Exploring the role of mind wandering and executive control in the vigilance decrement: A replication and online study using the ANTI-Vea task ------------------------------------------------------------------------ **Background** Sustained attention or vigilance is the ability to perform a task for an extended period of time. When performing a prolonged task, as time-on-task passes, vigilance tends to decrease, a phenomenon scientifically known as the vigilance decrement. The vigilance decrement is typically observed as a progressive decrease in the detection of infrequent but crucial signals and a progressive increase in response slowness and variability (Hancock, 2017). In the past few years, a dissociation has been proposed between two components of vigilance: (a) executive vigilance (EV), which refers to maintaining active control to execute a specific response for detecting critical signals that occur infrequently, and (b) arousal vigilance (AV), which is necessary to maintain optimal levels of activation to quickly and automatically react to environmental stimuli (Luna et al., 2018). Although several theories have been developed to explain the vigilance decrement, there is still an open debate concerning the mechanisms that lead a progressive loss of vigilance. On the one hand, the resource depletion hypothesis holds that attention works through limited resources and that vigilance tasks are difficult to perform, so that when performing prolonged tasks resources are progressively depleted over time and vigilance decreases (Caggiano & Parasuraman, 2004; Warm, Dember, & Hancock, 1996) On the other hand, the mind-wandering hypothesis holds that vigilance tasks are monotonous and boring, causing attentional resources to wander from the task at hand towards task-irrelevant-thoughts, making difficult to maintain attention in the task and therefore resulting in decreased vigilance (Smallwood & Schooler, 2006). To account for the vigilance decrement phenomenon, Thomson et al. (2015) propose the resource-control theory, which is derived from a combination of predictions by both the resource depletion and mind-wandering hypotheses. According to the resource control theory, the amount of attentional resources available is fixed and do not change over time. However, mind wandering (MW) is our default state, and task-irrelevant-thoughts consume attentional resources that should be dedicated to the primary task. To avoid resources being devoted to task-irrelevant-thoughts, Executive control is necessary to direct attentional resources to maintain vigilance on the task at hand and prevent MW. Importantly, it is executive control that decreases over time, therefore attentional resources being progressively deployed to task-irrelevant-thoughts, consequently leading to decreased vigilance. To test some of the predictions of the resource-control theory, Luna et al., (2022) analyzed a larger sample of data with the ANTI-Vea task, which is suited to simultaneously measure executive control and both types of vigilance. The authors found that executive control and EV decreased over time, and they observed an association between decreasing EV and cognitive control. These results provided empirical evidence that supports the predictions of the resource-control theory regarding the decline in cognitive control and its correlation with a decline in vigilance. However, the task used by Luna et al. (2022) did not include a direct measure of mind-wandering, thus making it difficult to test other critical predictions of the resource-control theory. In order to test the predictions of the resources control theory, we conducted a preliminary experiment incorporating measures of mind wandering (MW), i.e., thought probes (TP), in the ANTI-Vea task. We conducted a study in which different amounts of TP were tested between-participants across groups to determine the optimal TP rate. **Preliminary results** Ninety individuals participated in the study, divided into three groups. One group performed the task with only 4 TP (N=30), another with only 8 TP (N=30), and a third with only 12 TP (N=30). Standard analyses (see 'Design' section) were conducted for the ANTI- Vea task (Luna et al., 2021). For RT (see Fig 1) we found the typical effects of alertness, F (1, 82) = 63.92, p = <.001, η2p =.44, [.28, .56], orienting, F (1.85, 151.73) = 112.78, p = <.001, η2p =.58, [.48, .65] and congruency, F (1, 82) = 97.70, p =< .001, η2p = .54, [.40, .65]. We found a significant interaction between group and orienting, F (3.70, 151.73) = 2.74, p =.034, η2p < .06, [.00, .13], but no significant interaction was observed for alertness (F < 1) or congruency (F < 1). We found the expected effects for percentage of error for alertness, F (1, 82) = 37.86, p = <.001, η2p =.32, [.16, .46] and congruency, F (1, 82) = 8.37, p = .005, η2p = .09, [.01, .23] and no effect for orienting (F < 1). As in RT, we only found a significant interaction between group and orienting, F (3.98, 163.02) = 2.91, p =.024, η2p < .07, [.00, .13], but no significant interaction was observed for alertness (F < 1) or congruency (F < 1). ![Figure 1. Mean correct RT (superior graphs) and percentage of errors (inferior graphs) for the ANTI trials: alertness (left), orientation (center), and congruency (right) as a function of the group. Error bars represent 95% confidence intervals of the mean.][1] Additionally, a decrease in vigilance was observed through the task in the EV trials (see Fig. 2) in hits, F (4.39, 360.24) = 6.62, p = <.001, η2p =.07, [.03, .12] and false alarms, F (4.40, 360.78) = 3.37, p =.008, η2p =.04, [.00, .17]. No significant differences were found between the groups in any of the EV measures (F < 1). ![Figure 2. Hits and false alarms (FA) across time on task as a function of the group. Error bars represent 95% confidence intervals. ][2] In the AV trials (see Fig. 3), a decrease in vigilance was observed through the blocks as indicated by an increase in RT, F (2.69, 220.33) = 8.99, p = <.001, η2p =.10, [.04, .15], SD of RT, F (3.87, 317.13) = 16.53, p = <.001, η2p =.17, [.10, .23] and lapses, F (3.56, 291.67) = 8.81, p = <.001, η2p =.17, [.10, .23]. No significant differences were found between the groups in any of the AV measures (F < 1). ![Figure 3. Mean RT (left graph), SD of RT (central graph), and lapses percentage (right graph) across time on task as a function of the group for AV trials. Error bars represent 95% confidence intervals. ][3] Moreover, an increase in the report of MW over time was observed F (2.72, 222.69) = 60.80, p = <.001, η2p =.43, [.35, .48], with no significant differences found between the groups (F < 1) (see Fig 4). ![Figure 4. TP reporting score across time on task as a function of the group. Error bars represent 95% confidence intervals. ][4] These findings suggest that: (a) the vigilance decrement is observed when TP are embedded in the ANTI-Vea, (b) TP embedded in the ANTI-Vea demonstrates that MW increases across time-on-task, and (c) the number of TP does not significantly affect the main effects of the ANTI- Vea task. In the present study, we will aim to replicate these findings with a larger sample size to conduct further analyses, such as correlations between mind wandering measures and vigilance decrement, mind wandering measures and executive control decrement and correlation between vigilance decrement and executive control decrement. **Objectives** The aim of this study is to examine the relation between components of resource control theory, including MW. In this study, we propose to replicate our previous findings in a larger sample size and through an online data collection. To measure MW in the ANTI-Vea task, we will test the same number of TP used in our previous experiment, which showed that a suitable measure of MW can be obtained without affecting the measurement of the vigilance decrement. **Method** **Participants** Participants will be under-graduated students from the University of Granada who will be invited through an institutional email list. During the initial phase, participants will complete online surveys for a chance to win a financial prize through a lottery system. Next, participants will be asked to participate in an online procedure. In this study, given that all measures in the previous study showed good sensitivity to detect all effects of interest with the used sample (30 participants per group), the sample size used in the preliminary study will be doubled for each group in the present study, considering the online data collection procedure. Thus, 180 participants (60 per group) who have completed the online questionnaire will be randomly selected based on the following criteria: being between 18 and 40 years old, having completed all the questionnaires, and having correctly answered the control questions included in the survey to ensure understanding of the items. Signed consent will be obtained from all participants, in accordance with the ethical standards established in the 1964 Declaration of Helsinki (last updated in Seoul, 2008). Participants must have normal or corrected-to-normal vision. They will receive a reward of 10 euros per hour for their participation in the study. The study has been approved by the University of Granada's Ethical Committee (2442/CEIH/2021). **Materials** **The ANTI-Vea task** The online version of the ANTI-Vea task (https://anti-vea.ugr.es/) will be used to measure behavioral measures of executive control and vigilance decrements (Luna et al., 2018). The standard task is composed of six experimental blocks of 80 trials each one, in which three types of subtasks are combined: (a) ANTI (60%, 48 trials per block), in which the participants' task is to respond to the direction of the central arrow in a horizontal string of five arrows to assess the independence and interactions of phasic alertness, orienting, and executive control; (b) EV (20%, 16 trials per block), to measure the executive component of vigilance, in which the target is vertically displaced from the central