Does arousal matter for positive affect? An exploration of high and low arousal positive affect in everyday life and associations with health behaviors and symptomology

Open Access
Jones, Dusti
Graduate Program:
Biobehavioral Health
Doctor of Philosophy
Document Type:
Date of Defense:
March 06, 2020
Committee Members:
  • Jennifer Elise Graham-Engeland, Dissertation Advisor/Co-Advisor
  • Jennifer Elise Graham-Engeland, Committee Chair/Co-Chair
  • Stephanie Trea Lanza, Committee Member
  • Joshua M. Smyth, Committee Member
  • Martin John Sliwinski, Outside Member
  • Thomas J Gould, Program Head/Chair
  • positive emotion
  • ambulatory assessment
  • affective arousal
  • health behaviors
  • health symptomology
Theories suggest that higher positive affect (PA) should be associated with indicators of better health (e.g., more adaptive health behavior, lower symptomology). Yet, empirical studies vary widely in their observations of such associations. A limited understanding of PA in everyday life and discrepancies in how PA is conceptualized – such as whether PA items comprise primarily high or low arousal items – may help explain the lack of consistent associations between PA and health indicators. The proposed dissertation sought to characterize PA in everyday life by arousal (Aim 1), to examine the extent to which measures of high and low arousal PA relate to one another (Aim 2), and to examine whether high and low arousal PA are uniquely associated with health indicators in everyday life (Aim 3). Participants (N=121 adults, ages 25-65yrs) completed assessments of PA, symptomology, and perceived health using their own smartphones six times a day for seven days, and completed waking and bedtime surveys to assess health behaviors. Health indicators included daily self-reported health behaviors (i.e., sleep, diet, physical activity), and momentary reports of symptomology (i.e., pain, discomfort, fatigue) and perceived health. There was a substantial degree of overlap (i.e., moderate to large effect sizes) between measures of PA (e.g., comparing retrospective vs aggregated PA, high and low arousal PA). Reported levels of low arousal PA in everyday life were relatively high: well above the mid-point of the scale and higher than reports of high arousal PA. Low arousal PA generally exhibited distinct temporal trajectories as well as different individual difference correlates (e.g., gender, age) compared to high arousal PA; an exception was that neuroticism predicted both less high arousal and low arousal PA in everyday life. High and low arousal PA were uniquely linked with several health indicators, with indications of differences by arousal. For example, high arousal PA was more robustly associated with health behaviors at the between-person level whereas low arousal PA was linked with subjective health quality (perceived health, sleep quality) at the between-person level and with symptomology at the between and within-person level. The present study provided a careful exploration of an under-examined aspect of PA: arousal. Level of arousal has important implications for how PA may be associated with health indicators, but the majority of studies have ignored this aspect of affect. Specifically, results suggest there may be differences in temporality, antecedents, and health-relevant associations of high and low arousal PA. Causality and directionality could not be determined from the present study. However, results from the present study provide the basis for more nuanced and rigorous hypothesis development and testing. The present dissertation provided preliminary comparisons of reports of high and low arousal PA in everyday life: Although measures of PA exhibited strong overlap, results overall suggest that there is value in distinguishing PA by arousal. Distinguishing between PA by arousal in future research may improve understanding of how PA links with health.