Characterizing cognitive heterogeneity on the schizophrenia-bipolar disorder spectrum

T. E. Van Rheenen, K. E. Lewandowski, E. J. Tan, L. H. Ospina, D. Ongur, E. Neill, C. Gurvich, C. Pantelis, A. K. Malhotra, S. L. Rossell, K. E. Burdick

Research output: Contribution to journalArticlepeer-review

146 Scopus citations


Background Current group-average analysis suggests quantitative but not qualitative cognitive differences between schizophrenia (SZ) and bipolar disorder (BD). There is increasing recognition that cognitive within-group heterogeneity exists in both disorders, but it remains unclear as to whether between-group comparisons of performance in cognitive subgroups emerging from within each of these nosological categories uphold group-average findings. We addressed this by identifying cognitive subgroups in large samples of SZ and BD patients independently, and comparing their cognitive profiles. The utility of a cross-diagnostic clustering approach to understanding cognitive heterogeneity in these patients was also explored. Method Hierarchical clustering analyses were conducted using cognitive data from 1541 participants (SZ n = 564, BD n = 402, healthy control n = 575). Results Three qualitatively and quantitatively similar clusters emerged within each clinical group: a severely impaired cluster, a mild-moderately impaired cluster and a relatively intact cognitive cluster. A cross-diagnostic clustering solution also resulted in three subgroups and was superior in reducing cognitive heterogeneity compared with disorder clustering independently. Conclusions Quantitative SZ-BD cognitive differences commonly seen using group averages did not hold when cognitive heterogeneity was factored into our sample. Members of each corresponding subgroup, irrespective of diagnosis, might be manifesting the outcome of differences in shared cognitive risk factors.

Original languageEnglish
Pages (from-to)1848-1864
Number of pages17
JournalPsychological Medicine
Issue number10
StatePublished - 1 Jul 2017
Externally publishedYes


  • Key words Clustering
  • cognition
  • heterogeneity
  • neuropsychology
  • psychosis spectrum


Dive into the research topics of 'Characterizing cognitive heterogeneity on the schizophrenia-bipolar disorder spectrum'. Together they form a unique fingerprint.

Cite this