin F Miller 2, John Jonides 3
PMCID: PMC12440381  NIHMSID: NIHMS1993023  PMID: 40963703

Abstract

The current study examined whether the suppression of overt attention to a salient distractor requires attentional resources. In a feature-search task, participants searched for a constant shape among different shapes while ignoring a uniquely colored distractor. Trial-by-trial fluctuations in attentional resources were assessed via thought probes that elicited mind-wandering reports and via pre-trial pupil sizes. The results show that initial eye movements to the distractor were suppressed regardless of the availability of attentional resources. However, when mind-wandering, the presence of the distractor prolonged target looking time. Thus, the initial deployment of overt attention in this task does not require attentional resources and can proceed automatically, presumably due to strong selection history. Nonetheless, the distractor might still disrupt later processing stages, an effect exacerbated by mind-wandering. These results also suggest that initial eye movements do not fully reflect the extent of distractor interference during the entire course of visual search.

Keywords: visual search, attentional capture, mind-wandering, eye movements

Decades of research show that visual attention is often captured by salient but irrelevant objects in the environment. Consider the additional-singleton task, in which observers search for an item with a unique shape, such as a diamond among circles, all of the same color. In this task, visual search is disrupted by the presence of a differently colored item despite this “color singleton” never being the target (Theeuwes, 1992). A popular indicator of attentional capture is the landing position of initial eye movements, which reflect the initial deployment of overt attention. In the additional-singleton task, the presence of the color singleton distractor leads to an oculomotor capture effect, evidenced by a sizable portion of initial eye movements landing on the distractor (Gaspelin et al., 2017; Theeuwes et al., 2003).

But if salient distractors predominantly capture attention, people will have immense difficulties looking for a correct target. Recent studies have consistently shown that, in a variation of the original additional-singleton task, the color singleton does not produce an oculomotor capture effect (for a review, see Gaspelin & Luck, 2018b). In this variation, participants search for a fixed target among heterogeneous items (e.g., searching for a circle among squares, hexagons, triangles, etc.). On half of the trials, one of the non-target items appeared in a unique color. In this task, then, participants can engage in the search for a specific shape feature rather than searching for a unique shape. In this “feature-search” condition, response times (RTs) on distractor-present trials are typically slightly faster compared to distractor-absent trials. Critically, the analysis of initial eye movements reveals an oculomotor suppression effect, with initial fixations being less likely to land on the color-singleton compared to an average non-singleton item. It is crucial to clarify that the term “suppression” in this context specifically refers to a behavioral effect and should not be conflated with suppression at the neurophysiological level, such as neural firing inhibition (Wöstmann et al., 2022).

The oculomotor suppression effect observed in some feature-search tasks indicates that attentional capture by a salient distractor can be mitigated under certain conditions. However, the question arises whether the color-singleton in tasks that show suppression actually possesses the potential to capture visual attention (Theeuwes, 2004). The concern is that by using heterogeneous search items, the color singleton no longer stands out from the display and thus can be ignored due to its low physical salience. Wang and Theeuwes (2020), using a behavioral index of distractor processing, reported that the color-singleton in this kind of feature-search task still captured attention in displays with a high number of search items, in which the color singleton is presumed to be more salient (but see Stilwell & Gaspelin, 2021). However, Stilwell et al. (2023) quantified physical salience by measuring the exposure duration required to detect the presence of a color-singleton in a search display. They found that color singletons deemed more salient by this method nonetheless produced a stronger oculomotor suppression effect in a subsequent feature-search task.

This debate concerning physical salience is an example of the prevailing uncertainty regarding the nature of the suppression effect observed in the feature-search task. As many have pointed out (Anderson, 2021; Lamy, 2021; Leonard, 2021), distractor suppression at the behavioral level is likely an outcome of diverse cognitive processes. These processes can be interpreted from multiple perspectives. For example, ongoing research is exploring the extent to which distractor suppression is influenced by processes that proactively versus reactively guide attention (e.g., Gaspelin & Luck, 2018a), processes that up-weight target features versus those that down-weight distractor features (e.g., Chang & Egeth, 2019), processes that are feature-specific versus those that are feature-blind (e.g., Won et al., 2019), and processes that operate automatically versus those that engage attentional resources (e.g., Lien et al., 2023). The current study focuses on the automaticity of the oculomotor suppression effect. Understanding whether distractor suppression occurs automatically is crucial, not only for comprehending the nature of suppression itself but also for enhancing search efficiency in real-life scenarios.

A key criterion for determining the automaticity of cognitive processes involves assessing whether they are influenced by the availability of attentional resources (Yantis & Jonides, 1990). If observers need to exert attentional resources to counteract interference from a distractor, performance is likely to deteriorate when attentional resources allocated to the task are scarce. In contrast, if the mechanisms underlying the oculomotor suppression effect function automatically, this effect should not depend on the availability of attentional resources. Previous investigations into this matter have often involved assigning observers a secondary task, such as memorizing a list of items prior to the visual search (Gao & Theeuwes, 2020) or mixing a search task with a rapid serial visual presentation task (Lien et al., 2023). If the cognitive processes in question operate in an automatic, effortless manner, the additional task should minimally affect the primary task. However, even without introducing an external task, observers experience natural variations in their attentional state, leading to temporal fluctuations in available attentional resources for the primary task. One such attentional state is mind-wandering, the impact of which on visual search will be examined in the current study.

Mind-wandering

Our mind often drifts away from the ongoing task to task-unrelated thoughts, a phenomenon known as mind-wandering (Smallwood & Schooler, 2006). Typically measured using thought probes, mind-wandering has been reported to occur frequently (30% ~ 50%) in both laboratory and real-life settings (Kane et al., 2007; Risko et al., 2013; Seli et al., 2015; Smallwood et al., 2012; Varao-Sousa et al., 2018). Meanwhile, despite the substantial oculomotor suppression effect observed in the feature-search task, that suppression is not complete. In this task, initial eye movements still erroneously land on the distractor some 4 ~ 8% of the time (Adams et al., 2022; Gaspelin et al., 2017; Gaspelin & Luck, 2018b). This raises the question: is mind-wandering responsible for these distractor-landing eye movements?

Mind-wandering consumes the same attentional resources that are needed for a primary task (Smallwood & Schooler, 2006). Thus, it is no surprise that performance is compromised during mind-wandering in tasks that require attentional resources (Mooneyham & Schooler, 2013). For example, a handful of studies have shown that mind-wandering was associated with poor response inhibition and increased response variability in tasks that require sustained mental effort (Bastian & Sackur, 2013; Cheyne & Carriere, 2009; Christoff et al., 2009; Kane & McVay, 2012, p. 2; Seli et al., 2014; Smallwood et al., 2004). In the realm of visual attention, specifically, Reichle, Reineberg, and Schooler (2010) showed that during mind-wandering, the usual association between word frequency and fixation duration disappeared, suggesting that the eyes were not actively controlled by online comprehension processes to extract lexical information from the words. Similarly, Zhang et al. (2019) showed that, during mind-wandering, readers were less likely to re-read texts when they encountered information that was inconsistent with their current understanding of the text. Therefore, to the extent that the oculomotor suppression of the distractor requires attentional resources, the effect might disappear during mind-wandering.

In contrast, automatic cognitive processing remains largely intact during mind-wandering (Mooneyham & Schooler, 2013). Mind-wandering appears to occur more frequently in easy than difficult tasks (Seli, 2018). During mind-wandering, participants often display automatic response tendencies, such as failing to withhold rare target responses in go/no-go tasks (McVay & Kane, 2009). Moreover, in tasks that require implicit learning, mind-wandering has been shown to have minimal impact on performance (Brosowsky et al., 2021). These results indicate that automatic processing predominately governs behavior during mind-wandering.

Although multiple studies have investigated the relationship between mind-wandering and eye movements (Faber et al., 2020; Krasich et al., 2020; Reichle et al., 2010; Steindorf & Rummel, 2020; Zhang et al., 2020), scant attention has been paid to this relationship in the context of visual search. Visual search often requires executing accurate eye movements, such as avoiding looking at a distractor, to achieve optimal performance. Yet, there is a paucity of research examining how search performance might be affected during mind-wandering. To gain a more complete understanding of the impact of mind-wandering on cognitive performance, it is crucial to investigate visual search patterns during mind-wandering.