position of ANTI trials and should be detected while ignoring the direction of the arrow; and (c) AV (20%, 16 trials per block) in which a counter should be stopped as fast as possible as in the Psychomotor Vigilance Test (Lim & Dinges, 2008) to measure the arousal component of vigilance. Each trial has a duration of 4100ms. **Mind wandering measures** The same TP used in our previous studies will be added as independent trials to the online ANTI-Vea task. Participants will have to answer the following question: “Where was your attention just before the appearance of this question?” Participants will respond by moving the cursor on a continuous scale ranging from "completely on-task" (extreme left, coded as -1) to "completely off-task" (extreme right, coded as 1). TP will appear 4, 8 or 12 times per block. TP presentation will be pseudo-randomized, so that at least there will be 5 trials of the ANTI-Vea task of TPs’ interval in every block. **Questionnaires** Participants will complete the following questionnaires. First Step • Attentional Control Scale (ACS) • Five Facet Mindfulness Questionnaire (FFMQ) • Mind Wandering Questionnaire (MWQ) • Barkley Adult ADHD Rating Scale IV (BAARS IV) - Current Symptoms • Barkley Childhood ADHD Rating Scale IV (BAARS IV) - Inattention and Hyperactivity-Impulsivity • Difficulties in Emotion Regulation Scale - Short Form (DERS-SF) • Irrational Procrastination Scale (IPS) • Demographic questions. Second Step • NASA Task Load Index • Dundee Stress State Questionnaire (DSSQ) **Procedure** The larger sample of participants will complete the first stage of questionnaires at home. Then the selected participants (60 per group) for the second phase will be divided into three groups in order to analyze the appropriate number of probes to include in the task to measure changes in MW across time-on-task without affecting the measures of the vigilance decrement. In phase two, participants will complete two questionnaires and the online version of the ANTI-Vea task with TP. **Design** The standard analyses for the ANTI-Vea task (Luna et al., 2021) will be conducted, incorporating the group (4TP, 8TP, or 12TP) as a between-participant factor in all analyses. For the ANTI trials, mixed ANOVAs will be conducted, taking into account the following three within-participant factors: • Phasic alertness (with 2 levels: tone and no tone) • Orienting (with 3 levels: valid cue, invalid cue, and no cue) • Executive control (with 2 levels: congruent and incongruent). Changes in EV will be analyzed through mixed ANOVAs, which takes into account hits (correct identification of vertically displaced targets), false alarms (incorrect identification of non-displaced targets as being vertically displaced), sensitivity, or response bias as dependent variable, and experimental blocks as within-participant factor. AV will be analyzed using mixed ANOVAs, including the mean or standard deviation of the Reaction Time (RT), or lapses rate (i.e., RT ≥ 600 ms) as dependent variable, and blocks as a within-participant factor. To analyze changes in executive control across time-on-task, mixed ANOVAs as in Luna et al. (2022) will be conducted. The dependent variables will include the interference effect for RT, Accuracy (ACC), as well as the inverse efficiency score (Mean RT/% correct). The experimental blocks will be treated as a within-participant factor. Changes in MW will be evaluated using a mixed ANOVA, with the response on the scale as the dependent variable. The experimental blocks will be treated as a within-participant factor. To analyze the correlation between the components of the resource control theory, we will compute, for each participant, a linear slope across blocks of the EV decrement, AV decrement, and cognitive control decrement, and correlate the linear slopes with the linear slopes of MW increment. Additionally, the across blocks correlation for each participant between the average of each block in the EV, AV, and EC measurements and the average of each block in the MW measurement will be analyzed using Pearson correlation, following (Thomson et al., 2014). Finally, following standard analyses of mind-wandering studies (Bastian & Sackur, 2013; Seli et al., 2013; Thomson et al., 2014), the five trials of the ANTI- VEA before each TP will be analyzed to compare participant performance (RT and percentage of errors) when they report being on-task versus when they report being off-task **Hypothesis** • According to our analysis of preliminary data, and in agreement with the study by Robison et al. (2019), we expect that the report of MW will not differ based on the number of TP administered. • It is expected that the typical main effects and interactions usually observed with the ANTI-Vea will be replicated (Luna et al., 2021). Furthermore, based on our preliminary data, we expect that administering TP during the ANTI-Vea will not affect such main effects and interactions usually observed with the ANTI-Vea. Thus, group will not modulate the effects