The main limitation of this study was that there were key differences between the three units used to obtain participants. They differed in terms of participant illness severity, how much participants were supervised following meals, levels of distraction and whether they could be on their own or not to complete the study. Completers and drop-outs differed significantly in ethnicity and there were marginally significant differences in other measures. In addition, the sample size was relatively small and participants varied in demographic factors and clinical indicators, creating a rather heterogeneous subject group. However, this heterogeneity might suggest that the highly significant effects observed are unlikely to depend critically on small procedural details or highly restricted participant demographics. Studies with more homogenous samples may help detect mediating factors for different cognitive tasks. A further question for future research is to investigate the maintenance over time of the effects reported in this article. It is acknowledged that the high drop-out rate from the study may represent a challenge in using this as an intervention in treatment settings.