Abstract:
The assessment of suicidal thoughts in depressed patients in the past has been limited in that it was subject to the decision of the assessors and therefore could be inaccurate. The purposes of the current study were to develop Dual-Process Tests for detecting suicidal ideation in depressed patients, and to construct a classification equation for predicting suicidal ideation. This study included both behavioral and electroencephalogram parameters. The sample consisted of 68 patients with depressive disorder who had different levels of suicidal ideation, aged 18-50 years. Research tools included computerized Dual-Process Tests, the Firestone assessment of self-destructive thoughts, and EEG recorders. ANOVA and MANOVA, discriminant analysis and multiple regression were adopted to test research hypotheses. The results showed that 1) The Dual-Process Tests were most suitable for use (M=3.74); 2) The response accuracy and D scores were significantly different at the .05 level; 3) The amplitude and latency of N200, P300, and alpha and beta waves were statistically different at the .05 level; 4) group classification equations successfully predicted for suicidal ideation by incorporating the amplitude of N200, the latency of P300 in congruence conditions, the D scores, and the relative power of the alpha wave. The variances were significantly explained by 81 percent from the given equations at the .01 level; 5) The Dual-Process Tests and the Firestone assessment of self-destructive thoughts had a high correlation (r=.78), significant at the .01 level; and 6) the Dual-Process Tests significantly predicted self-harm incidences or suicide attempts of the patients within three months at the .05 level. Thus, it can be concluded that the Dual Process Tests for detecting suicidal ideation in depressed patients had both congruent and predictive validities.