Altered small-world brain networks in schizophrenia patients during working memory performance

Area of Science:

• Neuroscience
• Cognitive Science
• Psychiatry

Background:

• Working memory (WM) deficits are a core feature of schizophrenia.
• Schizophrenia patients (SZ) display altered brain activity and functional connectivity during WM tasks compared to healthy controls (HC).

Purpose of the Study:

• To investigate differences in small-world network metrics derived from fMRI data between SZ and HC during a WM task.
• To assess how WM load impacts brain network topology in SZ and HC.

Main Methods:

• Functional magnetic resonance imaging (fMRI) data were collected from 35 SZ and 35 HC performing the Sternberg Item Recognition Paradigm (SIRP) at three WM load levels.
• Partial correlations of BOLD signal time courses between task-related ROIs (defined by ICA) were used to construct functional connectivity networks.
• Networks were thresholded into the small-world regime, and metrics like clustering coefficient and local efficiency were computed.

Main Results:

• At medium WM load, SZ networks showed significantly lower clustering coefficients and local efficiency than HC.
• Network metrics in SZ significantly changed with increasing WM load, unlike the relatively stable metrics in HC.
• In SZ, poorer task performance (longer reaction time) correlated with lower clustering coefficient and local efficiency at medium WM load.

Conclusions:

• Brain functional connectivity in SZ is more diffuse and less locally connected at intermediate WM loads compared to HC.
• SZ demonstrate inefficient and variable network structures in response to WM load, contrasting with the stable, clustered networks in HC.
• These findings highlight topological differences in brain networks underlying WM impairments in schizophrenia.