Network analysis suggests working memory as a key target for facilitating subjective well-being

In response to rising rates of stress-related psychopathology, prevention programs have proliferated. These programs often target young adults, as the transition to adulthood marks a sensitive period for the emergence of psychiatric disorders. These programs typically target multiple processes and are therefore time-consuming. Here, we seek to advance our understanding of the relationships between multiple target processes and well-being as a first step toward simplifying these trainings to facilitate dissemination and scalability.

We assessed common prevention targets (i.e., emotion regulation strategy use, reward responsiveness, working memory, and delay discounting) and subjective well-being in 463 healthy young adults. First, we utilized a regularized partial correlation network to estimate relationships between target processes and subjective well-being. Subsequently, we applied a directed acyclic graph (DAG) with a Bayesian hill-climbing algorithm to infer potential causal relationships between these constructs.

Working memory was a key node associated with well-being across network models. Other nodes associated with well-being in the partial correlation network (i.e., reward responsiveness, reappraisal) appeared to be downstream consequences of well-being in the DAG. Though well-being was associated with greater habitual use of reappraisal and lower habitual use of rumination and catastrophizing in the partial correlation network, emotion regulation strategies were not directly linked to well-being in the DAG.

Using a Bayesian network approach, our findings point toward working memory as a potential key mechanism of well-being. Future work is needed to replicate these findings and determine the robustness of the identified network.