Suicide and Self-Injury
Spontaneous facial action exhibited during suicide assessments with adolescents and young adults
Ilana Gratch, M.S.
PhD Student
Columbia University
New York, New York
Maura A. Beaton, M.S.
Graduate Student
Columbia University
New York, New York
Daniella Ekstein, B.A.
Masters Student in Clinical Psychology
Teachers College, Columbia University
thornhill, Ontario, Canada
Alex Grattery, B.A.
Research Assistant
Columbia University
New York, New York
Emily Hubbard, B.A.
Research Assistant
Teachers College Columbia University
Brooklyn, New York
Supriya Kumble, M.A., Other
Graduate Research Assistant
Columbia University
New York, New York
Curreen Luongo, B.A.
Research Assistant
Teachers College, Columbia University
Brooklyn, New York
Emily N. O'Connor, M.A.
Research Assistant
Columbia University
New York, New York
Juno Pinder, M.A.
Student
Columbia University
New York, New York
Christine B. Cha, Ph.D.
Associate Professor
Teachers College, Columbia University
New York, New York
Suicide is the second leading cause of death among 15-24-year-olds in the United States, yet only 15% of mental health professionals report feeling very confident assessing youth suicide risk. Researchers have therefore sought to identify objective, empirically observable behaviors that can inform suicide risk detection. Despite progress in this area, behavioral markers have primarily been operationalized in the form of reaction times and task performance, which, while offering helpful approximations, only scratch the surface of what is possible with the rich, dynamic nature of behavioral data. The present study aims to examine non-verbal behaviors (i.e., spontaneous facial action) exhibited during suicide assessments, and test whether these behaviors can be used to identify adolescents and young adults with recent suicidal behavior. Facial action has long been thought to communicate emotion and intention, yet only two previous studies have tested facial action in a clinical context as a marker of suicide risk. In the present study, participants were recruited through social media (Facebook, Instagram) and partook in an hour-long Zoom interview with master’s level research staff, including an administration of the Columbia Suicide Severity Rating Scale (C-SSRS), and various self-report questionnaires. Participants are 70 18-24-year-olds with either: (1) past-year actual, interrupted, aborted suicide attempts or preparatory behaviors (n = 35), or (2) no history of suicidal thoughts/behaviors (n = 35). Participants are diverse with regard to gender (59% female, 17% male, 21% nonbinary/gender nonconforming), sexual orientation (50% lesbian/gay/queer/asexual), and race/ethnicity (19% Hispanic/Latino, 57% White/European American, 29% Asian, 11% Black/African American, 1% Native American/Alaska Native, 1% Native Hawaiian/Other Pacific Islander). Zoom interviews were video-recorded with participant consent using the Open Broadcaster Software. Spontaneous facial action exhibited by participants during the first two questions of the C-SSRS (i.e., In the past month, have you wished you were dead or wished you could go to sleep and not wake up? and In the past month, have you actually had any thoughts of killing yourself?) will be coded using the Facial Action Coding System (FACS). FACS is the most comprehensive approach to taxonomizing observable, anatomical movements on the face known as Action Units (AUs). Human coders note the occurrence and intensity of any of the 32 AUs for each frame of video. Coding is underway and being conducted by four certified Facial Action Coding System (FACS) master’s level research staff who are blinded to participants’ suicide-related histories. General linear models will be run to compare the proportion of the interview during which the interviewee exhibits certain AUs. More specifically, in line with extant studies, we hypothesize suicidal participants’ facial action will feature fewer smiles contracting the orbicularis oculi (AUs 6+12), and more chin raising (AU 17) and dimpling (AU 14). Findings from the present study can be used to inform the development of a wider array of behavioral markers of suicide risk including non-verbal behaviors exhibited during suicide assessments.