Resting-state degree centrality and Granger causality analysis of facial working memory in patients with first-episode schizophrenia

doi:10.21203/rs.3.rs-4840714/v1

Download PDF

Research Article

Resting-state degree centrality and Granger causality analysis of facial working memory in patients with first-episode schizophrenia

https://doi.org/10.21203/rs.3.rs-4840714/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: This study focused on the relationship between facial working memory (WM) and resting-state brain function abnormalities in patients with schizophrenia.

Methods: Resting-state functional magnetic resonance imaging (rs-fMRI) data were collected from 28 first-episode schizophrenia (FSZ) patients and 33 healthy controls (HCs). Degree centrality (DC) and Granger causality analysis (GCA) were used to assessbrain region connectivity. A match-to-sample task was used to examine visual WM for faces and houses. Correlations between DC and facial WM scores were analysed. Brain regions were selected as regions of interest (ROIs) and subjected to further GCA.

Result: The results revealed that WM accuracy was lower in FSZ patients than in HCs for both loads and stimuli (P < 0.010). FSZ patients presentedspecific facial WM impairmentsat high loads (t = 2.21, P = 0.031). DC values of the right middle frontal gyrus (MFG) were linked to facial WM accuracy (P < 0.050, FDR correction). GCA indicated inhibited connectivity from the right MFG to the right inferior frontal gyrus (IFG) and right thalamus and from the right postcentral gyrus to the right MFG in FSZ patients (P < 0.050, FDR correction). DC values of the right thalamus correlated with negative symptom scores (r = -0.44, P = 0.018) and affective symptom scores (r = -0.57, P = 0.002).

Conclusions: Our findings suggest that FSZ patients may have specifically impaired facial WM ability, which may be associated with altered functions in multiple brain regions. Some of these functions are associated with clinical symptoms, which may provide insight into the underlying neural mechanisms of schizophrenia.

schizophrenia

facial working memory

degree centrality (DC)

Granger causality analysis (GCA)

resting-state functional magnetic resonance imaging (rs-fMRI)

Schizophrenia is a prevalent severe mental illness that typically begins in early adulthood. According to epidemiological surveys, its prevalence in China is approximately 0.67% [1]. Clinical manifestations include "positive symptoms", such as hallucinations and delusions, alongside "negative symptoms", including affective, social, and cognitive deficits [2]. Throughout illness, deficits in various cognitive processes, notably working memory (WM), are consistently observed, with impaired WM being a core symptom [3].

WM is defined as a limited-capacity storage system crucial for temporary retention and processing of sensory information [4]. It plays a pivotal role in higher-order cognitive functions such as logical thinking, reasoning, and verbal comprehension [5]. Previous studies have posited that social functioning deficits in schizophrenia may be closely linked to impaired WM, particularly facial WM [6]. Facial WM, which is integral to visual WM, involves encoding and maintaining facial features critical for various face-related tasks essential for normal social functioning [7]. Numerous studies have consistently reported impaired facial WM in schizophrenia patients [8–10]. Conklin et al. [11] reported impaired facial memory in schizophrenia patients compared with healthy controls (HCs), even after controlling for spatial attention and verbal memory differences. Calkins et al. [8] reported significant deficits in immediate and delayed facial memory in schizophrenia patients without impairments in visual object memory, suggesting independent facial WM deficits.

Structural and task-state functional magnetic resonance imaging (fMRI) studies have inconsistently shown that reduced grey matter volume in the bilateral fusiform gyrus in first-episode and chronic schizophrenia patients is correlated with impaired facial feature memory [12, 13]. Task-state fMRI suggests that facial memory involves multiple brain regions from the occipital to prefrontal lobes [14, 15]. Yoo et al. [16] reported increased activation in the left fusiform gyrus, right supramarginal gyrus, bilateral middle frontal gyrus (MFG), bilateral insula, and left middle temporal gyrus during facial WM tasks in schizophrenia patients. Conversely, Shefali et al. [17] reported reduced bilateral prefrontal lobe activation during similar tasks. Resting-state fMRI (rs-fMRI) studies exploring the neural mechanisms underlying facial WM deficits in patients with schizophrenia are rare.

Rs-fMRI, which assesses brain functional states at rest, reveals interactions among multiple functional networks critical for various cognitive functions, such as vision, hearing, language, and attention. Degree centrality (DC) measures the importance of nodes in these networks [18], with voxel DC positively correlated with functional network importance [19]. Wang et al. correlated WM deficits in schizophrenia patients with DC values in several frontal brain regions and clinical symptoms [20, 21]. Sebastian et al. linked WM deficits in schizophrenia patients to DC values in the intraparietal sulcus [22], albeit in tasks involving alphanumeric and graphic WM, not facial WM.

Given that facial WM likely engages multiple brain regions, few studies have explored interactions and functional connectivity between these regions in patients with schizophrenia. Granger causality analysis (GCA), which uses rs-fMRI, assesses how activity in one brain region influences another [23, 24]. Researchers have applied GCA to identify abnormalities in cognitive deficit-related brain networks in schizophrenia, such as impaired inhibition and collaboration between default mode and external attention networks [25], abnormal causal connections in networks of patients with verbal hallucinations [26], and abnormal connectivity in first-episode schizophrenia patients related to cognitive impairments [27].

