Abnormal interactions of verbal- and spatial-memory networks in young people at familial high-risk for schizophrenia

Abstract

Background

Working memory impairment (especially in verbal and spatial domains) is the core neurocognitive impairment in schizophrenia and the familial high-risk (FHR) population. Inconsistent results have been reported in clinical and neuroimaging studies examining the verbal- and spatial-memory deficits in the FHR subjects, due to sample differences and lack of understanding on interactions of the brain regions for processing verbal- and spatial-working memory.

Methods

Functional MRI data acquired during a verbal- vs. spatial-memory task were included from 51 young adults [26 FHR and 25 controls]. Group comparisons were conducted in brain activation patterns responding to 1) verbal-memory condition (A), 2) spatial-memory condition (B), 3) verbal higher than spatial (A–B), 4) spatial higher than verbal (B–A), 5) conjunction of brain regions that were activated during both A and B (A ∧ B). Group difference of the laterality index (LI) in inferior frontal lobe for condition A was also assessed.

Results

Compared to controls, the FHR group exhibited significantly decreased brain activity in left inferior frontal during A, and significantly stronger involvement of ACC, PCC, paracentral gyrus for the contrast of A–B. The LI showed a trend of reduced left-higher-than-right pattern for verbal-memory processing in the HR group.

Conclusions

Our findings suggest that in the entire functional brain network for working-memory processing, verbal information processing associated brain pathways are significantly altered in people at familial high risk for developing schizophrenia. Future studies will need to examine whether these alterations may indicate vulnerability for predicting the onset of Schizophrenia.

Introduction

Schizophrenia is a chronic and heritable mental disorder that is characterized by positive and negative symptoms and neurocognitive impairments (Bang et al., 2015). First-degree relatives (offspring and siblings) of patients with schizophrenia have an almost 10-fold increased risk of developing schizophrenia compared with the general population (Gottesman, 1994). Those first-degree relatives of the schizophrenia patients, who are still within the age range of risk for developing schizophrenia (ages from 13 to 30 defined in Li et al. (2007b))), are considered at heightened risk (FHR). From the literature, FHR subjects for developing schizophrenia have received increased attention in clinical and neuroimaging studies aimed at determining the underlying neurobiological vulnerability that leads to, and predict the onset of schizophrenia.

Clinical studies have suggested that in the FHR individuals, neurocognitive deficits, especially in the verbal- and spatial-working memory domains, exist prior the presentation of positive and negative symptoms, and can be used as strong predictors of schizophrenia (Pukrop and Klosterkotter, 2010, Dickson et al., 2014, Scala et al., 2014, Bang et al., 2015, Hou et al., 2016). Among these studies, Bang et al. found that behavioral performance scores of the tests for verbal- and spatial-working memory in the FHR subjects were intermediate between low risk controls and patients with first-episode schizophrenia (FES) (Bang et al., 2015). A Meta-analysis of 25 studies [published between January 1987 and February 2013, including clinical longitudinal data from 905 FES, 560 HR (including both clinical HR and FHR), and 405 healthy controls (HC)] suggested that verbal-memory deficits were already established in the HR subjects before the prodromal phase of psychosis, while no further decline occurred during follow-up assessments (Bora and Murray, 2014). In a new study, Dickson et al. also found verbal- and spatial memory deficits in the FHR subjects when compared to HC; whereas by adjusting statistical analyses for IQ, no significant between-group differences in any neurocognitive domains were remained (Dickson et al., 2014). The inconsistency among these existing neurobehavioral studies may be caused by the heterogeneity of the study samples, different tasks and assessment criteria being used, as well as other factors.

Although the etiology of schizophrenia is still unknown, neuroimaging studies have established that disturbed activities and connectivities in the functional brain networks for sensory and cognitive processing underlie neurocognitive deficits in patients with schizophrenia and the FHR population (Li et al., 2009, Li et al., 2010, Li et al., 2012). Several spatial-working memory tasked-based fMRI studies have reported that compared to HC, the FHR subjects had decreased activation in the dorsolateral prefrontal cortex (DLPFC) and the inferior parietal cortex (IPC) (Callicott et al., 2000, Keshavan et al., 2002). One early fMRI study reported that during a verbal-working memory task, the FHR subjects showed greater task-elicited activation in prefrontal cortex (PFC) and anterior cingulate cortex (ACC) (Thermenos et al., 2004). A recent resting-state electroencephalogram (EEG) study demonstrated that compared to HC, FES had increased theta-band resting-state connectivity across midline, sensorimotor, orbitofrontal regions and the left temporoparietal junction, whereas the FHR subjects displayed intermediate theta-band connectivity patterns that did not differ from either FES or HC. Mean theta-band connectivity within the above network partially mediated verbal-working memory deficits in FES and FHR (Andreou et al., 2015).

Of note, as two of the core features of cognitive impairment in schizophrenia, verbal- and spatial-working memory deficits are often observed together, instead of distinctively in the diagnosed patients and even early in the HR status (Scala et al., 2013). However, most of the existing neuroimaging studies investigated the neural mechanisms associated with only verbal- or only spatial working memory processing. Inconsistent results of these studies (partly briefly review in last paragraph) can from sample differences and lack of understanding on interactions of the brain regions for processing verbal- and spatial-working memory. The current study proposed to investigate the pattern of interactions between the verbal- and spatial-working memory processing networks in the FHR population, and its differences from that in a well-match sample of low risk controls.