(PS3-A8) Examining Positivity Avoidance and Dampening Strategies: A Community Network Analysis

           Reward Devaluation Theory (RDT; Winer & Salem, 2016) posits that individuals learn to avoid prospectively rewarding stimuli due to past pairings of positivity with negative outcomes. Self-report measures exist which operationalize the biases outlined in RDT. Fear of happiness is a construct which involves feelings of uneasiness and anticipatory avoidance of happiness, due to fears that happiness will be replaced by negative outcomes. Gilbert et al. (2012) and Joshanloo (2013) have created fear of happiness scales (FHS), which quantify this maladaptive thought pattern. Additionally, the Response to Positive Affect Scale (RPA; Feldman et al., 2008), contains a dampening subscale, which measures the extent to which individuals downplay or destroy positive emotions. This concept is similar to fearing happiness, but is a responsive strategy. Given the apparent overlap between these measures, we examined the dynamic relationship between individual items of these measures, as well as the extent to which these items clustered into their parent measures.

            Participants (N = 759) completed the Gilbert and Joshanloo FHS, as well as the RPA. Network analysis was utilized to examine the dynamic interaction among items of the three measures (Borsboom & Cramer, 2013). Additionally, a community analysis was employed to examine which items were highly intercorrelated and thus, clustered together. Bridge symptoms were assessed, which are nodes that are important for connecting the communities that emerge. The data analytic plan was pre-registered via AsPredicted.org (#71150) before conducting analyses.

            After combining for redundant nodes, PCA.last.wrong (expected influence = 1.05), cheerful (expected influence = 1.24), and never_last (expected influence = 1.20) emerged as the nodes with the highest expected influence in the network. These items revolve around the idea that good feelings do not last and are usually followed by bad outcomes. The community analysis revealed that items largely clustered into their parent measures. Four communities emerged: an RPA and Joshnaloo FHS community, as well as two distinct Gilbert FHS communities. The FHS_deserve and frightened items were the only items from the Gilbert FHS to form a separate community. These items, which reflect feeling undeserving of happiness (FHS_deserve) and a fear of allowing oneself to become happy (frightened) also emerged as the strongest bridge symptoms, suggesting a strong influence in the connections across communities.

            The three most influential nodes in the network revolve around worry that things will go wrong or happiness will not last (PCA.last.wrong), being cheerful leads to bad outcomes (cheerful), and that good feelings never last (never_last), which are all theoretically related to RDT. It appears that due to a fear that negativity follows happiness, strategies regarding the anticipatory avoidance and responsive dampening of happiness are employed. Given the strong influence of these items, these biases may be good targets for positive affect interventions, which attempt to upregulate the positive valence system (Craske et al., 2016).