This study aims to assess facial and house WM abilities in first-episode schizophrenia (FSZ) patients and HCs via a delayed matching-to-sample (DMTS) paradigm, analysing the neural mechanisms underlying facial WM deficits through the DC and GCA of rs-fMRI data. Additionally, correlations between functional/connectivity alterations in brain regions associated with facial WM deficits and clinical symptoms in schizophrenia patients were explored to elucidate the neural mechanisms of schizophrenia.

Participants

Patients with FSZ, either as outpatients or inpatients at the Affiliated Brain Hospital of Guangzhou Medical University, were enrolled in this study. Patient screening was conducted by two physicians with intermediate titles or higher. The inclusion criteria for the patient group were as follows: (1) meeting the DSM-IV diagnostic criteria for schizophrenia; (2) illness duration of less than 2 years and antipsychotic medication use for less than 6 months; (3) aged 18–50 years, Han Chinese ethnicity, and ≥ 9 years of education; (4) no electroconvulsive therapy (ECT) within the past six months; (5) biological parents of Han Chinese ethnicity; and (6) normal or corrected-to-normal vision in both eyes. The exclusion criteria included the following: (1) serious somatic or neurological comorbidities; (2) other mental disorders; (3) substance abuse or addiction (excluding nicotine); (4) history of epilepsy, febrile convulsions, or coma; and (5) contraindications to MRI.

HCs were recruited via posters, with the following inclusion criteria: (1) did not meet any Structured Interview for Psychosis Risk Syndromes (SIPS) criteria; and (2) were matched for sex, age, and education years with the patient group. The exclusion criteria were identical to those for the schizophrenia group.

Clinical evaluation

The severity of psychotic symptoms in patients was assessed via the Positive and Negative Syndrome Scale (PANSS) [28], a scale developed by Kay et al. [28] in 1987. The PANSS evaluates the presence and severity of psychotic symptoms across multiple domains and is categorized into a five-factor model proposed by Wallwork [29]. This model includes positive symptoms, negative symptoms, cognitive symptoms, excitement symptoms, and affective symptoms. The evaluation was conducted by two attending physicians or higher who underwent a conformance test.

Evaluation of WM abilities

In this study, the subjects were asked to complete a neutral face WM task and a house WM task constructed on the basis of the DMTS paradigm, which was the same as in a previous study [30]. The experimental procedure is shown in Fig. 1.

(Fig. 1)

At the end of the experiment, trials with reaction times longer than 3 s were excluded, and the average accuracy of each subject in the face and house WM tasks was calculated separately. Facial WM may simultaneously encompass multiple cognitive processes, such as basic visual processing, general object WM, and face-specific WM. In addition to directly comparing facial WM, we also compared face-specific WM indicators. First, a general linear regression model was used to predict the accuracy of facial WM on the basis of the accuracy of house WM. The residual between the predicted and actual values of facial WM accuracy was subsequently used as a specificity indicator to characterize facial WM ability, removing the general cognitive ability component included in the facial WM task accuracy measure. The face-specific accuracy (FSA) was expressed according to the following equations:

FSA = y-\(\:\widehat{y}\) (2 − 1)

\(\:\widehat{y}\) =ax + b (2–2)

The y in Eq. (2 − 1) represents the measured value of facial WM accuracy, and \(\:\widehat{y}\) represents the predicted value of facial WM accuracy as predicted by house WM accuracy. x in Eq. (2–2) represents the measured value of the accuracy of the house WM.

Rs-fMRI data acquisition

Rs-fMRI images of each subject were acquired using a Philips 3.0T MRI scanner. The scanning procedure was performed in the MRI room of the Affiliated Brain Hospital of Guangzhou Medical University. The scanning parameters of the rs-fMRI used were the same as those used in a previous study [31].

Rs-fMRI data preprocessing

Rs-fMRI images were preprocessed via the RESTPLUS toolkit [32]. Initially, the data were converted to the NIFTI format, followed by discarding the first 10 time points, slice timing correction, realignment for motion correction, and exclusion of subjects with translation > 2 mm or rotation > 2°. Spatial normalization via echo planar imaging (EPI) and Gaussian smoothing (8.0 mm FWHM) were applied, followed by detrending and regression against head motion parameters, global white matter signals, and cerebrospinal fluid signals. Pearson correlation analysis was used to compute dynamic connectivity (DC) across whole-brain voxels. Voxels with correlations > 0.2 were summed to yield weighted DC values, normalized by individual average DC values and smoothed quadratically before statistical analysis.

