Network dynamics of affect and physical activity in a heterogeneous clinical sample with high negative affect: an ecological momentary assessment study
OA Version
Citation
Abstract
In clinical psychology, contemporary classification systems for mental disorders assume that abnormal behaviors are expressions of latent disease entities. Although the latent disease model has been influential in both medicine and clinical psychology, there are substantive and empirical reasons for disputing its validity in psychiatric disorders. An alternative to the latent disease model is the network approach. Rather than assuming that symptoms arise from an underlying disease entity, the network approach posits that disorders exist as systems of interrelated elements of a network. Although several studies have examined the network structure of depression and related emotional disorders using a cross-sectional design, there has been an increasing emphasis on the network dynamics that underlie emotional psychopathology. Several intensive longitudinal studies have been conducted on affect in clinical psychology; however, none have attempted to characterize the network dynamics of positive and negative affect as they relate to objective measurements of physical activity using smartphone-based sensing data.
The current study examined the network dynamics of positive affect, negative affect, perceived stress, and physical activity in a heterogeneous clinical sample of 34 individuals with high levels of negative affect. Participants underwent a two-week ecological momentary assessment phase. Results of the study revealed that in both the temporal and contemporaneous dynamic networks same-valence nodes exhibited positive associations and opposite-valence nodes evidenced negative associations. Physical activity exhibited significant auto-correlation, yet it was unrelated to other affect nodes in the network. Furthermore, critical slowing down, as indicated by temporal autocorrelation, in the affect and physical activity nodes were not predictive of symptom changes. However, momentary stressors were predictive of temporal autocorrelation in physical activity and affect. Baseline symptoms and maladaptive emotion regulation strategies were predictive of dynamics in negative affect. Finally, impulse response function analyses revealed that variance in negative affect nodes is accounted for largely by increases in negative affect, whereas variance in positive affect nodes is accounted for by increases in both positive and negative affect. These results provide insight into factors that influence the dynamics of positive and negative affect in patients with emotional disorders.
Description
License
Attribution-NonCommercial-NoDerivatives 4.0 International