More importantly, mind-wandering provides a unique opportunity to examine whether oculomotor suppression of the distractor requires attentional resources. Gaspelin and Luck (2018b, p.83) stated that one of the premises for distractor suppression is that “participants are in a state of good attentional control.” Later, Luck et al. (2020, p.4) further argued that “the capture of attention by salient singleton stimuli can be prevented if the attentional control system is appropriately configured.” Currently, there exists certain ambiguity as to what constitutes “a state of good attentional control” or an “appropriately configured” attentional control system (Zhang et al., 2021). Does such an appropriately configured attentional control system mean that the observer must be ready to exert attentional resources to the task? Mind-wandering represents an attentional state that selectively impairs processes requiring attentional resources while sparing automatic processes. To the extent that the observer must exert attentional resources to suppress initial eye movements to a distractor, the observer should be unable to do so during mind-wandering.

Alternatively, processes underlying the oculomotor suppression effect might be capable of operating automatically. It is important to recognize that the feature-search task induces strong selection history of the target and distractor features. The identity of the target is fixed throughout the entire task, and the distractor, although a color singleton on a given trial, always appears in the same color throughout the task. These aspects of the task give rise to strong implicit learning, where participants can implicitly learn across several trials to look for (and ignore) an item with certain feature values (Awh et al., 2012; Vatterott & Vecera, 2012). Studies have shown that this learning can take as few as 20 trials to acquire (Gaspelin & Luck, 2018b). Moreover, when implicit learning is not possible (either by switching target colors across trials or by cueing a random color on each trial), the presence of a distractor was found to capture attention (e.g., Gaspelin et al., 2019; Wang & Theeuwes, 2018). Because these history-based selection biases are considered to operate in an effortless, automatic manner (Theeuwes, 2018), participants may be able to avoid looking at the distractor even during mind-wandering.

Pre-trial Pupil Sizes as an Alternative Measure of Mental State

Mind-wandering is typically measured by thought probes that occasionally appear during a task. While thought probing is a reasonably valid approach, there has been longstanding interest in identifying objective, non-intrusive ways of assessing participants’ mental state. Pupil size emerges as a potential candidate in this regard.

Initially used in early studies as an indicator of cognitive load (Kahneman & Beatty, 1966), pupil size recently regained attention, bolstered by the adaptive gain theory (Aston-Jones & Cohen, 2005). This theory associates pupil size with activity in the locus coeruleus (LC), which is integral to the synthesis of norepinephrine (NE) and has widespread projections across the neocortex. The LC-NE system is pivotal in modulating arousal, attention, and responses to stress (Benarroch, 2009). According to Aston-Jones and Cohen (2005), there exists an inverted-U relationship between tonic (baseline) LC activity and performance on tasks that require attentional resources. Peak performance is achieved with an intermediate level of tonic LC activity, during which individuals maintain focus on the current task (exploitation mode). Conversely, excessively high levels of tonic LC activity can result in a distractible state (exploration mode), whereas very low levels of tonic LC activity are associated with sleep and drowsiness. Importantly, a correlation between tonic (baseline) pupil size and LC activity was found in both monkeys (Rajkowski et al., 1993) and humans (Gilzenrat et al., 2003), suggesting that baseline pupil size serves as an indicator of the LC-NE system’s activities. Consistent with the adaptive gain theory, previous studies show that performance dips when pupil size is very small or very large (Gilzenrat et al., 2010; Kristjansson et al., 2009; Murphy et al., 2011; van den Brink et al., 2016). To our knowledge, the link between baseline pupil size and the capacity to disregard distractors remains unexplored.

The Current Study

The main objective of the present study was to investigate whether the oculomotor suppression of a color-singleton, as reflected by the position of initial eye movements in the feature-search task, varies depending on the availability of attentional resources at the moment of search. To accomplish this goal, we administered the feature-search task, in which participants searched for a fixed target among heterogeneous items, with a color-singleton appearing on half of the trials. We used two measures to evaluate the observer’s capacity to exert attentional resources. The primary measure involved presenting thought probes after certain trials to obtain subjective reports of mind-wandering. We also measured participants’ baseline pupil size before the start of the search display. In addition to analyzing initial eye movements, we also examined additional eye movement measures, which will be detailed in the Results section.

Method

Participants

We determined the sample size based on a power analysis using the Superpower package (Lakens & Caldwell, 2019). This power analysis was based on the comparison between on-task trials and mind-wandering trials as measured by thought probes. The size of the oculomotor suppression effect in Gaspelin et al. (2017, exp. 2) was dz = 1.62. While there might be a reversed effect (i.e., oculomotor capture) during mind-wandering, a parsimonious estimation would be no effect (dz = 0). Moreover, considering that some analyses were restricted to probed trials, the actual oculomotor suppression effect, if any, might be lower than what is reported in the literature. Therefore, we conducted the power analysis based on a suppression effect with dz = .8 for on-task trials and a null effect (dz = 0) for mind-wandering trials. With alpha = .05 and a hypothesized correlation of .3 among within-subject factors, a sample size of 40 can achieve over 80% power for detecting the interaction term between attention (on-task/mind-wandering) and distractor presence (present/absent), as well as for establishing the suppression effect for on-task trials only.

We recruited 49 undergraduates from the University of Michigan to participate in this study for course credit. Of the 49 participants, we excluded data from seven participants due to incomplete data, one due to “brain fog” from medications, and one due to color blindness. The final sample size consisted of 40 participants (Mean Age: 18.95, SD Age: 1.04, 60% female, 65% Caucasian).

Apparatus

Participants completed the study in a quiet environment with dimmed lighting. The visual search task was presented on a 20.1-inch computer screen at approximately 80 centimeters from the participant. The task was implemented using OpenSesame (Mathôt et al., 2012) with functions from the PyGaze package (Dalmaijer et al., 2014). Eye movements were recorded by an EyeLink 1000 System at a sampling rate of 500 Hz.

Task and Stimuli

Visual Search Task

The visual search task was adapted from Gaspelin et al.’s (2017) experiment 2. A schematic illustration of the task display is shown in Figure 1. The search array consisted of 6 items distributed equally around an imaginary circle with a radius of 4.5° visual angle. The target was always a diamond (0.8°*0.8°) for a random half of the participants and a circle (0.9° radius) for the other half. The target location was randomized on each trial. Non-singleton items consisted of squares, hexagons, and the unselected target item (circle for the diamond group, diamond for the circle group). These non-singleton items were selected randomly on each trial with the constraint that each shape did not appear more than twice. For half of the trials, all items in the search array appeared in either red or green, randomized across participants. For the other half of the trials, one randomly selected non-target item appeared in the opposite color, thus becoming a color-singleton distractor. Inside each item, there was a black line segment (.37°) with either a vertical or horizontal orientation, randomly chosen for each item. The task was to report the orientation of the line segment inside the target item. Participants were explicitly asked to ignore the color singleton because it would never be the target.

Figure 1.

Figure 1

Open in a new tab

An example distractor-absent trial (left) and distractor-present trial (right). In both cases, the target is the diamond. Items are not drawn to scale. In the actual task, the items were farther away from the center and smaller in size to encourage eye movements.

A trial started with a fixation cross at the center of the screen. Participants had to maintain their gaze within a 1.5° radius of the fixation cross for 2000 milliseconds to trigger the search array. Then, the fixation cross disappeared, and the search array appeared and remained visible until response. If the fixation cross failed to trigger a search array within 10 seconds, the search array would automatically appear and the eye tracker was recalibrated at the end of the trial. Participants responded to the orientation of the line inside the target item using the “z” key and the “/” key on a keyboard. After a response was made, the screen remained dark for 600 milliseconds. Participants first completed a practice block of 80 trials, followed by 6 experimental blocks of 80 trials each.

Thought Probes

A thought probe was presented after 10% of the trials at pseudo-random positions in each block. The thought probe was presented after distractor-present and distractor-absent trials equally often. Each probed trial was preceded by at least 3 and at most 20 non-probed trials.

The thought probe asked participants to “select the one that best describes your conscious experience during the previous trial”. Participants could choose either “on-task” by pressing the “up-arrow” key, “unintentional mind-wandering” by pressing the “left-arrow” key, or “intentional mind-wandering” by pressing the “right arrow” key. Intentional mind-wandering was defined as “you intentionally decided to think about things that are unrelated to the task”, while unintentional mind-wandering was defined as “your thoughts drifted away despite your best intentions to focus on the task” (Seli et al., 2015). After responding to the thought probe, another screen appeared to ask the participant to move their hands back to the response keys. Participants then pressed the spacebar to resume the task.

