Symposia
Suicide and Self-Injury
E. Samuel Winer, Ph.D.
Associate Professor
The New School for Social Research
New York, New York
Amanda Collins, MA
Graduate Student
Mississippi State University
Mississippi State, Mississippi
Courtney Mason, MA
Graduate Student
Mississippi State University
Mississippi State, Mississippi
Michael R. Gallagher, B.S.
Graduate Student
Mississippi State University
Starkville, Mississippi
Matt Hanna, BA
Graduate Student
The New School for Social Research
New York, New York
Hilary DeShong, PhD
Assistant Professor
Mississippi State University
Mississippi State, Mississippi
Anhedonia, or the loss of interest or pleasure in people or things once enjoyed, is related to suicide risk (Winer et al., 2014; 2016a) when accounting for common symptoms of depression that have been examined in relation to suicidal behavior (e.g., Winer et al., 2014). Individuals with lower levels of self-compassion may be more susceptible to experiencing anhedonia and devaluation of positivity (Karl et al., 2017; Tobin & Dunkley, 2021). One possible reason for this is that individuals with lower levels of self-compassion may believe that they are not worthy of experiencing positivity or think that positive experience may lead to negativity (Winer et al., 2016b, 2017, 2019), and thus engage in harmful coping strategies, including NSSI and self-blame. Self-compassion is related to NSSI and self-blame (Gratz, 2001; Karl et al., 2017; Phillips, 2019; Tuna & Gencoz, 2020). However, how these components are interrelated, and how anhedonia and reward devaluation may play a role, has not yet been investigated via network analysis. Thus, the current study examined how self-compassion, NSSI, anhedonia, fear of happiness, and self-blame interrelate in a series of networks.
Participants pre-selected for a history of self-injury and/or a previous suicide attempt were recruited (N = 271 after listwise deletion). We used the goldbricker function from the networktools package in R to identify redundant nodes and combined them with the reduce_net function. Then, we estimated the networks with the estimateNetwork function and utilized the igraph package to detect communities of nodes.
Six networks were examined in all. Results from the first five networks suggest that NSSI (self-punishment), self-blame, reduced self-compassion, and fear of happiness (or anhedonia) shared almost all positive edges.
In the sixth network, all items representing NSSI, self-blame, reduced self-compassion, anhedonia, and fear of happiness were included in the network. Nodes representing self-blame and worthlessness and not unhappiness, represented the most influential nodes in the network. Lastly, when examining the edge-weights between nodes of different communities, self-blame and anhedonia shared the strongest edge. Thus, self-blame may contribute to the development of anhedonia, or vice versa, even when accounting for a wide variety of other symptoms and coping mechanisms.