A number of methodological limitations have been mentioned above, and will not be reiterated here. However, the present study was premised on two key assumptions: (1) that a 10-day Vipassana course represents intensive training in mindfulness meditation, fundamentally similar in nature to the mindfulness-based interventions described in the literature; and (2) that the innovative Internal Switching task (IST) employed here provides an accurate measure of improvements in executive function. These assumptions will be briefly explored, followed by an examination of other pertinent issues.
The Validity of the Vipassana Course as an Intervention. The 10-day course provided a unique natural population for the investigation of the psychological effects of intensive mindfulness meditation practice. All of the meditators included in the study had zero prior experience with mindfulness meditation, allowing precise measurement of the effects of mindfulness training.
One consideration is the inability to accurately assess the amount of actual meditation practice each participant engaged in during the course, the amount of application they applied to mastering the technique, and any individual difference variables which may have impacted upon the rate at which they were able to do this. There is a strict timetable on such courses, which meditators are actively encouraged to adhere to. There is also an intentional lack of potential distractions . It can therefore be reasonably assumed that the meditators in the present study performed approximately 110 hours of actual meditation practice over the 10 days. Furthermore, only meditators who completed the full course were included in the study.
The exploratory nature of this study should be reiterated here. To the best knowledge of the present researcher, this represents the first study to assess the impact of a 10-day Vipassana course using validated measures and a rigorous experimental design, and the first to directly examine internal attention-switching in such a population. Accordingly, the results presented here should be interpreted cautiously, and used as a guide for future research.
The Validity of the Internal Switching Task. The results supported the notion that the IST is a valid tool for measuring internal switching costs. This is consistent with the literature outlined above. However, switching costs were found to be invariant across conditions (neutral and affective) and groups (meditators and nonmeditators). The present results suggested that RTs were more plastic, demonstrated a significantly bigger effect as a result of the intervention, and thus may provide a more useful measure of executive function. Indeed, these results suggest that the meditators exhibited generally decreased RTs following the intervention.
Group Differences at Baseline. It was originally hoped that the control sample would consist of people drawn from waitlists for the same Vipassana courses as the experimental sample was drawn from. However, problems with recruitment (outlined in a footnote in the method) resulted in the majority of the control sample being drawn from the community. As a result, while both groups were matched with the meditators on age, gender and years of education, they demonstrated significant baseline differences in their levels of depression and mindfulness. Two possible explanations for this are provided here. Firstly, these results may reflect a self-selection bias, whereby certain traits and characteristics may motivate people to participate in Vipassana courses. Indeed, the pattern evident in the meditators' baseline scores suggests that they may have been drawn to the practice as a way of gaining insight into the factors underlying their dysphoria. Alternatively, the nonmeditators may represent an unusually healthy sample, and may have been exhibiting evaluation apprehension (Pelham & Blanton, 2003), a response bias whereby self-report items are rated according to their perceived social desirability. Support for this notion derives from the fact that the direct measures of executive function employed here were not significantly different at baseline.
Both possibilities are likely, as this sample consisted of friends, family, and colleagues of the investigator, many of whom were psychology students. They were matched with the meditators on age, gender, and years of education, but not on baseline measures of the independent variables in the present study. In particular, there were significant discrepancies on the BDI and the MAAS.
There were a number of potential implications of this. While both groups' baseline scores on the BDI were below the cutoff used in previous studies to indicate depressed samples (Farrin, Hull, Unwin, Wykes & David, 2003), the nonmeditators scored significantly lower than the meditators. This may have resulted in a restricted range of potential postintervention improvement, compared to the meditators. However, the large effect size for the time by group interaction on this measure suggests that the reduction observed amongst the meditators would likely have been significant had they begun with comparable preintervention scores to the nonmeditators. Furthermore, while it is uncertain why the nonmeditators demonstrated higher preintervention mindfulness scores than the meditators, there is no upper limit on the MAAS. This makes the postintervention improvement even more significant.
The Use of Self-Report Scales. The present study also employed a number of self-report scales to assess changes in affect and metacognitive processing. A number of potential complications arise from this. As already discussed, there is an inherent risk of self-report biases confounding the data, which may be exacerbated by demand characteristics. Meditators learn on the course that one of the principal aims of Vipassana meditation is to develop mindfulness, which naturally results in decreased depression and anxiety. They are thus arguably more likely to underreport such symptoms postintervention. Conversely, as mindfulness meditation aims to increase sensitivity to thoughts, feelings, emotions and sensations, scales which ask participants to rate their experience of each of these may result in an overreporting. Both of these effects may serve to explain some of the anomalous findings in the present study, such as the failure to detect significant changes in the meditators' anxiety and rumination levels. Furthermore, it may partially explain the apparent baseline discrepancy between the meditators and nonmeditators on the self-report measures of depression and rumination, in the absence of such significant differences on objective, performance measures. Researchers may wish to consider this when designing future mindfulness studies.