GCA followed similar preprocessing steps (format conversion, initial time point removal, slice timing correction, realignment, spatial normalization, smoothing, detrending, filtering, and regression). ROIs linked to facial WM accuracy guided whole-brain GCA via REST (version 1.8) software [33]. Path coefficient analysis at the voxel level quantified causal effects, with positive values indicating excitatory effects and negative values indicating inhibitory effects. Regression coefficients (β) computed the average ROI time series against whole-brain voxels, generating β maps of X-to-Y and Y-to-X for each subject. Fisher's Z transformation was used to normalize the β coefficients for subsequent statistical analyses of causal connections.

Statistical analysis

Data analysis was performed with SPSS 22.0. Descriptive statistics (means ± SDs) were used to characterize the measurements. Independent samples t tests and chi-square tests were used to assess group differences. Two × 2 two-factor analysis of variance (ANOVA) was used to analyse facial and house WM accuracy across patient and healthy control subgroups under low/high cognitive load conditions. Multiple regression was performed with DC values as the dependent variables, facial WM accuracy as the independent variable, and covariates (sex, age, and full-scale IQ [34]). False discovery rate (FDR) correction (P < 0.050, cluster > 10 voxels) was used to determine significant results in the voxelwise analyses. Correlation studies have employed Pearson's or Spearman's correlations on the basis of data distribution to explore associations between rs-fMRI findings and clinical symptoms. The test level α was less than 0.050, and a two-sided test was used.

Demographic and clinical characteristics of the subjects

The two groups showed no difference in general demographic features, including sex (χ² = 0.46, P = 0.500), age (t = 0.63, P = 0.528) and education (t = 0.67, P = 0.505). Compared with that of HCs, the IQ of FSZ patients was lower (t = 2.72, P = 0.009). The general demographics of the participants and the descriptive characteristics of the patients are shown in Table 1.

Table 1

Information on demographic, clinical characteristics of the subjects.
	Patients with FSZ (n = 28)	HCs (n = 33)	χ²/t	P
Gender (male/female)	16/12	16/17	0.46^b	0.500
Age (years)	25.21 (6.91)	24.24 (5.03)	0.63	0.528
Education (years)	11.11(3.40)	11.67 (3.11)	0.67	0.505
IQ	97.20 (13.43)	106.15 (12.09)	2.72	0.009
Duration of illness (months)	9.07 (7.09)	N/A	N/A	N/A
Antipsychotics drug use (mg)^a	198.95 (226.24)	N/A	N/A	N/A
PANSS_NEG	17.93 (4.63)	N/A	N/A	N/A
PANSS_DisCog	4.18 (1.76)	N/A	N/A	N/A
PANSS_POS	14.29 (4.13)	N/A	N/A	N/A
PANSS_EXC	7.54 (3.54)	N/A	N/A	N/A
PANSS_AnxDep	5.61 (2.45)	N/A	N/A	N/A
Values are presented as mean (s.d.). N/A, not applicable; PANSS, Positive and Negative Syndrome Scale; FSZ, First-episode schizophrenia; HCs, Healthy controls. PANSS_NEG, PANSS negative symptoms; PANSS_DisCog, PANSS Cognitive dysfunction symptoms; PANSS_POS, PANSS positive symptoms; PANSS_EXC, PANSS excitatory symptoms; PANSS_AnxDep, PANSS Anxiety/depression symptoms.
a equivalent to Chlorpromazine dosage[35]; b Chi-square test.

(Table 1)

Behavioural performance on the WM task

For the psychophysical experiments, only trials with a reaction time of less than 3 s were included in subsequent analyses. The accuracy of the two stimuli was then calculated as the ratio of the number of correct responses to the total number of trials. We focused on the difference in WM accuracy between the two groups for different stimuli. The accuracy of each condition is shown in Table 2 and Fig. 2.

Table 2

WM result of the subjects.
	Patients with FSZ (n = 28)	HC (n = 33)	t	P
Low-load face stimulus accuracy	0.79(0.12)	0.88(0.11)	3.17	0.002
High-load face stimulus accuracy	0.63(0.08)	0.70(0.07)	4.08	< .001
Low-load house stimulus accuracy	0.79(0.14)	0.91(0.10)	3.54	0.001
High-load house stimulus accuracy	0.65(0.08)	0.70(0.08)	2.69	0.009
face stimulus accuracy	0.71(0.09)	0.79(0.07)	3.98	< .001
house stimulus accuracy	0.72(0.11)	0.80(0.08)	3.47	0.001
Low-load FSA	-0.01(0.10)	0.01(0.10)	0.97	0.338
High-load FSA	-0.03(0.09)	0.02(0.10)	2.21	0.031
Values are presented as mean (s.d.). N/A, not applicable; FSZ, First-episode schizophrenia; HCs, Healthy controls; FSA, face special accuracy.

