The EAR is an unobtrusive portable audio recorder that researchers have used to study

Mood is an essential feature if not the defining characteristic of ongoing conscious experience. As both philosophers and psychologists have noted, consciousness bereft of mood is rare or nonexistent (DeLancey, 1996, Watson, 2000). Such a constant presence implies an important function, and indeed, mood does more than just provide affective tone (e.g. “the blues” or “rose-colored glasses”) to people’s moment-to-moment phenomenological worlds.

In daily life, mood acts as a subtle yet pervasive causal agent. Mood states alter the way humans recall autobiographic memories (Miranda & Kihlstrom, 2005), make decisions (Loewenstein & Lerner, 2003), resist or fall prey to persuasion (Schwarz & Clore, 2007), perceive and are perceived by others (Forgas, 2003), and influence how people respond to social situations (Zajonc, 2000). Momentary moods even exert traceable effects on the immune system and the regulation of physical health (Kiecolt-Glaser et al., 2002, Pressman and Cohen, 2005). Finally, mood dysregulation is a defining feature of many psychiatric disorders and is implicated in their onset, development, and recovery.

As pervasive as mood effects are, humans often remain largely unaware of the factors that are responsible for fluctuations in their moment-to-moment mood (Gilbert, 2006). It is a common experience to find oneself fumbling to understand why one is feeling a certain way. Not only are the causes of mood often beyond conscious access, but mood fluctuations themselves may go unrecognized. Therefore, it is important to develop a more complete understanding of patterns underlying variability in mood.

Over the past two decades, emotion researchers have made critical progress in identifying dispositional and contextual influences on everyday affective experiences. This article explores the role of a third component of daily mood fluctuations and their behavioral consequences: rhythmic processes, including both circadian and socio-cultural rhythms. It is well established using forced desynchrony protocols, the gold-standard in chronobiological methodology,1 that there is a circadian component to self-reported mood (Boivin et al., 1997). More recently, a number of studies have further explored this phenomenon, noting that self-reported positive affect (PA) varies according to a daily rhythm, while self-reported negative affect (NA) does not (Murray, 2007, Murray et al., 2002, Thayer, 1997, Watson et al., 1999, Wood and Magnello, 1992). Some studies (e.g., Murray et al., 2002) have experimentally controlled for social interaction, thus buttressing the argument that the rhythm in PA is endogenous, biologically-driven (i.e., circadian) and not merely diurnal (i.e., a daily pattern with uncertain causation). The disparity in patterns has been linked to different motivational systems regulating PA and NA.

These findings raise a critical question—is the diurnal pattern confined to how people describe their inner mood landscape, or does it also surface in their affect-associated behaviors in the external world? Answering this question is important for at least two reasons. First, from a multiple-method perspective (e.g., Larsen & Prizmic-Larsen, 2006) it is critical to establish whether the evidence of diurnal patterns underlying affect generalizes across methods and replicates if affect is assessed free of self-report. Second, it is of theoretical interest to test the extent to which diurnal fluctuations in experienced affect “spill over” into observable behavior. Affect display is partially under conscious control and often modulated according to social norms (e.g., a churchgoer may feel like laughing during service but decide to suppress it; a student may feel like singing in class but realize it is inappropriate to do so). Finally, studying diurnal rhythms in affect-associated daily behaviors is also in line with Baumeister and his colleagues’s recent plea for “affirmative action for action” (p. 401) research in social and personality psychology—a field that currently experiences a dearth of studies that measures “actual behavior” (Baumeister, Vohs, & Funder, 2007).

For the current study, we used the Electronically Activated Recorder or EAR (Mehl, 2007, Mehl et al., 2001), a relatively novel unobtrusive observation method, to sample affect-associated behavior directly from the natural stream of daily life and independent of self-reports. The EAR tracks participants’ real-world behaviors by periodically recording snippets of ambient sounds from their momentary environments. The sampled ambient sounds are then coded for aspects of participant’s moment-to-moment social behaviors, interactions, and environments (Mehl et al., 2006, Mehl and Pennebaker, 2003, Mehl et al., 2007).

This study extends prior research on diurnal rhythms in affect in four important ways: First, it subjects the hypothesis of diurnal rhythms in affect to a cross-method validation that is independent of self-reports and based on affect-associated behaviors that are socially important but infrequently studied (e.g., singing, arguing). Second, it tests whether the phenomenon is limited to the experience of affect or can extend to its behavioral expression in daily life. Third, it uses progress in statistical modeling and a theoretically specified mathematical function. Diurnal variability in affect is often studied by using repeated measures ANOVA on between 3 and 10 data points per person per day (e.g., Adan and Sanchez-Turet, 2001, Wood and Magnello, 1992); we directly fit a theoretically predicted sinusoidal curve to fine-grained EAR data consisting of more than 60 data points per person per day. Finally, we explore the idea that personality traits can moderate the identified diurnal patterns. Based on a recent finding that Neuroticism (N) may be associated with a lack of circadian rhythmicity in PA and a blunted amplitude in the core body temperature rhythm (Murray, Allen, Trinder, & Burgess, 2002), we examined whether N would be related to a blunted and Extraversion (E) to an amplified diurnal rhythm of PA-associated behaviors.

Methods

The data for this study were derived from two EAR projects that examined the personality implications of daily life. For details about Sample 1 see Mehl et al. (2006); for details about Sample 2 see Vazire and Mehl (in press).

Preliminary analyses

Preliminarily, we fitted two simple models (the unconditional means and unconditional growth models) for each of the five behaviors (1) to estimate the overall variability in the behaviors; (2) to assess the extent to which this variability resided within or between people; and (3) to test for any linear change in the behaviors over the course of a day. Analyses confirmed the existence of significant within- and between-person variability and indicated that (with one exception) the

Discussion

In summary, across two independent samples, we found that three behaviors associated with PA (socializing, laughing, and singing) followed 24-h sinusoidal patterns centered around participants’ mean waketimes. Even after accounting for the variance associated with socializing, both laughing and singing retained significant fits to the cosinor model, supporting our hypothesis—but effectively not ruling out the possibility—that they are not entirely driven by socio-cultural, environmental factors

Cited by (57)

• Two experience sampling studies examining the variation of self-control capacity and its relationship with core affect in daily life

  2018, Journal of Research in Personality

  Firstly, we measure both trait self-control and self-reported state self-control capacity in a multi-session multi-day experience sampling protocol to differentiate the variation of self-control capacity at person-to-person, day-to-day, and moment-to-moment levels. We also aim to examine whether moment-to-moment variation can be explained by a time-of-day effect, and especially as a diurnal pattern, using a statistical technique called cosinor fitting (e.g., Hasler, Mehl, Bootzin, & Vazire, 2008; Murray et al., 2009). There are clear indications for diurnal patterns in people’s core affect, such as arousal levels (e.g., Smolders, De Kort, & Van den Berg, 2013; Stone et al., 2006; Wood & Magnello, 1992), and valence (Murray et al., 2009; Stone et al., 2006).

Recommended articles (6)