of orienting, alertness, and congruency in RT and errors. Additionally, it is anticipated that there will not be an interaction between the group and the decrease in EV, observed as a decrement in hits and false alarms across the blocks. Finally, we also anticipate that the AV decrement, observed as an increase in mean and variability of RT and lapses across blocks, will not be modulated by the group. • It is expected that a decrease in EV, a decrease in AV, and a decrease in executive control (observed as an increase of interference in RT, errors, and the inverse efficiency score across the block) will be observed in this study, as has been found in our preliminary data and in previous research (Luna et al., 2021, 2022). • Following Luna et al., (2022), we expect that the decrease in executive control will be associated with the decrease in EV. • According to the resources control theory (Thomson et al., 2015), an increase in MW is anticipated over the course of the task, as was found in our preliminary data. Furthermore, a significant negative correlation (i.e., the larger the increase in MW the larger the decrease in EV) is expected to be found between the positive slopes of the increment of MW and the negative slope of the decrease in executive control. It is also hypothesized that there will be an across block (for each participant) significant negative correlation between the across-blocks (for each participant) increase in MW and the across-blocks decrease in EV (i.e., hits). • Based on the study by Thomson et al. (2014), performance will be worse in general in the trials preceding reports of “out of the task” states than in those preceding trials reporting “on task” states. More specifically, the interference effect (measured as mean RT, errors, and inverse efficiency score) will be larger, and responses will be slower and less accurate in the former than the latter. **References** Bastian, M., & Sackur, J. (2013). Mind wandering at the fingertips: Automatic parsing of subjective states based on response time variability. *Frontiers in Psychology, 4*. https://doi.org/10.3389/fpsyg.2013.00573 Caggiano, D. M., & Parasuraman, R. (2004). The role of memory representation in the vigilance decrement. *Psychonomic Bulletin & Review, 11*, 932–937. Hancock, P. A. (2017). On the Nature of Vigilance. *Human Factors, 59*(1), 35-43. https://doi.org/10.1177/0018720816655240 Langner, R., & Eickhoff, S. B. (2013). Sustaining attention to simple tasks: A meta-analytic review of the neural mechanisms of vigilant attention. *Psychological Bulletin, 139*(4), 870-900. https://doi.org/10.1037/a0030694 Lim, J., & Dinges, D. F. (2008). Sleep Deprivation and Vigilant Attention. *Annals of the New York Academy of Sciences, 1129*(1), 305-322. https://doi.org/10.1196/annals.1417.002 Luna, F. G., Barttfeld, P., Martín-Arévalo, E., & Lupiáñez, J. (2021). The ANTI-Vea task: Analyzing the executive and arousal vigilance decrements while measuring the three attentional networks. *Psicológica Journal, 42*(1), 1-26. https://doi.org/10.2478/psicolj-2021-0001 Luna, F. G., Marino, J., Roca, J., & Lupiáñez, J. (2018). Executive and arousal vigilance decrement in the context of the attentional networks: The ANTI-Vea task. *Journal of Neuroscience Methods, 30*6(May), 77-87. https://doi.org/10.1016/j.jneumeth.2018.05.011 Luna, F. G., Tortajada, M., Martín-Arévalo, E., Botta, F., & Lupiáñez, J. (2022). A vigilance decrement comes along with an executive control decrement: Testing the resource-control theory. *Psychonomic Bulletin and Review, 29*(5), 1831-1843. https://doi.org/10.3758/s13423-022-02089-x Robison, M. K., Miller, A. L., & Unsworth, N. (2019). Examining the effects of probe frequency, response options, and framing within the thought-probe method. *Behavior Research Methods, 51*(1), 398-408. https://doi.org/10.3758/s13428-019-01212-6 Seli, P., Carriere, J. S. A., Levene, M., & Smilek, D. (2013). How few and far between? Examining the effects of probe rate on self-reported mind wandering. *Frontiers in Psychology, 4*(JUL). https://doi.org/10.3389/fpsyg.2013.00430 Smallwood, J., & Schooler, J. W. (2006). The restless mind. *Psychological Bulletin, 132*, 946–958. Thomson, D. R., Besner, D., & Smilek, D. (2015). A Resource-Control Account of Sustained Attention: Evidence From Mind-Wandering and Vigilance Paradigms. *Perspectives on Psychological Science, 10*(1), 82-96. https://doi.org/10.1177/1745691614556681 Thomson, D. R., Seli, P., Besner, D., & Smilek, D. (2014). On the link between mind wandering and task performance over time. *Consciousness and Cognition, 27*, 14-26. https://doi.org/10.1016/j.concog.2014.04.001 Warm, J. S., Dember, W. N., & Hancock, P. A. (1996). Vigilance and workload in automated systems. In R. Parasuraman & M. Mouloua (Eds.), *Automation and human performance: Theory and applications. Human factors in transportation* (pp. 183–200). Hillsdale, NJ: Erlbaum. [1]: https://osf.io/zmvkx [2]: https://osf.io/yxwge [3]: https://osf.io/cj83a [4]: https://osf.io/dnxs4