Wellbeing is an important predictor for overall health and favourable outcomes in life, i.e. less psychopathology, more self-worth, and positive work functioning. Like many complex traits, feelings of wellbeing fluctuate over time (i.e. across the day or week) and across different contexts (Eid & Diener, 2004; Li et al., 2014; Lyubomirsky, 2001). Different measures of such affect variability or affect dynamics can be computed. First, momentary affect reflects the average level of positive or negative affect across the measurement period. Second, the within-person standard deviation (SD) reflects the variability of affect across the measurement period. The SD is time insensitive and does not account for the moment-to-moment changes (Trull et al., 2015). Third, the first-order autocorrelation (AR) reflects the correlation between affect measurements at consecutive time points, i.e., temporal dependency. This measure is often referred to as inertia, i.e. the resistance to change (Koval et al., 2013; Kuppens et al., 2010; Suls et al., 1998). Fourth, the root square of the mean square successive difference (rMSSD) captures the fluctuations between moments and takes into account both the variability and temporal dependence (Jahng et al., 2008). The rMSSD is often referred to as affective instability (Koval et al., 2013; Trull et al., 2015). Fluctuations in positive and negative affect are thought to be adaptive and important for wellbeing and help to respond to environmental changes and demands (Carver, 2015; Frijda & Mesquita, 1994; Kashdan & Rottenberg, 2010). However, affective fluctuations are only functional within certain boundaries. If emotions do not change or change too strongly or abruptly this might signal dysregulation or is related to psychopathology. For example, abnormal affect dynamics have been associated with emotional dysregulation and a vulnerability of psychological problems later in life (Kuppens et al., 2012). Furthermore, depression has been found to be related both to a higher mean level of negative affect, and to a change in affect dynamics (Aan het Rot et al., 2012). The meta-analysis of Houben et al. (2015) showed that in adults, high variability and instability of emotions are associated with worse general wellbeing and depressive symptoms, with stronger effects of negative affect dynamics on general wellbeing than positive affect dynamics. However, most studies only take in account the linear relations between affect dynamics and mental health outcomes, whereas nonlinear effects, i.e. quadratic, could explain the relation as well. For example, both very high and very low levels of variability could be related to less wellbeing or more psychopathological symptoms. The first goal of the current study is therefore to explore the relation between affect dynamics and general measures of wellbeing in a large sample, taking into account both linear and quadratic effects. Individual differences Individual differences in the size of affect dynamics have been reported. Some people show relatively stable levels of wellbeing, e.g., stable levels of positive and negative affect over the day or week, while others fluctuate a lot (Eid & Diener, 1999; Gadermann & Zumbo, 2007). Genetically informative samples can be used to investigate the role of genetic and environmental factors playing a role in the individual differences. Mainly based on standardized and well-validated surveys, it is known that individual differences in general wellbeing are partly explained by genetic factors, i.e., the heritability is around 40% (Bartels, 2015; Nes & Røysamb, 2015). In contrast, the causes of individual differences in momentary wellbeing and affect dynamics have been estimated in only a few studies. Riemann and colleagues assessed positive and negative affect across five different mood-inducing situations in 300 twin pairs. The estimated heritability of momentary positive affect was small across the different situations, around 5-20% (Riemann et al., 1998). In a real-life setting, Menne-Lothmann et al. (2012) investigated momentary positive affect in female twins (N=260 twin pairs) using the experience sampling method. Positive affect ratings were averaged for each person across the 50 measurements to a single score of momentary positive affect. Genetic influences on this rough average momentary positive affect did not reach significance, with a heritability of 19% (95% CI = 0-50%). Similarly, Zheng et al. (2016) assessed daily positive affect in 275 twin pairs for a month and reported a non-significant heritability estimate for the average daily positive affect (18%, 95% CI = 0-56%). Finally, in a daily diary study of 237 twin pairs, genetic influences on daily positive affect were found to be small, but significant, i.e., around 25% (Burt et al., 2015). With respect to affect fluctuations, Jacobs et al. (2013) reported a heritability estimate of 18% for positive affect variability, using the experience sampling design for five days in 279 twin pairs. In a daily-diary design, Zheng et al. (2016) reported a heritability estimate of 34% (95% CI = 17% - 48%) for positive affect variability. However, in a recent daily-diary study, a non-significant heritability of 9% (0–25%) was reported for positive affect inertia, i.e., the extent to which individuals’ emotions carry over from one time point to the next (Zheng & Asbury, 2019). The inconclusive findings and large confidence intervals for the heritability of momentary positive affect, affect fluctuations and variability, and other momentary measures of wellbeing indicate the need for more research to affect dynamics in daily life in a large sample of twins. The second goal of this study is therefore to investigate the genetic architecture of the different affect dynamic measures in our large sample of twins of the Netherlands Twin Register.