Abstract
In adult samples, tightly-controlled laboratory studies indicate the presence of circadian rhythms in positive (and negative) affect. Naturalistic studies also suggest the presence of diurnal positive affect rhythms in adults, the characteristics (acrophase, mesor, and amplitude) of which vary by self-report circadian preference-greater evening preference is associated with later acrophase, lower mesor, and lower amplitude in positive affect. We examined the extent to which diurnal affect rhythms are associated with 4 different measures of circadian timing, including dim light melatonin onset, in a sample of high-school adolescents who reported at least one drink of alcohol in their lifetime (N = 126, 17.3 ± 0.87 years, 55.6% female). Cosinor models found support for robust diurnal rhythms in positive, but not negative, affect. The overall modeled positive affect rhythm had an acrophase at 3:39 PM, a mesor of 9.77, and an amplitude of 1.61. Later circadian timing was associated with later acrophase in positive affect rhythms across the following measures: circadian preference (3:00 PM vs 4:20 PM, p < .001), chronotype (3:20 PM vs 4:11 PM, p = .014), and actigraphy-based midsleep (3:08 PM vs 4:16 PM, p = .014). We did not find significant associations between circadian phase (dim light melatonin onset) and positive affect rhythms. We also explored weekday-weekend differences in positive affect rhythms, finding significantly higher mesor (9.71 vs 9.99, p = .004) and lower amplitude (1.69 vs 1.26, p = .008) on the weekends than weekdays. In sum, compared to their peers, adolescents with later sleep and circadian timing experience a delayed peak in positive affect during the day, which may have consequences for behavioral activation and depressed mood. These findings underscore the importance of considering the role of sleep and circadian factors in affective processes during adolescence.