(Table 2)

(Fig. 2)

WM results for the facial and house stimulus

In Experiment 1, the subjects remembered the target and judged whether the stimulus that followed was the same as the target. We first performed 2 × 2 ANOVAs on the accuracy result. As shown in Fig. 2, the interaction between the condition load and the subject group was not significant (F = 0.22, P = 0.643), indicating that the facial WM accuracy of the two groups was not affected differently depending on the condition load. Significant main effects were observed for the subject group (F = 23.90, P < 0.001) and the condition load (F = 96.66, P < 0.001), which indicated a lower accuracy for FSZ participants and for trials with 2 items. Post hoc analysis revealed that accuracy was lower for FSZ individuals than for HCs under the two load conditions (low load: t = 3.17, P = 0.002; high load: t = 4.08, P < 0.001). Compared with HCs, independent samples t tests indicated that facial WM accuracy in patients with FSZ was lower (t = 3.98, P < 0.001).

In Experiment 2, we performed 2×2 ANOVAs on the WM results for the house stimulus. As shown in Fig. 2, the interaction effect between the condition load and subject group on accuracy was not significant (F = 2.30, P = 0.132). Significant main effects were observed for the subject group (F = 20.22, P < 0.001) and the condition load (F = 88.17, P < 0.001). Post hoc analysis revealed that accuracy was lower for FSZ individuals than for HCs under both condition loads (low load: t = 3.54, P = 0.001; high load: t = 2.69, P = 0.009). Compared with HCs, FSZ patients had lower house WM accuracy according to independent sample t tests (t = 3.47, P = 0.001).

FSA results

For the FSA results, we performed 2×2 ANOVAs first. The interaction effect between the condition load and the subject group on accuracy was not significant (F = 0.68, P = 0.413). Significant main effects were observed for the subject group (F = 4.94, P = 0.028). Post hoc analysis revealed that the FSA was lower in FSZ patients than in HCs at high loads (t = 2.21, P = 0.031) but not at low loads (t = 0.97, P = 0.338).

Correlation analysis between DC and facial WM in the FSZ group

In patients in the FSZ group, multiple regression analyses revealed that the DC values of the right MFG were associated with facial WM accuracy (P < 0.050, FDR correction). Detailed information on the brain regions associated with facial WM in FSZ patients is shown in Table 3.

Table 3

Correlation analysis result between DC and face WM of FSZ patients.
Brain region	Peak MNI coordinates			BA	t (Correlation)	Cluster size (voxels)	Patients with FSZ (n = 28)	HC (n = 33)	t (independent samples t-test)	P
Brain region	x	y	z	BA	t (Correlation)	Cluster size (voxels)	Patients with FSZ (n = 28)	HC (n = 33)	t (independent samples t-test)	P
Right MFG	48	9	36	N/A	5.18	14	1.27 (0.26)	1.44 (0.27)	2.56	0.013
MFG, Middle Frontal Gyrus.

(Table 3)

GCA of the right MFG in FSZ patients

The brain region associated with facial WM, the right MFG, was used as the ROI. The GCA results revealed that in FSZ patients, compared with HCs, the effective connectivity from the right MFG to the right inferior frontal gyrus (IFG) and the right thalamus as well as from the right postcentral gyrus to the right MFG was inhibited in the resting state (P < 0.050, FDR correction). Detailed information on the brain regions associated with GCA is shown in Table 4 and Fig. 3.

Table 4

Comparisons of GCA result of right MFG between two groups.
Brain region	Peak MNIcoordinates			BA	t	Cluster size (voxels)
Brain region	x	y	z	BA	t	Cluster size (voxels)
x→y
right IFG	54	21	0	N/A	-4.60	12
right thalamus	9	-18	0	N/A	-4.85	11
y→x
right postcentral	36	-36	57	3	-5.30	17
MFG, Middle Frontal Gyrus; IFG, inferior frontal gyrus.

(Table 4)

(Fig. 3)

Correlation analysis between MRI signals in the ROIs and clinical symptoms

The positive symptom scores, negative symptom scores and ROI signal values were normally distributed, so Pearson correlation analyses were used. The distributions of the cognitive symptom scores, excitement symptom scores, and affective symptom scores were skewed; thus, Spearman correlation analysis was used. Correlation analysis revealed that the DC values of the right thalamus were negatively correlated with negative symptom scores (r = -0.44, P = 0.018) and affective symptom scores (r = -0.57, P = 0.002). The results of the correlation analysis are shown in Fig. 4.

(Fig. 4)

In this study, the DMTS paradigm was employed to investigate deficits in facial and house WM among schizophrenia patients, revealing impairments in both domains. Analysis of the specific accuracy indicators for facial WM indicated that schizophrenia patients exhibit deficits in this area during high-load tasks, independent of their general visual WM abilities, but not during low-load tasks. Using rs-fMRI to explore the neural mechanisms underlying facial WM processing in schizophrenia patients, the right MFG DC value was found to be associated with facial WM. Focusing on the right MFG, the effective connectivity network for facial WM, including the right IFG, right thalamus, and right postcentral gyrus, significantly differed between patients and HCs. Within this network, the importance of the right thalamus was negatively correlated with negative and affective symptoms in schizophrenia patients.

