Verbal fluency deficits have been pointed out as a possible endophenotype in schizophrenia (Szöke et al., 2008). However, whether these deficits are specific or linked to semantic-verbal inability remains unclear. Additionally, this cognitive domain is already affected in early psychosis and do not improve despite early clinical interventions.

Authors tested the efficiency of a cognitive intervention specifically developed for improving fluency in psychosis.

Ninety patients with first-episode psychosis were randomly assigned to one of two groups: Cognitive rehabilitation group (REHACOP) or occupational therapy. Patients at the REHACOP group received one month structured group rehabilitation sessions (3 per week) to improve fluency. Repeated assessments of semantic fluency and phonological fluency were conducted before and after the treatment.

Compared to occupational therapy, the experimental group produced significant additional improvements in phonological fluency (F = 6.87, p < 0.01), but not in semantic fluency (F = 0.61, n.s). The composite verbal fluency score was also significant (F = 4.65, p < 0.05). The improvement remained 3 months after the treatment end.

The cognitive treatment using REHACOP has proven to be effective in treating phonological fluency deficits in first-episode psychosis, whereas socialization or communication in group therapy by itself do not. The differential pattern showed by semantic fluency is consistent with the proposal of Szöke et al 2008, who suggest that semantic fluency is a putative endophenotype for schizophrenia with links to genetic basis compared to phonological fluency.