Attentional Bias for Nonsuicidal Self-Injury Cues Among Self-Injuring Young Adults

Background and Purpose: Nonsuicidal self-injury (NSSI) is a prevalent problem behavior among young adults (e.g., 38.9% of university students) that is maintained by complex transdiagnostic processes. For self-injuring individuals, exposure to NSSI words or images on social media could result in the development of NSSI attentional bias (AB)—the preferential allocation of attentional resources to specific aspects of environmental stimuli. The role of AB in the maintenance of NSSI warrants particular investigation given the association between AB and other behaviors, e.g., suicide AB is predictive of future suicide attempts. In the present study, we developed and administered a dot probe task to assess AB to NSSI cues to a sample of young adults with a history of NSSI. We hypothesized that self-injuring individuals would exhibit a statistically significant NSSI AB, as well as heightened NSSI urge in response to NSSI cues.

Methods: Participants were 30 young adults between the ages of 18 and 26 with a history of NSSI. AB was assessed via a dot probe task measuring reaction times to target probes replacing neutral images compared with reaction times to target probes replacing NSSI-related images (e.g., injuries, razors) presented for either 200 or 2000 ms. In order to determine the specificity of these effects, we examined AB towards negatively-valenced images unrelated to NSSI (e.g., angry faces). Participants rated their urge to engage in NSSI in response to viewing the NSSI cues during the dot probe task. AB was assessed with paired t-tests for each stimulus duration. We used the Bonferroni correction to adjust for multiple comparisons.

Results: A paired samples t-test revealed participants showed a significant AB to NSSI cues presented for 200 ms, t(29) = -3.55, p= .001, indicating participants had significantly shorter reaction times to probes replacing NSSI cues relative to probes replacing neutral cues. A paired samples t-test revealed participants did not exhibit a significant AB to negative cues presented for 200 ms, t(29) = -2.35, p = .03, when accounting for multiple comparisons and an alpha set at .0125. There was no correlation between NSSI AB and negative AB for 200 ms cues, r = -.19, p = .32, and on average participants exhibited a greater AB to NSSI cues (M = 16.25, SD = 25.08) compared to negative cues (M = 9.37, SD = 21.86). A paired samples t-test indicated NSSI urge ratings following the dot probe task significantly increased from baseline levels, t(29) = -3.80, p = .001.

Conclusions and Implications: Individuals in this sample exhibited an AB towards NSSI cues presented for 200 ms—the first such demonstration in the scientific literature. Importantly, participants did not exhibit a significant AB towards negatively-valenced cues, indicating that self-injuring young adults have an AB that is specific for NSSI cues. Further, attending to NSSI cues significantly increased the urge to engage in NSSI, demonstrating that the task is tapping a clinically-relevant process. As such, the dot probe task may be a sensitive and specific tool for detecting NSSI risk.