To-do List Activity

This activity was adapted from Kopp, D’Mello, and Mills (2015) and was intended to increase the level of mind-wandering in the subsequent task. Participants were given 5 minutes to make a list of things they planned to do in the next five days. All participants completed this task before the visual search task.

Procedure

After signing the consent form, participants first completed a health and demographics questionnaire designed to obtain their demographic information and medical history. Then, we asked participants to complete the Ishihara color deficiency test (Ishihara, 2010) to screen for color blindness. Next, participants completed the to-do list activity, followed by the visual search task. After the visual search task, participants completed several questionnaires designed to measure attention and engagement. These questionnaires are not analyzed and can be found in the online OSF repository (https://osf.io/sfyr4/). Finally, participants were debriefed and thanked.

Data Analysis

Data Preprocessing

Trials were excluded (1) if they had over 30% of tracking loss (1.61% of trials), (2) if they had a response time over 3 standard deviations from the mean (0.45% of trials), (3) if they had a pre-trial pupil size over 3 standard deviations from the mean (0.11% of trials), (4) if no fixation recorded during the search array (0.16% of trials), or (5) if the fixation cross timed out (i.e., failing to trigger the search array within 10 seconds) during the pre-trial interval (1.01% of trials).

Search RTs were calculated based on correct trials only. Fixations that were greater than 2000 milliseconds or shorter than 80 milliseconds were discarded (5.09% of all fixations). To determine the landing position of fixations during the search array, we defined an annulus-shaped area of interest with an inner radius of 3° and an outer radius of 6°. Fixations inside the annulus were assigned to the closest item. Pupil sizes were recorded as the area of the pupil in an arbitrary unit determined by the Eyelink system. We selected the last 1000 milliseconds of pupil-size samples during the fixation cross. The selected samples went through a series of processing steps, including deblinking, removing outliers, smoothing, and interpolation of missing data. The means of the preprocessed samples were calculated to represent the pre-trial pupil sizes on each trial. Python scripts for these processing steps can be found in the online OSF repository (https://osf.io/sfyr4/).

Analysis of Thought Probe Reports

Mind-wandering reports obtained via thought probes constitute passive observations and led to an unbalanced design in the current study. Participants reported being on-task on 61.92% of trials, followed by unintentional mind-wandering on 28.87% of trials, and intentional mind-wandering on 9.21% of trials. Additionally, three participants reported no instances of mind-wandering in one or both conditions (distractor present/absent). We made two analytic choices regarding these data.

First, we opted to merge unintentional and intentional mind-wandering into a single category. This decision was made because (1) our primary goal was to compare mind-wandering and being on-task, and (2) the low probability of intentional mind-wandering trials could result in low statistical power for any tests pertaining to this condition.

Second, to create a balanced data set, we chose to exclude data from the three participants who did not report mind-wandering from analyses involving thought probe reports. This allowed us to use conventional methods such as repeated-measures ANOVA to analyze the thought-probe data. Nonetheless, their data were included in other analyses that did not rely on thought probe reports, as these analyses involved balanced data.

Bayes Factors (BF01) Calculation

To gauge support for the null hypothesis, we computed Bayes Factors (BF01) using the BayesFactor package (Morey et al., 2023). We used the ttestBF() function for paired-sample t-tests and the anovaBF() function for repeated-measures ANOVAs. Following the default settings of the package, both functions assumed the JZS priors for effects with prior scales set to medium (0.707 for t-tests and 0.5 for ANOVAs). For repeated-measures ANOVAs, we computed BF01 by comparing a model omitting the critical effect against the full model. Following convention, we interpret Bayes Factors (BF01) as follows: 1 < BF <= 3 suggests anecdotal evidence, 3 < BF <= 10 suggests moderate evidence, 10 < BF <= 30 suggests strong evidence, 30 < BF <= 100 suggests very strong evidence, and BF > 100 suggests extreme evidence.

Results

Oculomotor Suppression of the Distractor Across All Trials

We first sought to replicate the oculomotor suppression of the distractor across all trials. As shown in Figure 2A, search RTs were significantly faster when the distractor was present compared to when it was absent, with a mean difference of 19.75 milliseconds. A paired-sample t-test indicates that the difference was statistically significant, t (39) = 2.99, p = .005, dz = .47, BF01 = 1/7.67, which indicates moderate evidence against the null hypothesis.

Figure 2.

Figure 2

Open in a new tab

Search RT and initial eye movement results across all trials. Panel a shows that search RTs on distractor-present trials were faster compared to distractor-absent trials. ** p < .01. Panel b shows that, on distractor-present trials, participants were less likely to have their first eye movements on the color-singleton distractor than on an average non-singleton item. In both panels, error bars show 95% CIs.

The landing position of the participants’ initial fixations, as shown in Figure 2B, shows the early deployment of overt attention. Compared to an average non-singleton item, initial fixations were less likely to land on the color singleton distractor, with a mean difference of 3.85%. A paired-sample t-test indicated a significant difference, t (39) = 3.82, p < .001, dz = .60, BF01 = 1/60.19. This provides very strong evidence against the null hypothesis.

We also examined whether the probability of initial fixations landing on the target was influenced by the distractor’s presence. The probability of initial fixations landing on the target was slightly lower when the distractor was present than when it was absent, with a mean difference of 1.18%. However, this difference was not statistically significant, t (39) = 1.74, p = .090, dz = .27, BF01 = 1.49, suggesting anecdotal evidence in favor of the null hypothesis.

Overall, these results indicate an oculomotor suppression effect across all trials, in which initial eye movements were less frequently directed to the distractor compared to an average non-singleton item.

Distractor Suppression during Mind-wandering

After establishing the oculomotor suppression effect across all trials, we proceeded to examine whether there was a weaker oculomotor suppression effect during mind-wandering.

Search RT

Figure 3A presents the mean search RTs and 95% CIs for each condition. We conducted a repeated-measures ANOVA to examine the effects of probe report (mind-wandering vs. on-task), trial type (distractor present vs. absent), and their interaction term on search RT. The main effect of probe report was significant, F (1, 36) = 10.20, p = .003, partial eta-squared (ηp2) = .221, with an overall slower RT on mind-wandering than on-task trials. The BF01 for this main effect was 1/4.52, suggesting moderate evidence against the null hypothesis. The main effect of trial type was not significant, F (1, 36) = .33, p = .569, ηp2 = .009. The BF01 for this main effect was 4.53, suggesting moderate evidence in favor of the null hypothesis. The interaction between mind-wandering and trial type was also not significant, F (1, 36) = 3.03, p = .090, ηp2 = .078. The BF01 for this interaction effect was 1.49, suggesting anecdotal evidence in favor of the null hypothesis1.

Figure 3.

Figure 3

Open in a new tab

Search RT and initial eye movement results on mind-wandering and on-task trials as assessed by thought probes. Panel a shows search RT results. Search RTs were overall longer on mind-wandering trials compared to on-task trials. ** p < .01. Panel b shows the landing probability of initial fixations on distractor-present trials. The oculomotor suppression effect did not significantly differ depending on whether participants were on-task or mind-wandering. In both panels, error bars show 95% CIs.

Initial Landing Position

Figure 3B presents the landing probabilities of initial fixations on mind-wandering and on-task trials. We conducted a repeated-measures ANOVA to examine the effects of probe report (mind-wandering vs. on-task), item type (distractor vs. an average non-singleton), and their interaction term on the landing probability of initial fixations. Note that this analysis was performed using only distractor-present trials with the initial fixations landing on the color singleton distractor or a non-singleton item.

The main effect of probe report was not significant, F (1, 36) = .44, p = .509, ηp2 = .012. The BF01 for this main effect was 4.81, suggesting moderate evidence in favor of the null hypothesis. There was a significant main effect of item type, F (1, 36) = 13.93, p < .001, ηp2 = .279, with a lower probability of landing on the distractor compared to an average non-singleton item. The BF01 for this main effect was 1/535.20, suggesting extreme evidence against the null hypothesis. The interaction between probe report and item type was not significant, F (1, 36) = .30, p = .590, ηp2 = .008. The BF01 for this interaction effect was 3.74, providing moderate evidence in favor of the null hypothesis. These results demonstrate that participants exhibited suppression of initial eye movements to the distractor regardless of their attentional state as indicated by thought probe reports2.

