Background. In recent years, evidence has accumulated that a significant proportion of schizophrenic patients have severe memory impairment, which cannot be attributed to the effects of medication, chronicity or institutionalization. Our group has demonstrated that memory impairment is associated with poor psychosocial outcome and treatment resistance. Work on the classical amnesic syndrome has suggested that memory training is facilitated by adopting an ‘errorless learning’ approach, where subjects do not experience failure during learning. This is based on the theory that the preserved implicit memory of amnesic patients results in implicitly remembered incorrect responses interfering with target items, in the absence of a functioning explicit memory system to allow differentiation.

Method. We compared three groups of subjects, memory-impaired schizophrenic patients, memory unimpaired schizophrenic patients and healthy controls.

Results. An errorless learning approach conferred a significant advantage on the memory-impaired schizophrenic group, bringing their performance up to the level of both control groups. In contrast, adopting a traditional trial and error, or errorful approach resulted in markedly impaired performance in the memory-impaired schizophrenic group only.

Conclusions. We conclude that errorless learning approaches may be worthy of further evaluation in the cognitive rehabilitation of memory-impaired schizophrenic patients.