Sleeping poorly is robustly associated with a tendency to engage in spontaneous waking thought

Sleeping poorly is robustly associated with a tendency to engage in spontaneous waking thought. Ana Lucía Cárdenas-Egúsquiza, Dorthe Berntsen. Consciousness and Cognition, Volume 105, October 2022, 103401. https://doi.org/10.1016/j.concog.2022.103401

Highlights
•    A comprehensive study on the relationship between self-reported sleep and spontaneous thought tendencies.
•    A wide variety of sleep and spontaneous thoughts measures were included.
•    Sleep predicted spontaneous thoughts tendencies, controlling for trait negative affect.
•    Sleep did not predict the frequency of positive-constructive spontaneous thought.
•    Findings demonstrate the unique role of sleep in relation to spontaneous cognition.

Abstract: We spend approximately-one third of our lives sleeping, and spontaneous thoughts dominate around 20–50% of our waking life, but little is known about the relation between the two. Studies examining this relationship measured only certain aspects of sleep and certain forms of spontaneous thought, which is problematic given the heterogeneity of both conceptions. The scarce literature suggests that disturbed sleep and the frequency of spontaneous waking thoughts are associated, however this could be caused by shared variance with negative affect. We report a comprehensive survey study with a large range of self-reported sleep and spontaneous thought measures (N = 236), showing that poorer sleep quality, more daytime-sleepiness, and more insomnia symptoms, consistently predicted higher tendencies to engage in disruptive spontaneous thoughts, independently of trait negative affect, age and gender. Contrarily, only daytime sleepiness predicted positive-constructive daydreaming. Findings underscore the role of sleep for spontaneous cognition tendencies.

4. Discussion

We examined associations between different self-reported aspects of sleep, a variety of spontaneous thoughts tendencies and trait affect in a comprehensive survey study. The findings generally agreed with our predictions. First, subjective measures of disturbed sleep were consistently associated with higher tendencies to engage in disruptive spontaneous thoughts, but not positive constructive daydreaming. Second, subjective measures of disturbed sleep were associated with higher tendencies toward negative affect and lower tendencies to positive affect. Third, higher tendencies to engage in disruptive spontaneous thoughts were related to higher trait negative affect, whereas a tendency toward positive constructive daydreaming was related only to more positive affect. Fourth, subjective measures of disturbed sleep were significant predictors of disruptive, and, especially, poor attention-related spontaneous thoughts, above and beyond negative affect tendencies, age and gender. The emotionally valenced daydreaming styles showed different patterns: a tendency toward positive constructive daydreaming was predicted only by subjective daytime sleepiness. A tendency toward guilt-fear of failure daydreaming was not predicted by any of the subjective sleep measures, but it was predicted by higher negative affect tendencies.

The study adds to the literature by showing that the relationship between self-reported disturbed sleep and the tendency to engage in disruptive spontaneous thoughts is not driven solely by negative affect. Self-reported sleep quality, daytime sleepiness, and insomnia symptoms predicted a tendency to engage in various forms of spontaneous thoughts, beyond negative affect tendencies, age and gender. This suggests other mechanisms behind the relationship between subjective sleep and spontaneous thought tendencies, besides shared associations with negative affect. First, self-reported poor sleep quality (Nebes, Buysse, Halligan, Houck, & Monk, 2009), insomnia (Liu et al., 2014) and daytime sleepiness (Anderson, Storfer-Isser, Taylor, Rosen, & Redline, 2009) have been related to reduced executive cognitive control. Impaired executive control in people suffering from disturbed sleep could reduce their ability to prevent the mind from wandering. This is consistent with the executive-failure account of mind wandering, which proposes that mind wandering episodes occur due to failures in the executive control system (McVay & Kane, 2010). Second, poor sleep quality, insomnia symptoms and daytime sleepiness have been associated with disrupted connectivity of the resting state, default mode network (DMN, Killgore., 2015, Nie et al., 2015, Tashjian et al., 2018). Poor sleep has also been associated with altered connectivity of attentional networks (Tomasi et al., 2009). Speculatively, both changes may be associated with an increased tendency to engage in mind wandering (e.g., Christoff et al., 2009, Poerio et al., 2017, Van Calster et al., 2017). Third, poor sleep quality increases sleep pressure (i.e. the increasing need for sleep per time awake) and this, in turn, increases the occurrence of local sleep-like activity in specific brain areas and networks involved in specific tasks (Cajochen et al., 1995, Muto et al., 2016). Local sleep-like activity in task-specific brain areas has been proposed as a neurophysiological marker of mind wandering experiences (Andrillon et al., 2019, Jubera-Garcia et al., 2021). Another speculative possibility is that an increase in the tendency for experiencing spontaneous thoughts may be a compensatory or restorative response to a lack of good sleep, equivalent to REM-sleep compensation (Carciofo et al., 2014b, Scullin and Gao, 2018). Finally, pressing current concerns, such as financial or social problems or recent stressful events, may influence sleep quality and/or cause insomnia (Van Laethem et al., 2015) and at the same time increase the tendency to engage in spontaneous thought when being awake (Klinger, 2009). Further research is needed to clarify whether and how these mechanisms interact with one another to regulate the tendency for experiencing spontaneous thoughts.

