How emotions induced by reading influence sleep quality in young and older adults

Abstract

Objectives

This study examined whether emotions induced by reading before sleep influence sleep quality in young and older adults.

Method

Sixty older adults (64–75 years) and 60 young adults (18–35 years) were randomly assigned to three conditions: positive reading, neutral reading, and control. The reading groups read a short story at bedtime, whereas the control group kept its routine. Participants completed measures of affective states, subjective sleep parameters, and self-reported sleep quality related to emotions over 7 consecutive days.

Results

Older adults reported much longer sleep latency and lower sleep efficiency than young adults. In both reading conditions, older adults reported reduced sleep latencies compared to the control group. In the positive reading condition, older adults reported an increased sleep duration compared to younger adults and the other conditions. Young and older adults in the positive condition showed better self-reported sleep quality than those in the neutral conditions, regardless of age.

Conclusions

Reading at bedtime appears to reduce older adults’ time to fall asleep and increase their sleep duration. Positive emotions induced by reading short stories at bedtime seem to be a sleep-promoting factor that improves bedtime and wake time in young and older adults.

Keywords:

• Age-related differences
• emotion
• affective states
• reading
• sleep quality

Acknowledgements

This work was carried out within the scope of the project “Use-inspired basic research”, for which the Department of General Psychology of the University of Padova has been recognized as “Dipartimento di Eccellenza” by the Ministry of University and Research.

The authors would like to thank VS for his valuable contribution, and for lending his beloved and authorial pen to this research.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Applying all coefficients in the models here, a power analysis showed an estimated sample of 18–20 in each group should be sufficient to obtain a power of .80, an effect size of .30, and a p value less than .05 (using the R software’s pwr library). We aimed to gather as many participants as possible within a limited time to maximize the power, resulting in a sample size of 139 participants initially and 120 participants after exclusions.

2 A series of ANOVAs were run separately to check for differences among age groups and conditions regarding education and gender. The results confirmed that young and older people did not differ significantly in terms of years of education, Fs(1, 118) < 1, and gender, age group: χ2(1, 120)= 0.000, p = 1.000; condition: χ2(2, 120) = 0.000, p = 1.000 (Appendix, Table A1; Supplementary Material, Table 2S).

3 We recruited 73 healthy individuals (age range: 21–86; M = 57.08, SD = 21.09) for a pilot study, and they had to rate all 14 preselected stories to confirm their emotional valence as positive or neutral. The stories were rated using the SAM. A one-way analysis of variance revealed that the preselected positive and neutral stories differed in terms of valence, F(1, 72) = 190.59, p < .001, ηp2 = .157 (Positive stories: M = 6.70, SD = 1.41; Neutral stories: M = 5.47, SD = 1.41), and arousal, F(1, 72) = 20.24, p < .001, ηp2 = .019 (positive stories: M = 5.07, SD = 2.10; neutral stories: M = 4.50, SD = 1.93), confirming our selections.

Additional information

Funding

The authors received no funding from any external sources.