The results of this study demonstrate impaired WM in schizophrenia patients, which is consistent with findings from previous research. Prior studies have consistently reported cognitive impairments across various domains in schizophrenia patients, with WM being one of the most commonly affected functions [36]. A review highlighted that neurocognitive functions, including WM, are impaired in both first-episode and chronic schizophrenia patients [37]. Another study revealed deficits in visual WM abilities in schizophrenia patients and their first-degree relatives, suggesting that dysfunctional neural mechanisms related to WM could be potential endophenotypic markers [38]. Schizophrenia patients and their high-risk siblings exhibit impairments in both verbal and nonverbal (facial) WM [39], indicating that WM may serve as a genetic vulnerability marker for schizophrenia susceptibility. She et al. [30], using the same DMTS paradigm, investigated WM deficits in facial and house recognition. Their findings revealed impaired WM under both stimulus types and load conditions, which is consistent with the results of this study. These findings collectively underscore that impaired WM is a distinct feature of schizophrenia and may be a core cognitive characteristic of the disorder.

This study used specific indicators of facial WM to discover that schizophrenia patients exhibit deficits in facial WM during high-load tasks, independent of their general visual WM ability. Research has shown that schizophrenia patients have impaired facial perception abilities, including identity recognition, facial memory, and facial emotion recognition [40]. Another study indicated that schizophrenia patients experience visual and cognitive processing difficulties with facial information, revealing significantly reduced accuracy in visual detection, moderately decreased perceptual discrimination, and notably impaired facial WM [10]. However, She et al. [30] analysed the performance of 18 FSZ patients and HCs in face and house DMTS tasks and suggested that stimuli may have little effect on visual WM deficits in FSZ patients. Possible reasons for these differing results include the small sample size of the previous study and the lack of separate comparisons of the two load conditions when analysing the influence of stimulus factors. Additionally, we used a general linear model to control for the effects of the general visual WM factor, which may also account for the different outcomes.

This study used the DC value measured by rs-fMRI to determine the correlation between right MFG DC values and facial WM deficits in schizophrenia patients. For rs-fMRI, Sebastian et al. [22] used DC to study visual WM in schizophrenia patients and reported that WM capacity was negatively correlated with two centrality measures in the right intraparietal sulcus. The results of our study differ, likely because of the different visual stimuli used in the WM tasks; our study focused on faces, whereas Sebastian's study used coloured squares. Research on the use of DC to explore facial WM deficits in patients with schizophrenia is limited. However, other face-related studies suggest a link between the MFG and facial memory. For example, one study noted reduced activation in the right MFG in schizophrenia patients during a 2-back WM task [41]. Another study revealed widespread frontal lobe hypoactivation, including in the MFG, during visual WM tasks in schizophrenia patients [38]. Additionally, several studies have reported differences in right MFG activation in first-degree relatives of schizophrenia patients during WM tasks [38, 42, 43]. Given that our study included FSZ patients, it is possible that right MFG abnormalities during WM tasks may serve as potential endophenotypic markers.

This study used GCA rs-fMRI data to explore the effective connectivity network of facial WM in FSZ patients. Compared with other effective connectivity methods that require a priori model assumptions (such as structural equation modelling and dynamic causal modelling), the greatest advantage of GCA is its data-driven modelling approach, which requires no prior knowledge and has a simple computational model [24]. Compared with those of HCs, the effective connectivity from the right MFG to the right inferior frontal gyrus (IFG) and the right thalamus as well as from the right postcentral gyrus to the right MFG was inhibited in FSZ patients. The effective connectivity from the right postcentral gyrus to the right MFG was reduced. Previous GCA studies have shown that abnormal MFG connectivity is associated with various symptoms of schizophrenia. Zhao et al. [44] reported reduced effective connectivity from the posterior cingulate cortex to the right MFG in schizophrenia patients experiencing verbal hallucinations. Feng et al. [45] reported that increased effective connectivity from the right dorsal dentate nucleus to the right MFG in patients with schizophrenia was associated with clinical symptoms. Jiang et al. [46] reported increased effective connectivity from the thalamus to the MFG in schizophrenia patients. Increased effective connectivity from the thalamus to the frontal cortex in schizophrenia patients may be related to clinical symptoms and cognitive function [47]. However, no studies have investigated the effective connectivity between the MFG and IFG or the postcentral gyrus. However, abnormal connectivity among the MFG, IFG, and postcentral gyrus is often mentioned in schizophrenia patients and can be used to identify schizophrenia [48]. In WM studies, abnormalities in the parietal and frontal lobes are associated with WM deficits in schizophrenia patients [36]. First-degree relatives of schizophrenia patients show abnormal activation in the frontal and parietal lobes and thalamus during WM tasks [43]. The MFG and postcentral gyrus are involved in the development of WM circuits in healthy adolescents [49]. In summary, this evidence enhances the reliability of the facial WM network in the schizophrenia patients investigated in this study.

