Date of Award


Degree Type


Degree Name

Doctor of Philosophy




Dr. David Dozois

Second Advisor

Dr. Bruce Connell

Third Advisor

Dr. Rod Martin


Despite a growing body of research, a ∞mprehensive understanding of suicide remains elusive. The current dissertation involved the development of an integrated cognitive affective theory of suicidal thinking and behavior, and the initial assessment of components of this theoretical model in three separate, but related studies. In Study 1, the integrity of the model was assessed in a sample of 397 (83 male and 314 female) undergraduate students. The resulting fit indices (CFI=.87; GFI=.90; RMSEA=.09) of the structural equation modeling (SEM) analysis indicated an adequate fit between the hypothesized model and the data in the sample, with the model amounting for 90% of the variance. The SEM results indicated that rumination and negative affect directly impacted both cognitive distortions and ∞gnitive deficits. Cognitive distortions then had a direct impact on the development of suicide ideation, whereas cognitive deficits had an indirect impact on the development of ideation, through their reciprocal relationships with cognitive distortions, as well as with rumination and negative affect. In the interest of parsimony, a simplified model, in which ∞gnitive distortions and cognitive deficits were included, but affect and rumination were omitted, was also assessed. The resulting fit indices (CFI = .98; GFI = .98; RMSEA = .05) indicated a better fitting model than the initial model, with the simplified model ac∞unting for 98% of the variance. In Study 2, the relationships among the components in the model were further assessed in a community sample of 25 non-ideators, 25 ideators, and 11 recent suicide attempters. Patterns of correlations between the variables in the model paralleled those found in Study 1. Further, the components of the model varied across the suicide continuum as predicted by the model. Suicide ideation, cognitive distortions, and negative affect were significantly higher in ideators than in non- ideators, and higher in recent attempters than in ideators. Positive affect, in ∞ntrast, decreased significantly across these 3 groups. Ideators reported significantly higher levels of rumination than iii did non-ideators, but no significant differences in ruminative tendencies were noted between ideators and attempters. Between group differences in cognitive deficits varied depending on the deficit in question. Avoidant problem-solving was significantly higher in ideators than non-ideators. Although avoidant problem-solving increased across the entire continuum, between-group differences for ideators and attempters were not significant. Cognitive rigidity also increased across the entire continuum, and was significantly higher in attempters than in non-ideators. Self-reported problem-solving ability did not differ appreciably between any of the groups. Overall, compared to cognitive deficits, cognitive distortions demonstrated much stronger and more consistent relationships with suicide ideation and the severity of recent suicide attempts. As such, the findings of Study 2 provided further support for the relationships in the proposed model. Study 3, which was exploratory in nature, involved conducting psychological autopsy interviews with the next of kin of 4 individuals who had died by suicide, and analyzing the content of their suicide notes for evidence of affect, rumination, cognitive distortions and ∞gnitive deficits. The results of Study 3 indicated that high levels of negative affect and cognitive distortions were evident in the thinking and behavior of fatal suicide attempters in the month prior to the suicide. Perhaps due to the process-oriented nature of rumination and cognitive deficits, there was less evidence of these variables in both the interviews and the suicide note analysis. In general, the results of Study 3 indicated that affect, cognitive distortions and to a smaller degree, cognitive deficits may have utility for understanding fatal suicide attempts as well. A secondary goal of the current dissertation involved assessing the utility of the components in the model as predictors of suicide ideation. In Study 1, hopelessness was a significant predictor of suicide ideation, but ambivalence, rumination and negative affect exhibited predictive utility above and beyond hopelessness. These findings were replicated in Study 2. Specifically, hopelessness significantly predicted suicide ideation, but the addition of ambivalence iv and negative affect (self-directed anger) significantly enhanced the prediction. In Study 2, relative to the severity of intent in recent suicide attempters, the results of the regression analysis indicated that hopelessness was the only significant single predictor. These findings were discussed in terms of the cognitive-affective state that was associated with ideators and attempters in the current studies. Further, the limitations of the current studies were discussed, as were ∞ntributions, implications and future directions.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.