Assessing the information content of ERP signals in schizophrenia using multivariate decoding methods

Area of Science:

• Cognitive neuroscience
• Neuroimaging
• Psychiatric research

Background:

• Multivariate pattern classification (decoding) quantifies neural information in individuals.
• Decoding methods can reveal differences in information representation between psychiatric and non-psychiatric groups.

Purpose of the Study:

• To investigate differences in working memory representation between people with schizophrenia (PSZ) and healthy control subjects (HCS) using ERP decoding.
• To examine the relationship between decoding accuracy, neural signal, and noise in PSZ and HCS.

Main Methods:

• Examined event-related potentials (ERPs) during a working memory task with varying memory loads (1, 3, or 5 items).
• Applied spatial pattern analysis of ERPs to decode which side of the display was in working memory.
• Compared decoding accuracy and noise levels between PSZ and HCS.

Main Results:

• Decoding accuracy was higher in PSZ than HCS at memory load 1.
• Decoding accuracy correlated with the ratio of memory-related ERP activity to noise.
• Noise levels (trial-to-trial variability) were similar between PSZ and HCS.

Conclusions:

• Multivariate decoding is applicable at the individual level for studying cognitive function in psychiatric populations.
• Contrary to expectations, higher decoding accuracy in PSZ at low loads suggests distinct neural processing mechanisms.
• Findings challenge the assumption of uniformly increased neural noise in schizophrenia impacting cognitive decoding.