Past research indicates a causal relationship between emotional writing and health benefits (Smyth, 1998). At present, little is known about the mechanisms underlying change or if the emotional writing paradigm may be applied to a clinical setting. This present study reviewed current models of trauma and hypothesised three mechanisms of change leading to future health benefits: exposure, devaluation, and benefit-finding. Instructions for the standard writing paradigm were manipulated to isolate and increase engagement with each of these processes. It was hypothesised that if any one of these processes were to underlie health benefits, participants assigned to that condition would obtain more benefit than standard writing participants. Individual differences were also hypothesised to interact with each process to amplify or detract from their influence in leading to future benefit. A total of 201 university students were recruited from Griffith University. Participants were assigned to one of five writing conditions: Control, Standard, Exposure, Devaluation, and Benefit-Finding. Sessions were conducted once a week for three weeks. Physiological and self-report measures were taken before, during and after writing sessions. Follow-up assessments of psychological and physical health were taken at 2 and 6-months post-writing. Essay content analysis suggested that participants wrote in the instructed manner. Participants assigned to each of the groups experienced expected amounts of distress and affect changes. Overall, results failed to replicate the beneficial health effects for the standard emotional writing paradigm. There were no significant physical or psychological benefits for the standard trauma-writing participants in comparison to control. However, a trend in the appropriate direction was noted for illness visits at 6-months. Furthermore, in support of Greenberg and Stone's (1992) findings, standard writing participants who disclosed more severe and personal experiences evidenced significant illness visit reductions in comparison to control. Comparisons between standard and experimental trauma writing groups failed to support hypotheses that any one mechanism was responsible for physical health benefits. Examination of psychological self-report measures indicated exposure participants experienced the greatest reduction on the Impact of Events Scale at two months. However, these participants experienced greater reduction of positive affect and growth for the experience. They also became more anxious, depressed, and stressed at six-months follow-up. Process variables were examined within the exposure condition to explain these findings. Habituation was found to be strongly associated with the alternate outcomes. Individual differences. Including alexithymia, absorption, and negative affect, were also related to outcome. Benefit-finding participants experienced the greatest increase on a measure of post-traumatic growth at two-months and positive affect for the experience, but the finding was significant only in comparison to exposure and devaluation groups. The results of this study failed to identify the process of change, but suggest specific areas for future research. The findings demonstrate the importance of comprehensive health research to avoid blanket statements that suggest a paradigm either does or does not lead to health benefits. The results also support the manipulation of the writing paradigm to examine the role of emotion processing in trauma and health research.