Neuropsychiatric (NP) symptoms are relatively common in patients with SLE. The diverse and dramatic clinical presentations, the unclear pathogenesis, the lack of diagnostic test/s and uncertainties about the optimal management are some problems facing a clinician. When serum anti P antibodies were claimed to be highly correlated to lupus psychosis, this needed confirmation. An ELISA for measuring anti P antibodies was developed and validated. The prevalence of anti P antibodies was determined in different patient groups in a large retrospective study. Although anti P antibodies were highly specific for SLE, there was no correlation between the presence of these antibodies and lupus psychosis or other NP symptoms. Two prospective studies were carried out to eliminate any bias in our retrospective study. In one, none of the patients developed psychosis and these antibodies were not found to be specific for lupus depression or anxiety. In the other, anti P antibodies were measured in Malaysian Chinese SLE patients. No correlation was found between these antibodies and NP-SLE but a high prevalence of these antibodies was demonstrated in this group. Genetic studies showed that there was an increase in HLA-Dr2w16X subtype allele in anti P-positive patients but this did not reach significance. The usefulness of measuring antineuronal antibodies in helping to diagnose NP-SLE was examined but these antibodies were not better indicators of NP-SLE. Although the clinical correlations of anti P antibodies remain controversial, anti P antibodies were found to selectively bind to neuroblastoma cell surfaces <i>in vitro</i> but the nature of the surface antigen was not determined. Finally, sera from patients with lupus psychosis were found to significantly influence the response of neuroblastoma cells to agonist-induced stimulation and if confirmed, would offer an explanation for the reversible changes in cell function associated with psychiatric lupus.