The presence of attentional asymmetries in patients with schizophrenia was investigated with particular emphasis on the effects of stage of disease, medication status and clinical symptom severity. A modified version of Posner's covert orienting of visual attention task (COVAT) which included both spatial and non-spatial cues was administered to six volunteer samples of subjects which consisted of (i) 15 unmedicated and acutely psychotic male subjects with schizophrenia, (ii) 15 male subjects with schizophrenia who had been receiving medication for 14-21 days, (iii) 10 chronic male schizophrenic subjects who had been receiving medication for at least two years, (iv) 10 acutely psychotic male subjects with non-schizophrenic psychoses, (v) 15 subjects with unilateral brain frontal lobe (n = 6) or parietal lobe (n = 9) lesions, (vi) and 15 male control subjects. Measures of saccadic and pursuit eye movements were also obtained from unmedicated and recently medicated subjects with schizophrenia. COVAT attentional asymmetries were present in unmedicated subjects with schizophrenia for the 150 msec stimulus onset asynchrony (SOA). These asymmetries arose because reaction times (RTs) to right visual field targets were significantly slower than RTs to left visual field targets when targets followed invalid spatial or non-spatial cues. These asymmetries were qualitatively similar to those found in the patients with unilateral parietal lobe lesions. Attentional asymmetries partially resolved with brief periods of medication and completely resolved with long periods of medication. No asymmetries were found in controls nor in unmedicated subjects without schizophrenia. No asymmetries of ocular motor function were found. In schizophrenia, attentional asymmetries may reflect a deficit in the disengagement of visual attention from the right visual field and appear to be a stage marker for the disease. However this attentional deficit is dynamic and may reflect disruption to the neurocognitive network controlling attention at the level of the anterior cingulate cortex.