Oculomotor Suppression as a Function of Pre-trial Pupil Sizes

Next, we examined whether pre-trial pupil sizes predicted search RTs as well as the extent to which initial eye movements to the distractor were suppressed.

Search RT

For each participant, we categorized their mean pre-trial pupil sizes into quartiles. Figure 4A presents the mean search RTs and 95% CIs for each quartile and trial type. We analyzed the search RTs using a repeated-measures ANOVA with trial-type (distractor present vs. absent), pupil size quartiles (1/2/3/4, from smallest to largest), and their interaction term. The results show a main effect of trial type, F (1, 39) = 9.85, p = .003, ηp2 = .202, with overall faster RTs in the distractor-present condition than in the distractor-absent condition. The BF01 for this main effect was 1/7.74, suggesting moderate evidence against the null hypothesis. The main effect of pupil size quartile was not significant, F (3, 117) = 2.25, p = .086, ηp2 = .055. The BF01 for this main effect was 1/1.18, which suggests anecdotal evidence against the null hypothesis. Importantly, the interaction between pupil size quartiles and trial type was also not significant, F (3, 117) = .08, p = .970, ηp2 = .002. The BF01 for this interaction effect was 29.59, providing strong evidence in favor of the null hypothesis. Overall, these results indicate that search RTs in general were faster when the distractor was present than when it was absent, and this effect did not significantly vary by pre-trial pupil size quartiles.

Figure 4.

Figure 4

Open in a new tab

Search RT and initial eye movement results across quartiles of mean pre-trial pupil size. Panel a shows search RT results. On average, search RTs on distractor-present trials were faster compared to distractor-absent trials across all quartiles. ** p < .01. Panel b shows the landing probability of initial saccades on distractor-present trials. On average, initial eye movements were less likely to land on the color-singleton distractor than an average non-singleton item across all pupil size quartiles. In both panels, error bars show 95% CIs.

Initial Landing Position

Figure 4B presents the landing probabilities of initial fixations on each item for each pupil size quartile. To examine if oculomotor suppression of initial eye movements to the distractor varied by pupil size quartiles, we performed a repeated-measures ANOVA with item type (distractor vs. an average non-singleton), pupil size quartile (1/2/3/4), and their interaction term. Again, this analysis was performed using only distractor-present trials with initial fixations landing on the color-singleton distractor or a non-singleton item. There was a significant main effect of item type, F (1, 39) = 14.34, p < .001, ηp2 = .269, with a lower probability of landing on the distractor compared to an average non-singleton item. The BF01 for this main effect was smaller than 1/100, indicating extreme evidence against the null hypothesis. The main effect of pupil size quartiles was not significant, F (3, 117) = 2.20, p = .092, ηp2 = .053. The BF01 for this main effect was 29.24, providing strong evidence in favor of the null hypothesis. Importantly, the interaction between item type and pupil size quartiles was also not significant, F (3, 117) = 1.23, p = .303, ηp2 = .031. The BF01 for this interaction effect was 15.90, indicating strong evidence in favor of the null hypothesis. Overall, these results show that oculomotor suppression of initial eye movements to the distractor did not significantly vary by pre-trial pupil size quartiles3.

Distinguishing Early and Late Effects of Mind-wandering on Visual Search

The results obtained thus far indicate that the suppression of initial eye movements to the distractor remained robust, irrespective of the participant’s attentional state. However, it is important to recognize that the initial landing position reflects only the earliest phase in the deployment of visual attention. It is possible that mind-wandering may disrupt later phases of the visual search process. To test this possibility, we dissected the time course of a trial into two distinct phases: the time from trial onset to when a fixation first landed on the target (time to target) and the time from this initial target fixation to the execution of a response (time from target landing to response).

Time to Target

The time taken to locate the target would be delayed if participants inspected other items, such as the distractor, before landing on the target. Figure 5A presents the means and 95% CIs of time-to-target for distractor-present and distractor-absent trials when participants were on-task and mind-wandering.

Figure 5.

Figure 5

Open in a new tab

The impact of mind-wandering on early and late stages of visual search. Panel A shows the time from when the search array appeared to when the target item was first fixated. Panel B shows the time from when the target was first fixated to when a manual response was made. Panel C shows the total time participants spent examining the target. Panel D shows the same target-looking times on distractor-present trials as a function of target-to-distractor distance. In all panels, error bars show 95% CIs. * p < .05, ** p < .01.

We used a repeated-measures ANOVA to examine the effects of probe report (mind-wandering vs. on-task), trial type (distractor present vs. absent), and their interaction term on the time to land on the target. There was no significant main effect of probe report, F (1, 36) = .58, p = .452, ηp2 = .016. The BF01 for this main effect was 4.54, suggesting moderate evidence in favor of the null hypothesis. There was also no significant main effect of trial type, F (1, 36) = 2.13, p = .153, ηp2 = .056. The BF01 for this main effect was 1.88, which suggests anecdotal evidence in favor of the null hypothesis. The interaction between probe report and trial type did not reach significance, F (1, 36) = 1.77, p = .192, ηp2 = .047. The BF01 for this main effect was 2.08, suggesting anecdotal evidence in favor of the null hypothesis. These results do not show clear effects in which the time participants took to locate the target was interfered by mind-wandering or by the presence of a distractor4.

Time from Target Landing to Response

The time from target landing to response would be delayed if participants spent extra time inspecting search items (including but not exclusive to the target) even after the target had been located, before making a response. Figure 5B presents the means and 95% CIs for distractor-present and distractor-absent trials when participants were on-task and mind-wandering.

We used a repeated-measures ANOVA to examine the effects of probe report (mind-wandering vs. on-task), trial type (distractor present vs. absent), and their interaction term on the time from target landing to response. We found a significant main effect of probe-report, F (1, 36) = 12.53, p = .001, ηp2 = .258, with an overall longer time on mind-wandering trials compared to on-task trials. The BF01 for this main effect was 1/22.64, providing strong evidence against the null hypothesis. The main effect of trial type was not significant, F (1, 36) = .28, p = .601, ηp2 = .008. The BF01 for this main effect was 5.18, suggesting moderate evidence in favor of the null hypothesis. Importantly, we also found a significant interaction between probe report and trial type, F (1, 36) = 8.07, p = .007, ηp2 = .183, indicating that the distractor’s interference depended on whether participants were mind-wandering or not. The BF01 for this interaction effect was 1/5.92, indicating moderate evidence against the null hypothesis.

Comparisons of estimated marginal means show that, during mind-wandering, the time from target landing to response was significantly longer on distractor-present trials compared to distractor-absent trials, with a mean difference of 45.10 milliseconds, t (71) = 2.29, p = .025, dz = .31. However, when participants were on-task, the time from target landing to response was slightly but not significantly faster on distractor-present trials with a mean difference of 29.60 milliseconds, t (71) = 1.50, p = .138, dz = .33.

Item Looking Times After Target Landing

To further pinpoint the source of the slowdown after locating the target, we examined the looking times on the target, the distractor, and non-singleton items since when the target was first inspected. These measures reflect how much time participants inspected each item after the target had been located but before a response was made.

For looking time on the target, this measure was calculated by summing up the duration of all target-landing fixations on a given trial. Figure 5C presents the mean and 95% CI of target-looking times in each condition. We used a repeated-measures ANOVA to examine the effects of probe report (mind-wandering vs. on-task), trial type (distractor present vs. absent), and their interaction term on target looking time. The main effect of probe report was significant, F (1, 36) = 10.93, p = .002, ηp2 = .233, with an overall longer target looking time on mind-wandering trials compared to on-task trials. The BF01 for this main effect was 1/27.25, providing strong evidence against the null hypothesis. The main effect of trial type was not significant, F (1, 36) = 1.58, p = .217, ηp2 = .042. The BF01 for this main effect was 2.40, suggesting anecdotal evidence in favor of the null hypothesis. Importantly, there was a significant interaction between probe report and trial type, F (1, 36) = 7.06, p = .012, ηp2 = .164, indicating that the distractor’s interference on target looking time depended on whether participants were mind-wandering or not. The BF01 for this interaction effect was 1/4.34, indicating moderate evidence against the null hypothesis.

