Circadian mood variations in Twitter content

Abstract

BackgroundCircadian regulation of sleep, cognition, and metabolic state is driven by a central clock, which is in turn entrained by environmental signals. Understanding the circadian regulation of mood, which is vital for coping with day-to-day needs, requires large datasets and has classically utilised subjective reporting.MethodsIn this study, we use a massive dataset of over 800 million Twitter messages collected over 4 years in the United Kingdom. We extract robust signals of the changes that happened during the course of the day in the collective expression of emotions and fatigue. We use methods of statistical analysis and Fourier analysis to identify periodic structures, extrema, change-points, and compare the stability of these events across seasons and weekends.ResultsWe reveal strong, but different, circadian patterns for positive and negative moods. The cycles of fatigue and anger appear remarkably stable across seasons and weekend/weekday boundaries. Positive mood and sadness interact more in response to these changing conditions. Anger and, to a lower extent, fatigue show a pattern that inversely mirrors the known circadian variation of plasma cortisol concentrations. Most quantities show a strong inflexion in the morning.ConclusionSince circadian rhythm and sleep disorders have been reported across the whole spectrum of mood disorders, we suggest that analysis of social media could provide a valuable resource to the understanding of mental disorder.

Diurnal variations of psychometric indicators in Twitter content

“…Observing temporal variations in the expression of our internal state requires the investigation of large populations of individuals sampled at an adequate rate which has previously proven difficult [ 3 ]. Previous attempts have relied on subjective reporting with inherent problems of self-reporting and recall bias, they have also been limited in their size, or in the duration of the period of observation.…”

“…The specific content generated on the social media has been studied to observe temporal variations in the collective expression of our emotional processes [ 3 , 8 , 9 , 10 ]. These processes show a complex hierarchy of positively and negatively valenced emotions.…”

“…These processes show a complex hierarchy of positively and negatively valenced emotions. Finer distinctions separate them further: the emotion of anger has been shown to follow a stable pattern in the 24-h day at different points of the week and year, while the emotion of sadness has been found to vary in response to these changing environmental conditions, with higher levels of interaction observed with the onset of sunlight exposure [ 3 ]. Positive and negative states have been found to remain independent in various settings [ 11 , 3 , 8 ] suggesting different mechanisms behind their expression.…”

“…Finer distinctions separate them further: the emotion of anger has been shown to follow a stable pattern in the 24-h day at different points of the week and year, while the emotion of sadness has been found to vary in response to these changing environmental conditions, with higher levels of interaction observed with the onset of sunlight exposure [ 3 ]. Positive and negative states have been found to remain independent in various settings [ 11 , 3 , 8 ] suggesting different mechanisms behind their expression. In a previous study we identified two distinct patterns of emotional variations within the circadian rhythm, with the positive emotions showing a bi-phasic behaviour, and the emotion of anger showing variations that inversely mirror the variations of plasma cortisol concentrations peaking early in the morning and slowly decreasing for the rest of the day [ 3 ].…”

“…Positive and negative states have been found to remain independent in various settings [ 11 , 3 , 8 ] suggesting different mechanisms behind their expression. In a previous study we identified two distinct patterns of emotional variations within the circadian rhythm, with the positive emotions showing a bi-phasic behaviour, and the emotion of anger showing variations that inversely mirror the variations of plasma cortisol concentrations peaking early in the morning and slowly decreasing for the rest of the day [ 3 ].…”