In this study, we found that right thalamic DC values in schizophrenia patients were associated with negative and affective symptoms. The thalamus, a higher centre of sensation and the most important sensory relay station, is involved in various higher-order cognitive functions and mental activities, including sensory processing, attention, decision-making, and memory [50]. Previous studies have reported abnormal thalamic connectivity in schizophrenia patients with persistent negative symptoms [51]. Another study suggested that higher connectivity of the cerebellar-thalamus-cortical circuits at baseline significantly predicts poorer long-term reduction in negative symptoms [52]. Cheng et al. [53] reported that reduced thalamo-frontal functional connectivity (FC) in schizophrenia patients is associated with negative symptoms. In a study of patients with FSZ, connectivity between the right medial thalamus and the left middle temporal gyrus was associated with negative symptom scores [47]. On the other hand, disruption of thalamic activity may lead to incoordination of information transfer between regions involved in emotional processing and cognitive control [54]. Neuroimaging studies have shown structural and functional changes in the thalamus in patients with schizophrenia, which have been associated with affective disorders [55]. In a recent study, activation of the thalamus, a key brain region, was reduced during an emotion-sharing task in patients with schizophrenia [56]. A retrospective study suggested that structural and functional changes in the cortex and subcortex (e.g., the thalamus) may play a role in altering the interplay of affective and cognitive control in psychiatric disorders [57]. Additionally, it has been suggested that connections between the thalamus and sensorimotor cortex correlate with PANSS general psychopathology scores and that the PANSS affective symptom items are all part of the traditional PANSS scale of general psychopathology [58]. Overall, these findings contribute to a deeper understanding of the role of thalamic nodes in the pathophysiology of schizophrenia.

However, the limitations of this study should be noted. First, the small sample size may affect the reliability of the results. Second, the rs-fMRI results were not corrected via multiple comparison correction methods, which may impact the confidence of the findings. Finally, this study only used GCA for effective connectivity, and comparisons with other methods, such as structural equation modelling and dynamic causal modelling, are lacking. Therefore, further studies with larger, more representative samples with enriched research methodologies are needed to fully understand the resting-state neural mechanisms underlying impaired facial WM in patients with schizophrenia.

In summary, this study contributes to revealing the nature of facial WM deficits in patients with schizophrenia. Additionally, a neural network of effective resting-state connections for facial memory processing in these patients was preliminarily constructed. These neuroimaging findings may deepen our understanding of the underlying neural mechanisms of facial WM deficits in schizophrenia patients and provide a reference for the subsequent application of resting-state fMRI in the diagnosis, treatment, and intervention of schizophrenia.

Acknowledgements

Not applicable.

Authors’ contributions

Sumiao Zhou: Writing - original draft, Visualization, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization, Funding acquisition. Qijie Kuang: Writing - review & editing, Validation, Supervision, Project administration, Funding acquisition, Data curation. Huaqin Huang: Resources, Investigation, Data curation. Shenglin She: Validation, Supervision, Resources, Project administration, Data curation. Yingjun Zheng: Writing - review & editing, Validation, Supervision, Software, Project administration, Funding acquisition, Data curation. Xuanzi Li: Writing - review & editing, Validation, Supervision, Software, Resources, Funding acquisition, Data curation.

Funding

This study was funded by the Medical Scientific Research Foundation of Guangdong Province,China (A2024061, B202402), the Health Science and Technology Projects of Guangzhou (20241A011052) and the Planed Science and Technology Projects of Guangzhou (202201011338).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to partcipate