Further comparisons of the estimated marginal means show that, during mind-wandering, target looking time was significantly longer on distractor-present trials than on distractor-absent trials, with a mean difference of 46.00 milliseconds, t (72) = 2.73, p = .008, dz = .37. However, when participants were on-task, target looking time was slightly but not significantly shorter on distractor-present trials than on distractor-absent trials with a mean difference of 14.90 milliseconds, t (72) = 0.89, p = .378, dz = .21.

The looking time on the distractor was calculated by summing up the duration of distractor-landing fixations that occurred after the first target-landing fixation. A repeated-measures ANOVA found that the effect of probe report was not significant, F (1, 36) = .53, p = .473, ηp2 = .014. The BF01 for this effect was 3.21, providing moderate evidence in favor of the null hypothesis. Thus, mind-wandering did not increase distractor-looking time when the target had been located.

The looking time on non-singleton items was calculated by summing up the duration of fixations that landed on non-singleton items after the first target-landing fixation. We performed a repeated-measures ANOVA to examine the effects of probe report (mind-wandering vs. on-task), trial type (distractor present vs. absent), and their interaction term on this measure. The main effect of probe report was not significant, F (1, 36) = .39, p = .539, ηp2 = .011. The BF01 for this main effect 5.18, providing strong evidence in favor of the null hypothesis. The main effect of trial type was not significant, F (1, 36) = .79, p = .381, ηp2 = .021. The BF01 for this main effect was 4.70, indicating moderate evidence in favor of the null hypothesis. The interaction between probe report and trial type was also not significant, F (1, 36) = .37, p = .545, ηp2 = .010. The BF01 for this interaction effect was 3.23, suggesting moderate evidence in favor of the null hypothesis.

In sum, these results indicate that the presence of the distractor prolonged target looking time when participants were mind-wandering but not when they were on-task.

Target Looking Time by Distance to the Distractor

The fact that target-looking times during mind-wandering only increased on distractor-present trials suggests that target processing during mind-wandering was affected by the distractor. To further examine this possibility, we tested whether target-looking times varied as a function of the relative distance between the target and the distractor. As there were a total of six equally distributed items on each trial, there were three possible relative distances between the target and the distractor: close (1 item away), medium (2 items away), and far (3 items away). Figure 5D shows the average target-looking times and their 95% CIs as a function of the relative distance to the distractor and whether participants reported mind-wandering.

We conducted a repeated-measures ANOVA with probe report (mind-wandering/on-task), target-distractor distance (close, medium, and far), and their interaction term. Note that ten participants did not report mind-wandering at all target-distractor distances, resulting in an unbalanced design. For simplicity, we removed data from these participants, leaving 27 participants for this analysis. There was a significant main effect of probe report, F (1, 26) = 10.83, p = .003, ηp2 = .294, with an overall longer target looking time on mind-wandering trials compared to on-task trials across all target-distractor distances. The BF01 for this main effect was 1/19.92, providing strong evidence against the null hypothesis. Thus, results within this subset of participants are consistent with previous results by showing a longer target looking time on distractor-present trials during mind-wandering. The main effect of target-distractor distance was not significant, F (2, 52) = 1.65, p = .201, ηp2 = .060. The BF01 for this main effect was 4.89, indicating moderate evidence in favor of the null hypothesis. The interaction between probe-report and target-distractor distance was also not significant, F (2, 52) = .16, p = .854, ηp2 = .006. The BF01 for this main effect was 7.81, indicating moderate evidence in favor of the null hypothesis5.

Discussion

The present study examined whether oculomotor suppression of a color singleton requires attentional resources. Participants performed a feature-search task that has been specifically designed to elicit strong oculomotor suppression of the distractor. Mind-wandering reports, obtained from thought probes following a subset of trials, served as the primary measure for assessing the availability of attentional resources on those trials. The results show an intriguing pattern: During mind-wandering, initial eye movements toward the distractor were suppressed but the presence of the distractor prolonged the time from target localization to response generation, particularly target looking time, compared to when participants were on-task.

Suppression of Initial Eye Movements to the Distractor

Analyses based on thought probe responses and pre-trial pupil sizes consistently demonstrated that the oculomotor suppression of the distractor, as indicated by initial eye movements, remained effective even when the availability of attentional resources to the task was low. These results help clarify what constitutes an “appropriately configured” attentional state that enables the oculomotor suppression effect in the feature-search task (Luck et al., 2020). As far as initial eye movements are concerned, our results indicate that oculomotor suppression of the distractor in the current task does not require attentional resources.

The current results are consistent with a growing body of research highlighting implicit learning as a key factor in distractor suppression (Gaspelin et al., 2019; Lien et al., 2023; Theeuwes, 2018; Vatterott & Vecera, 2012). According to Luck et al. (2020), implicit learning can proactively modulate the attentional priority map via feature gain control mechanisms at the early stages of visual processing. This priority map, in turn, guides the allocation of visual attention. The fact that the oculomotor capture effect was still observed during mind-wandering suggests that these implicit-learning-induced feature gain control mechanisms are preserved. In fact, it is perhaps this strong implicit learning that led to mind-wandering in the first place – the learned distractor suppression potentially frees up attentional resources, thereby allowing participants to think about something else.

One might wonder what exactly was implicitly learned in this task that led to the oculomotor suppression effect. Here, it is important to reiterate that the feature-search task induces strong selection history of both target and distractor, as participants searched for a constant target shape while ignoring a distractor in a fixed color. Therefore, it is quite possible that there was implicit learning of both target and distractor features, leading to the up-weighting of target feature values and the down-weighting of distractor feature values (Chang & Egeth, 2019). If this is indeed the case, the current results imply that both forms of learning can operate automatically and are preserved during mind-wandering.

The current results are consistent with those of a recent study (Lien et al., 2023) employing a different paradigm that shows the automatic nature of implicitly learned distractor suppression. These authors combined an RSVP task with a capture-probe paradigm. For the RSVP task, participants detected the presence of a target digit in a sequence of digits. The search display appeared at a lag of 2 or 8 digits after the RSVP target, in order to manipulate the attentional resources available at the time of the search. Their search task also induced strong selection history, with a constant target item (a yellow circle) and a distractor color randomly chosen from two colors. On 30% of trials, probe letters appeared inside the search items, and participants were asked to recall the letters. They found that probe recall accuracy was lower at locations with distractor colors compared to neutral colors, and critically, this effect was similar at Lag 2 and Lag 8. The authors similarly concluded that implicitly learned distractor suppression does not require attentional resources.

What happens when implicit learning is not possible, such as when the target and/or the distractor colors randomly vary on each trial? In such scenarios, observers may need to reactively disengage attention from the distractor following initial capture (Geng, 2014; Theeuwes, 2018). A rapid disengagement of attention from the distractor may demand attentional resources, which may be compromised when the observer is mind-wandering. In this regard, what constitutes an “appropriately configured attentional state” is likely to vary task by task.

Pre-trial Pupil Size

The use of pupil size as a supplemental measure to thought probes was motivated by the fact that pupil size covaries with activity in the LC, a region that plays a crucial role in regulating the balance between on- and off-task focus. It is important to note, however, that in our study we only measured pre-trial (tonic) pupil size. Previous research on the relationship between task-evoked pupillary responses (phasic pupil size) has suggested that smaller task-evoked pupillary responses are associated with lower task performance and self-reported mind-wandering (Jubera-García et al., 2019; Mittner et al., 2014; Unsworth & Robison, 2016, 2018). Our task required participants to make rapid and overt eye movements away from the screen center, which may have occluded any effects on pupil dilation (although see Hayes & Petrov (2016) for a way to correct for this). The relationship between the oculomotor suppression effect and task-evoked pupil size remains an intriguing topic for future research.

It is also noteworthy that there might be other objective indicators of participants’ attentional state that can modulate the oculomotor capture effect. For example, Zhang et al. (2021) reported that pre-trial gaze variability — the spatial dispersion of gaze during the pre-trial interval — positively predicted the probability of initial fixations landing on the distractor. However, it is currently ambiguous as to what this variability measure is measuring even if it does modulate the oculomotor capture effect (Theeuwes, 2021).

A Late Distractor Interference During Mind-wandering

Our discussion thus far has centered solely on initial eye movements, which only reflect the earliest moment of visual search. During mind-wandering, the presence of the distractor prolonged the time from target localization to response generation, particularly the time to inspect the target, compared to when participants were on-task.