The tendency to engage in positive constructive daydreaming was not associated with, nor predicted by, any of the disturbed sleep measures, except for daytime sleepiness. This is consistent with previous research that failed to find an association between sleep quality and problem-solving daydreaming (Carciofo et al., 2014b, Denis and Poerio, 2017); between insomnia and positive constructive daydreaming (Starker and Hasenfeld, 1976, Starker, 1985) and between a tendency towards eveningness and problem-solving daydreaming (Carciofo et al., 2014b). These findings agree with the view that positive constructive daydreaming is a distinct form of spontaneous thought. It may serve as a source of problem solving, future planning, creativity and fantasy (Antrobus et al., 1966, Huba et al., 1977, Singer, 1966), as well as being related to compassion, simulating another person’s perspective and deriving meaning from events and experiences (Immordino-Yang, Christodoulou, & Singh, 2012). Therefore, and as suggested by Cárdenas-Egúsquiza and Berntsen (2022), positive constructive daydreaming may differ from disruptive spontaneous thoughts by at least two specific features. First, people who report higher tendencies to engage in positive constructive daydreaming also report higher tendencies toward positive affect, showing that this type of spontaneous thought is related to positively valenced emotions (Carciofo et al., 2014b), whereas most other forms of spontaneous thought are associated with negative affectivity (e.g., Killingsworth & Gilbert, 2010). Second, positive constructive daydreams may more often be initiated intentionally (Seli et al., 2016, Seli et al., 2017). People may frequently engage in positive daydreaming, such as planning an activity or solving a problem, in a deliberative fashion, which would then require executive functions to decouple from a current task or activity in order to sustain the daydream (Smallwood, 2013, Smallwood and Schooler, 2006).

In this line, one would expect people reporting disturbed sleep to show lower tendencies to engage in positive constructive daydreaming, due to reduced executive functioning. This is consistent with Carciofo et al. (2014b) who found more subjective daytime sleepiness related to less problem solving daydreaming. In the present study, however, we unexpectedly found more daytime sleepiness to be associated with higher tendencies to engage in positive constructive daydreaming. This discrepancy may reflect that the measure used in the present study assessed a broader concept of positive constructive daydreaming (i.e. fantasy, imagination, vivid imagery; McMillan et al., 2013) than problem solving and future planning as examined by Carciofo et al. (2014b).

The raw correlations showed consistent associations between self-reported measures of disturbed sleep and tendencies to engage in guilt-fear of failure daydreaming. However, disturbed sleep measures did not predict guilt-fear of failure daydreaming when controlling for negative affect tendencies in the regression analyses. This suggests that the association between self-reported insomnia symptoms and guilt-fear of failure daydreaming found in previous research (Starker, 1985, Starker and Hasenfeld, 1976) might have been driven by negative affectivity and not by insomnia symptomatology itself. This would be consistent with the maladaptive and negatively valenced nature of the guilt and fear-of-failure daydreaming style, which involves depressing, frightening and panicking daydreams related to fearing and failing responsibilities, failing loved ones, aggressing others and lying (Antrobus et al., 1966, Huba et al., 1977, McMillan et al., 2013, Singer, 1966).

Finally, younger age appeared to be a robust predictor of higher tendencies to engage in all the spontaneous thoughts measured in this study, except for involuntary mental time travel. This supports prior observations of decreased prevalence of mind wandering and daydreaming in older adults (Berntsen et al., 2015), and a less clear age-related decline for involuntary remembering (Berntsen et al., 2017, Maillet and Schacter, 2016).

The present study holds limitations. First, the cross-sectional and self-report nature of our data prevents reasoning about causality as well as generalizability to experimental studies. Here we used subjective measures of sleep, spontaneous thoughts and affectivity. We acknowledge that trait and state-level measures capture different aspects of the same phenomenon. However, we note that our results agree with previous survey and experimental studies measuring both trait and state-level sleep and spontaneous cognition (Cárdenas-Egúsquiza & Berntsen, 2022). Nonetheless, a bidirectional relationship in which the tendency to engage in spontaneous thoughts predicts the quality and duration of sleep is a possibility that should be further explored (Marcusson-Clavertz et al., 2019). Second, the influence of sleep duration (considered an aspect of sleep quality, Buysse et al., 1989, Kline, 2013) on the frequency of spontaneous thoughts was not examined in depth in the present study (but see the Supplementary Material Table S1 on correlations with sleep duration) and previous research has revealed inconsistent results (Robison et al., 2020, Walker and Trick, 2018). Third, we only applied the widely used PANAS to assess trait negative affectivity and its relation to spontaneous thoughts and sleep outcomes. However, other measures strongly related to negative affect, such as trait neuroticism, have been associated with more mind wandering, poorer sleep quality and a tendency towards eveningness (Carciofo, 2020, Carciofo and Jiang, 2021, Carciofo et al., 2016). Lastly, we acknowledge that the data were collected in November 2020, when lockdown restrictions to reduce the spread of the Covid-19 virus were still in place in several states in the United States of America, which may have influenced the findings (Leone et al., 2020, Simor et al., 2021). However, most of our results are consistent with previous research performed before the Covid-19 pandemic, suggesting that the relationships between subjective sleep outcomes and the tendency to engage in spontaneous thoughts are not contextual and seem to remain even during unusual situations.