In Chapter 2, experienced burden of and adherence to a short-duration monitoring period was examined in a cohort of participants suffering an affective disorder (n=100), remitted persons (n=190) and healthy controls (n=94). Participants were invited to complete five EMAs a day, for fourteen days. The perceived burden at the moment of assessing (“momentary burden”) was slightly higher when momentary mood was worse. Nevertheless, momentary burden remained stable over the course of the monitoring period. Moreover, when asked to evaluate the overall monitoring period in retrospect, no inter-group differences were identified. In terms of adherence rates, the participants with affective disorders showed very high adherence rates (94%), with a rate on a par with the other two groups. The results suggest that it is feasible to apply short-duration and intensive EMA monitoring in a cohort of persons suffering from an affective disorder. Though, the participants were selected from the NESDA cohort, and the participants had been participating in the NESDA study for over nine years. As such it can be assumed that the NESDA-participants are highly motivated. In Chapter 3 and 4, we examined different distinctive profiles in mood dynamics among a sample suffering from subclinical depression who were recruited in the general population (chapter 3) and MDD patients during their first week of treatment (chapter 4). The analyses revealed two profiles in de first sample, and four profiles in Chapter 4’s sample. The profiles differed in terms of average mood and mood variability, but not emotional inertia. Even though the profiles, in both samples, different substantially in sample size, the studies provided us with initial insights in the heterogeneity of average mood and mood variability, but not emotional inertia in both subclinical depressed persons and MDD patients. In Chapter 5, building further on Chapter 4, posited a first attempt in examining the utility of the smartphone-based EMA-derived profiles of mood dynamics as predictor for treatment responsiveness during and after treatment for MDD. Treatment outcome was operationalized by a decrease in depression severity (assessed pre-treatment, at three-, six- and twelve-months). The results of this study suggested that smartphone-based EMA-derived profiles of average mood and mood variability could potentially distill patients who have a more favorable prognosis following treatment for depression. Caution is needed when interpreting our results, amongst others as some of the profiles derived in Chapter 4 referred to relatively small number of patients, implying large standard errors. The purpose of the study in Chapter 6 was to examine the assumed temporal associations in the theoretical framework of behavioral activation (BA) during blended cognitive behavioral therapy (CBT). The smartphone-based EMA included weekly reports of 240 patients on their levels of engaging in pleasant activities, on the perceived pleasure as a result of these activities and their mood. The analyses revealed positive cross-sectional relations between activation, experienced pleasure and mood. In other words, in weeks a patient engaged in more pleasant activities, they also reported experiencing more pleasure. In weeks a patient experienced more pleasure, they also reported a better mood. In weeks a patient reported a better mood, they also engaged in more activities. However, the analyses did not reveal any statistically significant temporal (cross-lagged) relations. The results of this study were partly in line with the theory of BA, but no conclusions could be drawn about possible temporal precedence or causality.