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national committees and with the Helsinki Declaration. The Institutional Review Board of the Affiliated Brain Hospital of Guangzhou Medical University approved the study, and the ethics approval number for our study is (2018) No. (002). The patients/participants provided written informed consent to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Huang Y, Wang Y, Wang H, Liu Z, Yu X, Yan J, Yu Y, Kou C, Xu X, Lu J, et al. Prevalence of mental disorders in China: a cross-sectional epidemiological study. Lancet Psychiatry. 2019;6(3):211–24.
Marder SR, Cannon TD, Schizophrenia. N Engl J Med. 2019;381(18):1753–61.
Kaminski J, Gleich T, Fukuda Y, Katthagen T, Gallinat J, Heinz A, Schlagenhauf F. Association of Cortical Glutamate and Working Memory Activation in Patients With Schizophrenia: A Multimodal Proton Magnetic Resonance Spectroscopy and Functional Magnetic Resonance Imaging Study. Biol Psychiatry. 2020;87(3):225–33.
Repovs G, Baddeley A. The multi-component model of working memory: explorations in experimental cognitive psychology. Neuroscience. 2006;139(1):5–21.
Baddeley A. Working memory. Science. 1992;255(5044):556–9.
Gur RE, Moore TM, Calkins ME, Ruparel K, Gur RC. Face Processing Measures of Social Cognition: A Dimensional Approach to Developmental Psychopathology. Biol Psychiatry Cogn Neurosci Neuroimaging. 2017;2(6):502–9.
Bruce V, Young A. Understanding face recognition. Br J Psychol. 1986;77(Pt 3):305–27.
Calkins ME, Gur RC, Ragland JD, Gur RE. Face recognition memory deficits and visual object memory performance in patients with schizophrenia and their relatives. Am J Psychiatry. 2005;162(10):1963–6.
Sachs G, Steger-Wuchse D, Kryspin-Exner I, Gur RC, Katschnig H. Facial recognition deficits and cognition in schizophrenia. Schizophr Res. 2004;68(1):27–35.
Chen Y, Norton D, McBain R, Ongur D, Heckers S. Visual and cognitive processing of face information in schizophrenia: detection, discrimination and working memory. Schizophr Res. 2009;107(1):92–8.
Conklin HM, Calkins ME, Anderson CW, Dinzeo TJ, Iacono WG. Recognition memory for faces in schizophrenia patients and their first-degree relatives. Neuropsychologia. 2002;40(13):2314–24.
Onitsuka T, Niznikiewicz MA, Spencer KM, Frumin M, Kuroki N, Lucia LC, Shenton ME, McCarley RW. Functional and structural deficits in brain regions subserving face perception in schizophrenia. Am J Psychiatry. 2006;163(3):455–62.
Lee CU, Shenton ME, Salisbury DF, Kasai K, Onitsuka T, Dickey CC, Yurgelun-Todd D, Kikinis R, Jolesz FA, McCarley RW. Fusiform gyrus volume reduction in first-episode schizophrenia: a magnetic resonance imaging study. Arch Gen Psychiatry. 2002;59(9):775–81.
Gobbini MI, Haxby JV. Neural systems for recognition of familiar faces. Neuropsychologia. 2007;45(1):32–41.
Li J, Liu J, Liang J, Zhang H, Zhao J, Rieth CA, Huber DE, Li W, Shi G, Ai L, et al. Effective connectivities of cortical regions for top-down face processing: a dynamic causal modeling study. Brain Res. 2010;1340:40–51.
Yoo SS, Choi BG, Juh RH, Park JM, Pae CU, Kim JJ, Lee SJ, Lee C, Paik IH, Lee CU. Working memory processing of facial images in schizophrenia: fMRI investigation. Int J Neurosci. 2005;115(3):351–66.
Walther S, Federspiel A, Horn H, Bianchi P, Wiest R, Wirth M, Strik W, Muller TJ. Encoding deficit during face processing within the right fusiform face area in schizophrenia. Psychiatry Res. 2009;172(3):184–91.
Zuo XN, Ehmke R, Mennes M, Imperati D, Castellanos FX, Sporns O, Milham MP. Network centrality in the human functional connectome. Cereb Cortex. 2012;22(8):1862–75.
Garcia-Garcia I, Jurado MA, Garolera M, Marques-Iturria I, Horstmann A, Segura B, Pueyo R, Sender-Palacios MJ, Vernet-Vernet M, Villringer A, et al. Functional network centrality in obesity: A resting-state and task fMRI study. Psychiatry Res. 2015;233(3):331–8.
Wang F, Liu Z, Ford SD, Deng M, Zhang W, Yang J, Palaniyappan L. Aberrant Brain Dynamics in Schizophrenia During Working Memory Task: Evidence From a Replication Functional MRI Study. Schizophr Bull. 2024;50(1):96–106.
Wang F, Liu Z, Yang J, Sun F, Cheng P, Pan Y, Cheng Y, Tan W, Huang D, Zhang J, et al. The neural compensation phenomenon in schizophrenia with mild attention deficits during working memory task. Asian J Psychiatr. 2024;97:104077.
Markett S, Reuter M, Heeren B, Lachmann B, Weber B, Montag C. Working memory capacity and the functional connectome - insights from resting-state fMRI and voxelwise centrality mapping. Brain Imaging Behav. 2018;12(1):238–46.
Birnbaum R, Weinberger DR. Functional neuroimaging and schizophrenia: a view towards effective connectivity modeling and polygenic risk. Dialogues Clin Neurosci. 2013;15(3):279–89.
Stephan KE, Friston KJ. Analyzing effective connectivity with functional magnetic resonance imaging. Wiley Interdiscip Rev Cogn Sci. 2010;1(3):446–59.
