Selective speech perception alterations in schizophrenic patients reporting hallucinated "voices"

Objective: The authors tested a model of hallucinated "voices" based on a n eural network computer simulation of disordered speech perception. Method: Twenty-four patients with schizophrenia spectrum disorders who reported hal lucinated voices were compared with 21 patients with schizophrenia spectrum disorders who did not report voices and 26 normal subjects. Narrative spee ch perception was assessed through use of a masked speech tracking task wit h three levels of superimposed phonetic noise. A sentence repetition task w as used to assess grammar-dependent verbal working memory, and an auditory continuous performance task was used to assess nonlanguage attention. Resul ts: Masked speech tracking task and sentence repetition performance by hall ucinating patients was impaired relative to both nonhallucinating patients and normal subjects. Although both hallucinating and nonhallucinating patie nts demonstrated auditory attention impairments when compared to normal sub jects, the two patient groups did not differ with respect to these variable s. Conclusions: Results support the hypothesis that hallucinated voices in schizophrenia arise from disrupted speech perception and verbal working mem ory systems rather than from nonlanguage cognitive or attentional deficits.