The relatively late interference effect observed in our study raises questions about its underlying causes. First, this effect was identified exclusively in the presence of the distractor (refer to Figures 5B and 5C), implying that it is not attributable to a general slowdown in target processing, such as difficulties in retrieving the correct response associated with the line orientation. If that were the reason, we would expect a general increase in target inspection time on both distractor-present and distractor-absent trials during mind-wandering. This did not happen. Second, the observed impact of mind-wandering on these measures suggests that the cognitive processes involved do require attentional resources. Here, we propose two potential explanations.

First, the effect may reflect a filtering cost, namely a delay in responding to the target caused by the simultaneous presentation of distractors, even when these distractors do not capture attention (Folk & Remington, 1998). For example, Kahneman et al. (1983) demonstrated that word reading speed is reduced by the presence of irrelevant dot patches, despite these dot patches not sharing any critical features with the target words. Becker (2007) further showed that when both the target and distractor features are known in advance, the distractor incurs filtering costs in the absence of attentional capture. Thus, perhaps the color-singleton in the current task also produced a filtering cost even without producing an overt capture, and this cost was more pronounced during mind-wandering. Indeed, we did not find that mind-wandering increased distractor-looking times after the target had been found, indicating no overt shift of attention. However, it is noteworthy that filtering costs are typically assumed to be spatially unrelated, whereas, in our study, there was some limited evidence suggesting that the distractor’s interference was pronounced when it was close to the target (see Footnote 5). However, since we could not establish a significant interaction effect due to a smaller sample size in this analysis, this remains a question for future research.

Another possibility is that there was covert capture by the distractor, even though the eyes stayed on the target. Focus of attention can be independent of eye position (Posner, 1980). Despite a strong automatic tendency to direct initial eye movements toward the target and away from the distractor, covert attention may inadvertently drift to the distractor during mind-wandering. Recovering from these unintended attentional drifts would slow down the response time post-target identification. A recent study employing a feature-search task dissected a trial’s timeline similar to our approach (Hamblin-Frohman et al., 2022). They found that the overall benefit of distractor presence on search RTs was primarily (~65%) due to early attention-guiding processes (“target localization”) and, to a lesser extent (~30%), due to the facilitation of later processes (“decision speed”). Therefore, the attentional advantage resulting from strong history-driven bias in the feature-search task is rapid but diminishes over time, potentially allowing for accidental covert attention shifts in later processing stages. Distinguishing between the covert capture account and the filtering cost account may hinge on evidence indicating that covert attention shifted to the distractor’s location even when the eyes were fixated on the target.

Implications for Attentional Capture Research

The landing position of initial eye movements is one of the most popular criteria for assessing distractor interference. This approach is justified in that the landing position of initial saccades provides a clear measure of which item in the display is prioritized. However, the task extends beyond the first eye movement; the observer often needs to identify and process the target item to formulate an appropriate manual response. These subsequent processes have been relatively underexplored compared to the extensive research on initial landing positions (see also Zivony, 2021). Our results suggest that even when the distractor does not attract initial eye movements, it can still interfere with later stages of processing under certain conditions. Thus, the predominant focus on initial eye movements may provide a simplified and incomplete picture of how a distractor impacts visual search over the entire course of a trial.

Acknowledgments

This work was supported by the National Science Foundation [grant number: 1658268] awarded to the University of Michigan with JJ as Principal Investigator and the National Institute of Mental Health (Unique Federal Award Identification Number (FAIN): R21MH129909) awarded to the University of Michigan with JJ as Principal Investigator.

Footnotes

1

Despite a non-significant interaction effect, we conducted a marginal means analysis to explore the pattern of the means. As shown in Figure 3A, when participants were on-task, search RTs on distractor-present trials were slightly but not significantly faster compared to distractor-absent trials, with a mean difference of 42.3 milliseconds, t (65) = 1.48, p = .144, dz = .29. When participants were mind-wandering, search RTs were slightly but not significantly slower on distractor-present trials compared to distractor-absent trials, with a mean difference of 15.6 milliseconds, t (65) = 0.55, p = .587, dz = .08.

2

A significant oculomotor suppression effect was observed for on-task trials, with a mean difference of 5.87%, t (71) = 3.15, p = .002, dz = .62, as well as for mind-wandering trials, with a mean difference of 4.53%, t (71) = 2.43, p = .018, dz = .35.

3

A significant oculomotor suppression effect was observed for each pupil size quartile (1st quartile: t (64) = 3.20, p = .002, dz = .49; 2nd quartile: t (64) = 3.46, p = .001, dz = .61; 3rd quartile: t (64) = 4.06, p < .001, dz = .70; 4th quartile: t (64) = 2.58, p = .012, dz = .35).

4

During mind-wandering, participants were slightly (but not significantly) faster to locate the target on distractor-present trials compared to distractor-absent trials with a mean difference of 36.24 ms, t (71) = 1.97, p = .052, dz = .33. This distractor-presence benefit was reduced to 4.16 ms for on-task trials, t (71) = 0.23, p = .821, dz = .04. Thus, if anything, the distractor interfered less with target localization during mind-wandering, which is consistent with the idea that the early deployment of overt attention in this task is highly automatic.

5

Despite a non-significant interaction effect, we explored the pattern of the marginal means. When the target and the distractor were close, target-looking times were significantly longer on mind-wandering trials compared to on-task trials, with a mean difference of 85.2 milliseconds, t (77) = 2.22, p = .029. When the target and the distractor were either at a medium distance or at a far distance, target-looking times did not significantly differ between mind-wandering and on-task trials (mean difference at medium distance: 55.1 milliseconds, t (77) = 1.44, p = .156; mean difference at far distance: 61.6 milliseconds, t (77) = 1.60, p = .113). Thus, the effect appears to be numerically the strongest when the distractor was close to the target. However, these effects did not significantly different from one another given the non-significant interaction term, presumably due to the relatively small sample size in this analysis.

All data, code, and tasks are available at https://osf.io/sfyr4/. This study was not preregistered. The authors have no conflicts of interest to report.

Contributor Information

Han Zhang, Department of Psychology, University of Michigan.

Kevin F. Miller, School of Education and Department of Psychology, University of Michigan

John Jonides, Department of Psychology, University of Michigan.

