Participants were randomised to either the BFP or a control treatment consisting of commercial computer games (CG). Participants were not blind to group allocation but all cognitive and symptom raters were. The 40 hours of BFP or computer game activity was expected to occur 4 days a week, for an hour each day, over a period of 10-12 weeks. The primary outcome was cognitive function assessed using the MATRICS consensus cognitive battery MCCB) (Nuechterlein and others 2008). The secondary outcome was social and role functioning assessed with Global Functioning: Social and Role scales. All clinical and cognitive assessments using symptom, functioning and cognitive measures were performed at baseline, post-treatment (at 3 months) and at 9-month follow-up (i.e. 9 months post baseline or 6 months after post-treatment assessment).
