Moving forward in the neurocognitive study of mind-wandering: tracking the onset and time-course of mind-wandering

In the present work, we focused on mind-wandering (MW). MW is a multidimensional mental state that absorbs parts of our waking life and can be defined as a shift of attention away from a current activity towards internal thoughts unrelated to the ongoing activity. Although neurocognitive studies on MW have increased substantially over the last years, there are still a number of overlooked questions. In particular, a number of researchers have indicated a need for investigations of the dynamics of this process. To this end, it would be especially important to focus on the two basic elements of MW experiences, specifically, the moment of the onset of MW episodes (i.e., when the flow of thoughts starts) and the maintenance of these episodes over time (i.e., what happens during the continuation of the flow). In our studies, we focused on identifying the ignition moment (i.e., the onset) of MW experiences, as well as investigating its dynamics over time. Here we report three empirical studies employing a vigilance task that allowed elicitation and analysis of MW episodes in the laboratory. Pupillometry was also employed both in the second and third study to assess the association between physiological and self-report measures. The three studies used different versions of the vigilance task, in which participants were asked to detect infrequent target stimuli among a number of non-target stimuli and were eventually exposed to task-irrelevant verbal cues that could potentially act as triggers for MW episodes. MW was collected by using either a self-caught procedure (first and third studies) or a probe-caught procedure (second study). Specifically, the first study was carried out for investigating the cue-dependent nature of MW and verifying whether MW episodes could be linked to preceding triggers. Thus, in a between-subject design, we studied the causal role of meaningful external cues (i.e., verbal cues) in triggering MW experiences. We found that the exposure to the external cues increased the amount of MW and biased its temporal focus towards the past compared with a condition of no exposure to the cues. The second study was developed on the basis of the first one, with the main aim of associating a physiological measure (i.e., pupil diameter) to the onset and maintenance of MW experiences. The main finding was obtained by tracking pupil size over 6 seconds after MW triggers and non-triggers: we found a significantly larger pupil dilation following cues reported as triggers of the MW episodes compared to non-trigger cues. This suggested that the onset of MW and its unfolding over time were accompanied by a physiological marker (i.e., a pupil dilation). The third study was conducted with the main aim of replicating the results of the second study by using a different thought-sampling method, and extend them further. In particular, we used a self-caught procedure instead of a probe-caught procedure in order to track the pupil diameter following triggers of aware MW. We also examined whether and how the pupil dilation associated with MW was modulated by the emotional valence of MW. The main findings showed a significant increase in pupil diameter following triggers of aware MW compared to non-triggers, and this dilation appeared not to be modulated by the emotional content of MW. Collectively, these studies provided several contributions to neurocognitive research on MW. First, they demonstrated that the onset of MW episodes could be identified in the laboratory, since MW episodes were linked to external, meaningful and task-irrelevant stimuli. Second, they showed that a physiological index (i.e., pupil dilation) was associated with the onset of MW and accompanied its unfolding over time. Third, the use of the vigilance task with verbal cues and the self-caught procedure allowed us to also explore the latency of MW episodes (i.e., the time for the formation of thought and being aware of it). In the general discussion, we report some implications of these findings for further investigations in MW research.