References

  1. Adams OJ, Ruthruff E, & Gaspelin N (2022). Oculomotor suppression of abrupt onsets versus color singletons. Attention, Perception, & Psychophysics. 10.3758/s13414-022-02524-0 [DOI] [Google Scholar]
  2. Anderson BA (2021). Time to stop calling it attentional “capture” and embrace a mechanistic understanding of attentional priority. Visual Cognition, 29(9), 537–540. 10.1080/13506285.2021.1892894 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aston-Jones G, & Cohen JD (2005). An integrative theory of locus coeruleus-norepinephrine function: Adaptive gain and optimal performance. Annual Review of Neuroscience, 28(1), 403–450. 10.1146/annurev.neuro.28.061604.135709 [DOI] [Google Scholar]
  4. Awh E, Belopolsky AV, & 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]
  5. Bastian M, & Sackur J (2013). Mind wandering at the fingertips: Automatic parsing of subjective states based on response time variability. Frontiers in Psychology, 4, 573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Becker SI (2007). Irrelevant singletons in pop-out search: Attentional capture or filtering costs? Journal of Experimental Psychology: Human Perception and Performance, 33(4), 764–787. 10.1037/0096-1523.33.4.764 [DOI] [PubMed] [Google Scholar]
  7. Benarroch EE (2009). The locus ceruleus norepinephrine system: Functional organization and potential clinical significance. Neurology, 73(20), 1699–1704. 10.1212/WNL.0b013e3181c2937c [DOI] [PubMed] [Google Scholar]
  8. Brosowsky NP, Murray S, Schooler JW, & Seli P (2021). Attention need not always apply: Mind wandering impedes explicit but not implicit sequence learning. Cognition, 209, 104530. 10.1016/j.cognition.2020.104530 [DOI] [PubMed] [Google Scholar]
  9. Chang S, & Egeth HE (2019). Enhancement and Suppression Flexibly Guide Attention. Psychological Science, 30(12), 1724–1732. 10.1177/0956797619878813 [DOI] [PubMed] [Google Scholar]
  10. Cheyne JA, & Carriere JSA (2009). Anatomy of an error: A bidirectional state model of task engagement/disengagement and attention-related errors. Cognition, 111(1), 98–113. 10.1016/j.cognition.2008.12.009 [DOI] [PubMed] [Google Scholar]
  11. Christoff K, Gordon AM, Smallwood J, Smith R, & Schooler JW (2009). Experience sampling during fMRI reveals default network and executive system contributions to mind wandering. Proceedings of the National Academy of Sciences, 106(21), 8719–8724. 10.1073/pnas.0900234106 [DOI] [Google Scholar]
  12. Dalmaijer ES, Mathôt S, & der Stigchel S (2014). PyGaze: An open-source, cross-platform toolbox for minimal-effort programming of eyetracking experiments. Behavior Research Methods, 46(4), 913–921. [DOI] [PubMed] [Google Scholar]
  13. Faber M, Krasich K, Bixler R, Brockmole J, & D’Mello S (2020). The eye-mind wandering link: Identifying gaze indices of mind wandering across tasks. Journal of Experimental Psychology: Human Perception and Performance. [Google Scholar]
  14. Folk CL, & Remington R (1998). Selectivity in Distraction by Irrelevant Featural Singletons: Evidence for Two Forms of Attentional Capture. Journal of Experimental Psychology: Learning Memory and Cognition, 24(3), 847–858. [Google Scholar]
  15. Gao Y, & Theeuwes J (2020). Learning to suppress a distractor is not affected by working memory load. Psychonomic Bulletin & Review, 27(1), 96–104. 10.3758/s13423-019-01679-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gaspelin N, Gaspar JM, & Luck SJ (2019). Oculomotor inhibition of salient distractors: Voluntary inhibition cannot override selection history. Visual Cognition, 27(3–4), 227–246. 10.1080/13506285.2019.1600090 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gaspelin N, Leonard CJ, & Luck SJ (2017). Suppression of overt attentional capture by salient-but-irrelevant color singletons. Attention, Perception, & Psychophysics, 79(1), 45–62. 10.3758/s13414-016-1209-1 [DOI] [Google Scholar]
  18. Gaspelin N, & Luck SJ (2018a). Combined Electrophysiological and Behavioral Evidence for the Suppression of Salient Distractors. Journal of Cognitive Neuroscience, 30(9), 1265–1280. 10.1162/jocn_a_01279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gaspelin N, & Luck SJ (2018b). Distinguishing among potential mechanisms of singleton suppression. Journal of Experimental Psychology: Human Perception and Performance, 44(4), 626–644. 10.1037/xhp0000484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Geng JJ (2014). Attentional Mechanisms of Distractor Suppression. Current Directions in Psychological Science, 23(2), 147–153. 10.1177/0963721414525780 [DOI] [Google Scholar]
  21. Gilzenrat MS, Cohen JD, Rajkowski J, & Aston-Jones G (2003). Pupil dynamics predict changes in task engagement mediated by locus coeruleus. Society for Neuroscience Abstracts, 515, 19. [Google Scholar]
  22. Gilzenrat MS, Nieuwenhuis S, & Cohen JD (2010). Pupil diameter tracks changes in control state predicted by the adaptive gain theory of locus coeruleus function. Cognitive, Affective, & Behavioral Neuroscience, 10(2), 252–269. 10.3758/CABN.10.2.252 [DOI] [Google Scholar]
  23. Hamblin-Frohman Z, Chang S, Egeth H, & Becker SI (2022). Eye movements reveal the contributions of early and late processes of enhancement and suppression to the guidance of visual search. Attention, Perception, & Psychophysics, 84(6), 1913–1924. 10.3758/s13414-022-02536-w [DOI] [Google Scholar]
  24. Hayes TR, & Petrov AA (2016). Mapping and correcting the influence of gaze position on pupil size measurements. Behavior Research Methods, 48(2), 510–527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ishihara. (2010). Ishihara’s Tests for Colour Deficiency 2010: 14 Plate Book Concise Edition (1st ed.). Graham-Field. https://www.amazon.com/Ishihara-Chart-Books-Color-Deficiency/dp/B000FLTPZU [Google Scholar]
  26. Jubera-García E, Gevers W, & Van Opstal F (2019). Influence of content and intensity of thought on behavioral and pupil changes during active mind-wandering, off-focus, and on-task states. Attention, Perception, & Psychophysics. 10.3758/s13414-019-01865-7 [DOI] [Google Scholar]
  27. Kahneman D, & Beatty J (1966). Pupil diameter and load on memory. Science, 154(3756), 1583–1585. [DOI] [PubMed] [Google Scholar]
  28. Kahneman D, Treisman A, & Burkell J (1983). The cost of visual filtering. Journal of Experimental Psychology: Human Perception and Performance, 9(4), 510. [DOI] [PubMed] [Google Scholar]
  29. Kane MJ, Brown LH, McVay JC, Silvia PJ, Myin-Germeys I, & Kwapil TR (2007). For whom the mind wanders, and when: An experience-sampling study of working memory and executive control in daily life. Psychological Science, 18(7), 614–621. [DOI] [PubMed] [Google Scholar]
  30. Kane MJ, & McVay JC (2012). What Mind Wandering Reveals About Executive-Control Abilities and Failures. Current Directions in Psychological Science, 21(5), 348–354. 10.1177/0963721412454875 [DOI] [Google Scholar]
  31. Kopp K, D’Mello S, & Mills C (2015). Influencing the occurrence of mind wandering while reading. Consciousness and Cognition, 34, 52–62. 10.1016/j.concog.2015.03.003 [DOI] [PubMed] [Google Scholar]
  32. Krasich K, Huffman G, Faber M, & Brockmole JR (2020). Where the eyes wander: The relationship between mind wandering and fixation allocation to visually salient and semantically informative static scene content. Journal of Vision, 20(9), 10. 10.1167/jov.20.9.10 [DOI] [Google Scholar]
  33. Kristjansson SD, Stern JA, Brown TB, & Rohrbaugh JW (2009). Detecting phasic lapses in alertness using pupillometric measures. Applied Ergonomics, 40(6), 978–986. 10.1016/j.apergo.2009.04.007 [DOI] [PubMed] [Google Scholar]
  34. Lakens D, & Caldwell AR (2019). Simulation-Based Power-Analysis for Factorial ANOVA Designs. 10.31234/osf.io/baxsf [DOI]
  35. Lamy D (2021). The attentional capture debate: The long-lasting consequences of a misnomer. Visual Cognition, 29(9), 544–547. 10.1080/13506285.2021.1904076 [DOI] [Google Scholar]
  36. Leonard CJ (2021). Consensus emerges and biased competition wins: A commentary on Luck et al. (2021). Visual Cognition, 29(9), 560–562. 10.1080/13506285.2021.1912236 [DOI] [Google Scholar]
  37. Lien M-C, Ruthruff E, & Tolomeo D (2023). Evidence that proactive distractor suppression does not require attentional resources. Psychonomic Bulletin & Review. 10.3758/s13423-023-02422-y [DOI] [Google Scholar]
