Too tired to talk? Sleep, fatigue, and social functioning

Abstract

Sleep has been shown to have many health promoting functions, and is vital for the maintenance of physiological, cognitive, and emotional well-being. Much of what we know about sleep derives from studies showing what happens when individuals do not obtain sufficient sleep. A number of such studies have shown that sleep may also be vital for one of the key drivers of human dominance - our ability to be social. The overall aim of this thesis is to advance our knowledge of the relationship between sleep and social functioning, specifically focusing on the short-term effects of insufficient sleep, poor sleep quality, and related states such as sleepiness and fatigue, on our ability to perform a number of socially vital behaviours and functions.

Paper I longitudinally investigated potential bidirectional relationships between sleepiness/sleep and the amount of social activity individuals engage in. This paper followed 681 participants who completed a time-use questionnaire over three weeks requiring individuals to report their activity every 30 minutes. Participants also completed the Karolinska Sleepiness Scale (KSS) six times per day and the Karolinska Sleep Diary (KSD) each morning. Activity categorised as ‘social activity’ was counted and its association to sleepiness and sleep duration was analysed. We observed that higher sleepiness predicted a decrease in probability of conducting social activity in the near future and a decrease in the duration of social activity. Sleep duration did not predict next day social activity. In the opposite direction, we found that social activity moderated the association between time-of-day and sleepiness. Additionally, an increase in the total amount of socialising per day predicted that individuals would sleep less subsequently. The study has implications for understanding the dynamic role of sleep in human social behaviour and suggests that sleepiness, likely due to decreased social motivation, may hinder the maintenance of a healthy social life.

Paper II employed an experimental sleep deprivation paradigm to investigate the effect of acute sleep loss on communication ability. Participants were randomised to either one night of total sleep deprivation (N = 91) or normal sleep (N = 92). The following day, communication ability was assessed via performance in two collaborative tasks with a partner (a model-building task and a word-description task). During these collaborative tasks, aspects of speech prosody were also measured (speaking duration, speaking volume, and speaking volume consistency). Participants additionally completed individual verbal fluency assessments. Performance on the model-building task was worse if the model-builder was sleep deprived, whereas the sleep deprivation appeared to boost performance in the instruction-giver. All other aspects measured seem unaffected by sleep loss. The results suggest that sleep deprivation leads to changes in communicative performance during longer instructive tasks, while simpler word-description tasks appear resilient.

Paper III assessed whether sleep deprivation leads to changes in facial appearance. Subjects (same data collection as Paper II; sleep deprived = 91, normal sleep = 90) were photographed and their skin tone was measured using a spectrophotometer. Using the photographs, the degree of eye-openness, mouth curvature, and periorbital darkness was calculated. A separate sample of raters (N = 63) judged the appearance of subjects using the facial photographs in three dimensions: paleness, fatigue, and healthiness. The results did not reveal any evidence of changes in facial appearance following sleep deprivation, in neither objective nor subjective measurements. However, decreased skin yellowness, less eye openness, downward mouth curvature, and periorbital darkness all predicted how fatigued individuals were rated by others. These results exemplify the difficulty of assessing fatigue from facial features alone.

Paper IV examined whether sleep was related to cognitive empathy ability. Study 1 (N = 291) investigates the effect of normal variation in sleep duration and sleep quality on our ability to recognise and categorise the emotional state of others. Study 2 (same data collection as Papers II and III; sleep deprived = 90, normal sleep = 91) investigated whether one night of total sleep deprivation impacted the ability to recognise and categorise the emotional state of others. Both studies tested performance on a multimodal forced-choice task of emotion recognition, where participants had to correctly categorise an actor-portrayed emotion on screen. Self-reported sleep duration/quality (Study 1), and sleep deprivation (Study 2) did not predict overall emotion recognition accuracy. Follow-up tests found no emotion-specific effects, apart from one positive association suggesting that greater self-reported sleep quality could predict more accurate recognition of disgust (Study 1). The combination of these studies suggests that the ability to accurately categorise the emotions of others is resilient to acute periods of insufficient sleep.

In summary, the work in this thesis has combined different study designs and data-sources to investigate the relationship between sleep and social functioning, as well as the influence of related states such as sleepiness and fatigue. We find evidence that social motivation and communication depends on aspects of sleep, while facial appearance and cognitive empathy appear resilient. This thesis therefore exemplifies the complex relationship between sleep and social functioning, while providing exciting directions for future research.