Assessment of Self-Injurious Thoughts Using a Behavioral Test?

Print More

American Journal of Psychiatry/Harvard University

Abstract at

Mindful of the fact that youth who intentionally injure themselves are often unable or unwilling to talk about their behaviors, two Harvard professors found a way to improve the chances of predicting that a youth will self-injure by testing youths’ attitudes on the subject.

Mathew Nock, an assistant professor of psychology, and Mahzarin Banaji, the Cabot Professor of Social Ethics in the department of psychology, used the Implicit Association Test (IAT) – a computer application designed by Banaji – to measure the keystroke reaction times of participants as they grouped sets of words like “me” and “they” with the phrases “cutting” and “no cutting.” A second round of tests measured reaction times to sets of words in the categories of “good” and “bad,” with the words “cutting” and “no cutting.”

The premise underlying the test is that people will perform a behavioral response faster when the concepts or categories they’re classifying are associated in their minds.

“For suicide, self-injury and other things like drug use and violent ideas, people are often motivated to conceal their thoughts,” Nock said. “Or they may be unaware themselves about how they think or feel about different people or behaviors.”

Studies have shown that people who injure themselves are particularly motivated not to disclose their behavior, because the act of self-injury reduces or distracts them from intense feelings of pain or discomfort.

“It’s a way to treat their pain by themselves. And they don’t want others to take that away from them,” Nock said.

The tests – administered to 89 twelve- to 19-year-olds, 36 of whom had never self-injured – “revealed significant behavioral differences between the self-injurers and non-injurers,” according to the study, which appears in the May 2007 issue of The American Journal of Psychiatry.

Youth who engaged in self-injurious behavior were “more likely to quickly and consistently group words that were relevant to cutting with words related to the self,” the study reported. Although both groups of youth demonstrated an association between “cutting” and “bad,” that association proved weaker among self-injurers.

Study participants were also assessed using more conventional measures of self-injury, including self-reporting and personal history, that capture demographic data, IQ and previous mental disorders. Used in conjunction with such conventional testing, the IAT significantly improved the researchers’ ability to predict which of the participants were likely to injure themselves.

“The ultimate goal is that this [test] could be administered in the schools, in emergency rooms and in physician’s offices via laptop” to augment assessments that rely on self-reporting, Nock said.

Although there has been little research in the United States regarding the prevalence of self-injury among youth, a 2006 literature review by Cornell University estimated that between 4 and 38 percent of American youth have injured themselves.