Methodological Differences in Measuring Mind Wandering Influences Willingness to Re-Engage with Cognitive Tasks

What if mind wandering really isn’t all that bad, just misunderstood? There is a body of literature that often cites it as the issue in failing to perceive important events (McVay & Kane, 2009). On the other hand, there is a body of literature that cites mind wandering as a contributing factor for beneficial processes such as generating novel ideas or solutions to problems (Baird, Smallwood, & Schooler, 2011). However, the method by which data on mind wandering is measured influences the interpretations made by previous researchers regarding its benefits and detriments. Furthermore, little work has been done to assess how mind wandering influences re-engagement with a task that is being performed for extended periods of time. I focused on exploring how 1) Probe Placement influences task performance and willingness to re-engage with the task; 2) if mind wandering shares a similar cognitive mechanism as task switching by looking at performance differences between those with high versus low levels of mind wandering for a complex cognitive task (such as a task-switching paradigm) and the sustained attention to response task (SART); and 3) if ratings on scales measuring effort, difficulty, frustration, and available mental resources differ between those who mind wander and those who do not. For Study 1, I analyzed the results of 107 participants. Of those participants included in the analytic sample, 27 were randomly assigned to the interrupting probe with enjoyable music condition, 26 for the interrupting probe with unenjoyable music condition, 24 for the post probe with enjoyable music condition, and 30 for the post probe with unenjoyable music condition. Results show that the Participant Pool influenced the likelihood of mind wandering, such that those recruited from SONA Systems were 4.5 times more likely to report having high levels of mind wandering. Probe placement influenced errors of omission for SART performance. Those who received post probes were more likely to not respond to stimuli on the Go trials. Probe Placement also influenced their paid response indicating their willingness to re-engage with the task. Those in the interrupting probe group requested more money (mean = $62.38, SD = 31.47) compared to those who were in the post probe group (mean = $44.83, SD = 29.82). Levels of mind wandering also impacted participants’ paid responses. Those who were labeled as having high levels of mind wandering and were assigned to the interrupting probe condition had an average reported amount to be paid of $74.68 (SD = 22.92). However, those assigned to the interrupting probe condition and were labeled as having low levels of mind wandering had an average reported amount to be paid of $55.50 (SD = 33.73). For the post probe group, those identified as having high levels of mind wandering had an average reported amount to be paid of $51.50 (SD = 30.50) while those identified as having low levels of mind wandering had an average reported amount to be paid of $42.50 (SD = 29.61). For Study 2, I analyzed the results of 90 participants. Of those participants included in the analytic sample, 20 had been assigned to the enjoyable condition and 22 had been assigned to the not enjoyable condition for the SART task. While 25 had been assigned to the Enjoyable condition and 23 to the Not Enjoyable condition for the Task-Switching paradigm. There was a significant interaction between levels of mind wandering and congruency for task accuracy on the Task-Switching paradigm. Participants with low levels of mind wandering had an average accuracy of 0.94 (SD = 0.24) on congruent trials and an average accuracy of 0.92 (SD = 0.27) on incongruent trials. In contrast, participants with high levels of mind wandering had an average accuracy of 0.95 (SD = 0.22) on congruent trials and an average accuracy of 0.90 (SD = 0.30) for incongruent trials. Additionally, levels of mind wandering impacted accuracy and errors of commission for the SART task. Those in the Enjoyable condition with high levels of mind wandering had an average accuracy for No Go trials of 48% (x = 0.48, SD = 0.50) while those in the Enjoyable condition with low levels of mind wandering had an average accuracy of 66% (x = 0.66, SD = 0.47). Those in the Unenjoyable condition with high levels of mind wandering had an average accuracy of 57% (x = 0.57, SD = 0.50) and those in the Unenjoyable condition with low levels of mind wandering had an average accuracy of 68% (x = 0.68, SD = 0.47). Also, those with high levels of mind wandering and who were in the Enjoyment condition, an average rate of errors of commissions of 0.53 (SD = 0.50) while those with low levels of mind wandering and in the enjoyable condition had an average rate of errors of commissions of 0.34 (SD = 0.47). Furthermore, those with high levels of mind wandering who were in the unenjoyable condition had an average rate of errors of commissions of 0.43 (SD = 0.50). While those with low levels of mind wandering who were in the unenjoyable condition had an average rate of errors of commissions of 0.32 (SD = 0.47). Furthermore, I found a significant main effect of Task Type for participants’ paid responses. Those who went through the SART task requested an average dollar amount of $62.69 (SD = 31.50) while those who went through the Task Switching paradigm requested an average dollar amount of $49.46 (SD = 34.50). For Study 3, I used all 197 participants collected from the previous two studies. An exploratory factor analysis was conducted on the post-task questions administered to each participant. A Positive Affect factor, Stress factor, GTQ factor, Mental Demand factor, and Effort factor were identified. I then ran a linear model testing the effect probe condition (Interrupting Probe vs. Post Probe), task type (SART vs. Task-Switch), Levels of Mind Wandering grouping (High vs. Low), and sample source (MTurk vs. SONA) on willingness to re-engage. There was a statistically significant main effect of levels of mind wandering. those in the interrupting probe condition with High Levels of Mind Wandering, had a higher paid response (x = $74.68, SD = 22.92) compared to those with High Levels of Mind Wandering in the Post Probe condition (x = $56.40, SD = 33.20). However, those with Low Levels of Mind Wandering who were in the interrupting probe condition gave an average paid response of $55.50 (SD = 33.73), while those with Low Levels of Mind Wandering in the post probe condition gave an average paid response of $49.39 (SD = 32.19). The results show those with high levels of mind wandering are less willing to re-engage with the task. I then ran linear models testing the effect probe condition (Interrupting Probe vs. Post Probe), task type (SART vs. Task-Switch), and Levels of Mind Wandering grouping (High vs. Low) on the factors from the EFA. Probe placement and levels of mind wandering significantly influenced scores for the Positive Affect factor, while Probe Placement alone influenced scores on the GTQ factor. in the Interrupting Probe condition with high levels of mind wandering had an average Positive Affect score of -0.42 (SD = 0.95), while those in the Interrupting Probe condition with low levels of mind wandering had an average Positive Affect score of -0.03 (SD = 0.92). Those in the Post Probe condition with high levels of mind wandering had an average Positive Affect score of 0.24 (SD = 0.90). Finally, participants in the Post Probe condition with low levels of mind wandering had an average positive affect score of -0.02 (SD = 0.83), indicating those in the Post Probe condition had a more positive view of the cognitive task. For the GTQ factor, participants who received the mind wandering probe after the cognitive task had positive scores on the GTQ factor (x = 0.07, SD = 0.96), indicating more mental resources after completing the task, while those who received the interrupting probe had on average negative scores on the GTQ factor (x = -0.19, SD = 1.10) indicating fewer mental resources after completing the task. I also found that mind wandering affected scores for the Stress factor. Those with high levels of Mind Wandering generally had higher ratings of Stress (x = 0.51, SD = 0.86) towards the task compared to those Low levels of Mind Wandering (x = -0.25, SD = 0.97), indicating that those who had a high level of Mind Wandering found the tasks to be more demanding of their time and frustrating to engage with. The results show that participants who were labeled as High Levels of Mind Wandering generally viewed the tasks negatively and found them to be more stressful. Overall, results show that Probe Placement influences willingness to re-engage with the SART task. Additionally, those who report having high levels of mind wandering during the cognitive tasks are less willing to re-engage, report higher levels of frustration and lower levels of positivity towards the tasks.

Author

• Cisler, Dean Scott

• Mind Wandering
• Re-engagement
• SART
• Task-Switching

Identifier

hdl:1920/13855

OAI identifier oai:identifier

oai:MARS:1920/13855