  38. Luck SJ, Gaspelin N, Folk CL, Remington RW, & Theeuwes J (2020). Progress toward resolving the attentional capture debate. Visual Cognition, 1–21. 10.1080/13506285.2020.1848949 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mathôt S, Schreij D, & Theeuwes J (2012). OpenSesame: An open-source, graphical experiment builder for the social sciences. Behavior Research Methods, 44(2), 314–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. McVay JC, & Kane MJ (2009). Conducting the train of thought: Working memory capacity, goal neglect, and mind wandering in an executive-control task. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(1), 196–204. 10.1037/a0014104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Mittner M, Boekel W, Tucker AM, Turner BM, Heathcote A, & Forstmann BU (2014). When the Brain Takes a Break: A Model-Based Analysis of Mind Wandering. The Journal of Neuroscience, 34(49), 16286–16295. 10.1523/JNEUROSCI.2062-14.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Mooneyham BW, & Schooler JW (2013). The costs and benefits of mind-wandering: A review. Canadian Journal of Experimental Psychology/Revue Canadienne de Psychologie Expérimentale, 67(1), 11–18. 10.1037/a0031569 [DOI] [PubMed] [Google Scholar]
  43. Morey RD, Rouder JN, Jamil T, Urbanek S, Forner K, & Ly A (2023). BayesFactor: Computation of Bayes Factors for Common Designs (0.9.12–4.6) [Computer software]. https://cran.r-project.org/web/packages/BayesFactor/index.html
  44. Murphy PR, Robertson IH, Balsters JH, & O’connell RG (2011). Pupillometry and P3 index the locus coeruleus–noradrenergic arousal function in humans. Psychophysiology, 48(11), 1532–1543. 10.1111/j.1469-8986.2011.01226.x [DOI] [PubMed] [Google Scholar]
  45. Posner MI (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32(1), 3–25. [DOI] [PubMed] [Google Scholar]
  46. Rajkowski J, Kubiak P, & Aston-Jones G (1993). Correlations between locus coeruleus (LC) neural activity, pupil diameter and behavior in monkey support a role of LC in attention. Society for Neuroscience Abstracts, 19, 974. [Google Scholar]
  47. Reichle ED, Reineberg AE, & Schooler JW (2010). Eye movements during mindless reading. Psychological Science, 21(9), 1300–1310. [DOI] [PubMed] [Google Scholar]
  48. Risko EF, Buchanan D, Medimorec S, & Kingstone A (2013). Everyday attention: Mind wandering and computer use during lectures. Computers & Education, 68, 275–283. 10.1016/j.compedu.2013.05.001 [DOI] [Google Scholar]
  49. Seli P (2018). The role of task difficulty in theoretical accounts of mind wandering. Consciousness and Cognition, 8. [Google Scholar]
  50. Seli P, Carriere JSA, Thomson DR, Cheyne JA, Martens KAE, & Smilek D (2014). Restless mind, restless body. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(3), 660–668. 10.1037/a0035260 [DOI] [PubMed] [Google Scholar]
  51. Seli P, Cheyne JA, Xu M, Purdon C, & Smilek D (2015). Motivation, intentionality, and mind wandering: Implications for assessments of task-unrelated thought. Journal of Experimental Psychology: Learning, Memory, and Cognition, 41(5), 1417. [DOI] [PubMed] [Google Scholar]
  52. Smallwood J, Brown KS, Baird B, Mrazek MD, Franklin MS, & Schooler JW (2012). Insulation for Daydreams: A Role for Tonic Norepinephrine in the Facilitation of Internally Guided Thought. PLoS ONE, 7(4), e33706. 10.1371/journal.pone.0033706 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Smallwood J, Davies JB, Heim D, Finnigan F, Sudberry M, O’Connor R, & Obonsawin M (2004). Subjective experience and the attentional lapse: Task engagement and disengagement during sustained attention. Consciousness and Cognition, 13(4), 657–690. 10.1016/j.concog.2004.06.003 [DOI] [PubMed] [Google Scholar]
  54. Smallwood J, & Schooler JW (2006). The restless mind. Psychological Bulletin, 132(6), 946–958. 10.1037/0033-2909.132.6.946 [DOI] [PubMed] [Google Scholar]
  55. Steindorf L, & Rummel J (2020). Do your eyes give you away? A validation study of eye-movement measures used as indicators for mindless reading. Behavior Research Methods, 52(1), 162–176. [DOI] [PubMed] [Google Scholar]
  56. Stilwell BT, Adams OJ, Egeth HE, & Gaspelin N (2023). The role of salience in the suppression of distracting stimuli. Psychonomic Bulletin & Review. 10.3758/s13423-023-02302-5 [DOI] [Google Scholar]
  57. Stilwell BT, & Gaspelin N (2021). Attentional suppression of highly salient color singletons. Journal of Experimental Psychology: Human Perception and Performance, 47(10), 1313–1328. 10.1037/xhp0000948 [DOI] [PubMed] [Google Scholar]
  58. Theeuwes J (1992). Perceptual selectivity for color and form. Perception & Psychophysics, 51(6), 599–606. 10.3758/BF03211656 [DOI] [PubMed] [Google Scholar]
  59. Theeuwes J (2004). Top-down search strategies cannot override attentional capture. Psychonomic Bulletin & Review, 11(1), 65–70. 10.3758/BF03206462 [DOI] [PubMed] [Google Scholar]
  60. Theeuwes J (2018). Visual Selection: Usually Fast and Automatic; Seldom Slow and Volitional. Journal of Cognition, 1(1), 29. 10.5334/joc.13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Theeuwes J (2021). Response to commentaries to Luck et al. (2021). Progress toward resolving the attentional capture debate. Visual Cognition, 29(9), 637–643. 10.1080/13506285.2021.1978672 [DOI] [Google Scholar]
  62. Theeuwes J, de Vries G-J, & Godijn R (2003). Attentional and oculomotor capture with static singletons. Perception & Psychophysics, 65(5), 735–746. 10.3758/BF03194810 [DOI] [PubMed] [Google Scholar]
  63. Unsworth N, & Robison MK (2016). Pupillary correlates of lapses of sustained attention. Cognitive, Affective, & Behavioral Neuroscience, 16(4), 601–615. 10.3758/s13415-016-0417-4 [DOI] [Google Scholar]
  64. Unsworth N, & Robison MK (2018). Tracking arousal state and mind wandering with pupillometry. Cognitive, Affective, & Behavioral Neuroscience, 18(4), 638–664. 10.3758/s13415-018-0594-4 [DOI] [Google Scholar]
  65. van den Brink RL, Murphy PR, & Nieuwenhuis S (2016). Pupil Diameter Tracks Lapses of Attention. PLOS ONE, 11(10), e0165274. 10.1371/journal.pone.0165274 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Varao-Sousa TL, Smilek D, & Kingstone A (2018). In the lab and in the wild: How distraction and mind wandering affect attention and memory. Cognitive Research: Principles and Implications, 3(1), 42. 10.1186/s41235-018-0137-0 [DOI] [Google Scholar]
  67. Vatterott DB, & Vecera SP (2012). Experience-dependent attentional tuning of distractor rejection. Psychonomic Bulletin & Review, 19(5), 871–878. 10.3758/s13423-012-0280-4 [DOI] [PubMed] [Google Scholar]
  68. Wang B, & Theeuwes J (2018). How to inhibit a distractor location? Statistical learning versus active, top-down suppression. Attention, Perception, & Psychophysics, 80(4), 860–870. 10.3758/s13414-018-1493-z [DOI] [Google Scholar]
  69. Wang B, & Theeuwes J (2020). Salience determines attentional orienting in visual selection. Journal of Experimental Psychology: Human Perception and Performance, 46(10), 1051–1057. 10.1037/xhp0000796 [DOI] [PubMed] [Google Scholar]
  70. Won B-Y, Kosoyan M, & Geng JJ (2019). Evidence for second-order singleton suppression based on probabilistic expectations. Journal of Experimental Psychology: Human Perception and Performance, 45(1), 125–138. 10.1037/xhp0000594 [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Wöstmann M, Störmer VS, Obleser J, Addleman DA, Andersen SK, Gaspelin N, Geng JJ, Luck SJ, Noonan MP, Slagter HA, & Theeuwes J (2022). Ten simple rules to study distractor suppression. Progress in Neurobiology, 213, 102269. 10.1016/j.pneurobio.2022.102269 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Yantis S, & Jonides J (1990). Abrupt visual onsets and selective attention: Voluntary versus automatic allocation. Journal of Experimental Psychology: Human Perception and Performance, 16(1), 121. [DOI] [PubMed] [Google Scholar]
  73. Zhang H, Abagis TR, & Jonides J (2021). The malleability of attentional capture. Visual Cognition, 29(9), 571–574. 10.1080/13506285.2021.1915903 [DOI] [Google Scholar]
  74. Zhang H, Miller KF, Sun X, & Cortina KS (2020). Wandering eyes: Eye movements during mind wandering in video lectures. Applied Cognitive Psychology, 34(2), acp.3632. 10.1002/acp.3632 [DOI] [Google Scholar]
  75. Zhang H, Qu C, Miller KF, & Cortina KS (2019). Missing the joke: Reduced rereading of garden-path jokes during mind-wandering. Journal of Experimental Psychology: Learning, Memory, and Cognition. 10.1037/xlm0000745 [DOI] [Google Scholar]
  76. Zivony A (2021). Beyond guidance: It’s time to focus on the consequences of attentional capture. Visual Cognition, 29(9), 556–559. 10.1080/13506285.2021.1912235 [DOI] [Google Scholar]

RESOURCES