Dr. Hoffman and Colleagues Reply

To the Editor: Mr. Aleman and Dr. de Haan question the conclusions of our study, which found specific alterations for a speech perception task, but not for an auditory, nonlanguage attentional task, when comparing patients with schizophrenia who reported auditory hallucinations with nonhallucinating patients with schizophrenia and normal comparison subjects. We interpreted these data as supporting the hypothesis that information processing alterations underlying the generation of auditory hallucinations are relatively specific to speech processing systems. Mr. Aleman and Dr. de Haan argue that other nonlanguage capacities need to be considered as etiological factors.

They cite Mintz and Alpert (1972), who found an association between auditory hallucinations and increased vividness of auditory imagery on the basis of a subjective rating system. Later studies, however, have not shown an association between the increased vividness of auditory imagery and auditory hallucinations (1, 2). These studies in fact suggested that auditory hallucinations are accompanied by decreased vividness of auditory images.

A report by Morrison and Haddock (1997) was also cited by Mr. Aleman and Dr. de Haan. This study suggested that there is a bias in attributing internal thoughts to external sources in patients with schizophrenia who report auditory hallucinations, a finding that challenges our hypothesis regarding a language-specific mechanism leading to auditory hallucinations. It is possible, however, that the persistent experience of auditory hallucinations conditions these patients to expect an external, nonself locus of control for their thoughts and behaviors. If so, reality monitoring biases could be a consequence of auditory hallucinations rather than a cause.

We also question an explanatory model of auditory hallucinations based on reality monitoring because it predicts that the distribution of hallucinations in patients with schizophrenia should mirror the nature and content of the imagery of ordinary thought. However, many patients state that their hallucinations derive from an external, nonself source precisely because the content of these experiences is so uncharacteristic of their own thoughts. Most patients with schizophrenia who hallucinate report only auditory hallucinations of speech or “voices.” Yet there is no reason to believe that these patients are not also capable of experiencing, for instance, visual or musical imagery. Why certain types of images would be subjected to failures of reality monitoring is a question raised by this model. One explanation is that only images that are especially vivid are transformed into hallucinations. However, our work with patients who hallucinate has revealed that some auditory hallucinations are experienced as so indistinct and of such low volume that their verbal content is not perceptible.

We hope that our article and the important issues Mr. Aleman and Dr. de Haan raise will encourage additional language and nonlanguage research that examines the mechanism of auditory hallucinations in patients with schizophrenia.