Pu W, Luo Q, Palaniyappan L, Xue Z, Yao S, Feng J, Liu Z. Failed cooperative, but not competitive, interaction between large-scale brain networks impairs working memory in schizophrenia. Psychol Med. 2016;46(6):1211–24.
Gao J, Zhang D, Wang L, Wang W, Fan Y, Tang M, Zhang X, Lei X, Wang Y, Yang J, et al. Altered Effective Connectivity in Schizophrenic Patients With Auditory Verbal Hallucinations: A Resting-State fMRI Study With Granger Causality Analysis. Front Psychiatry. 2020;11:575.
Huang H, Shu C, Chen J, Zou J, Chen C, Wu S, Xiao L, Liu Z, Wang H, Zhou Y, et al. Altered corticostriatal pathway in first-episode paranoid schizophrenia: Resting-state functional and causal connectivity analyses. Psychiatry Res Neuroimaging. 2018;272:38–45.
Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull. 1987;13(2):261–76.
Fong TC, Ho RT, Wan AH, Siu PJ, Au-Yeung FS. Psychometric validation of the consensus five-factor model of the Positive and Negative Syndrome Scale. Compr Psychiatry. 2015;62:204–8.
She S, Zhang B, Mi L, Li H, Kuang Q, Bi T, Zheng Y. Stimuli may have little impact on the deficit of visual working memory accuracy in first-episode schizophrenia. Neuropsychiatr Dis Treat. 2019;15:481–9.
Kuang Q, Zhou S, Li H, Mi L, Zheng Y, She S. Association between fractional amplitude of low-frequency fluctuation (falff) and facial emotion recognition ability in first-episode schizophrenia patients: a fmri study. Sci Rep. 2022;12(1):19561.
Jia XZ, Wang J, Sun HY, Zhang H, Liao W, Wang Z, Yan CG, Song XW, Zang YF. RESTplus: an improved toolkit for resting-state functional magnetic resonance imaging data processing. Sci Bull. 2019.
Song XW, Dong ZY, Long XY, Li SF, Zuo XN, Zhu CZ, He Y, Yan CG, Zang YF, Harrison BJ. REST: A Toolkit for Resting-State Functional Magnetic Resonance Imaging Data Processing. PLoS ONE 2011; 6.
Khodaei M, Laurienti P, Dagenbach D, Simpson S. Brain working memory network indices as landmarks of intelligence. Neuroimage Rep. 2023; 3.
Gardner DM, Murphy AL, O'Donnell H, Centorrino F, Baldessarini RJ. International consensus study of antipsychotic dosing. Am J Psychiatry. 2010;167(6):686–93.
Gebreegziabhere Y, Habatmu K, Mihretu A, Cella M, Alem A. Cognitive impairment in people with schizophrenia: an umbrella review. Eur Arch Psychiatry Clin Neurosci. 2022;272(7):1139–55.
Tschentscher N, Woll CFJ, Tafelmaier JC, Kriesche D, Bucher JC, Engel RR, Karch S. Neurocognitive Deficits in First-Episode and Chronic Psychotic Disorders: A Systematic Review from 2009 to 2022. Brain Sci 2023; 13(2).
Stablein M, Storchak H, Ghinea D, Kraft D, Knochel C, Prvulovic D, Bittner RA, Reif A, Oertel-Knochel V. Visual working memory encoding in schizophrenia and first-degree relatives: neurofunctional abnormalities and impaired consolidation. Psychol Med. 2019;49(1):75–83.
Brahmbhatt SB, Haut K, Csernansky JG, Barch DM. Neural correlates of verbal and nonverbal working memory deficits in individuals with schizophrenia and their high-risk siblings. Schizophr Res. 2006;87(1–3):191–204.
Soria Bauser D, Thoma P, Aizenberg V, Brune M, Juckel G, Daum I. Face and body perception in schizophrenia: a configural processing deficit? Psychiatry Res. 2012;195(1–2):9–17.
Li X, Yi ZH, Lv QY, Chu MY, Hu HX, Wang JH, Zhang JY, Cheung EEF, Chan RCK. Clinical utility of the dual n-back task in schizophrenia: A functional imaging approach. Psychiatry Res Neuroimaging. 2019;284:37–44.
Jiang S, Yan H, Chen Q, Tian L, Lu T, Tan HY, Yan J, Zhang D. Cerebral Inefficient Activation in Schizophrenia Patients and Their Unaffected Parents during the N-Back Working Memory Task: A Family fMRI Study. PLoS ONE. 2015;10(8):e0135468.
Zhang R, Picchioni M, Allen P, Toulopoulou T. Working Memory in Unaffected Relatives of Patients With Schizophrenia: A Meta-Analysis of Functional Magnetic Resonance Imaging Studies. Schizophr Bull. 2016;42(4):1068–77.
Zhao Z, Li X, Feng G, Shen Z, Li S, Xu Y, Huang M, Xu D. Altered Effective Connectivity in the Default Network of the Brains of First-Episode, Drug-Naive Schizophrenia Patients With Auditory Verbal Hallucinations. Front Hum Neurosci. 2018;12:456.
Feng S, Huang Y, Li H, Zhou S, Ning Y, Han W, Zhang Z, Liu C, Li J, Zhong L et al. Dynamic effective connectivity in the cerebellar dorsal dentate nucleus and the cerebrum, cognitive impairment, and clinical correlates in patients with schizophrenia. Schizophr Res. 2024.
Jiang Y, Duan M, Chen X, Zhang X, Gong J, Dong D, Li H, Yi Q, Wang S, Wang J, et al. Aberrant Prefrontal-Thalamic-Cerebellar Circuit in Schizophrenia and Depression: Evidence From a Possible Causal Connectivity. Int J Neural Syst. 2019;29(5):1850032.
Wang LN, Lin S, Tian L, Wu H, Jin WQ, Wang W, Pan WG, Yang CL, Ren YP, Ma X, et al. Subregional thalamic functional connectivity abnormalities and cognitive impairments in first-episode schizophrenia. Asian J Psychiatr. 2024;96:104042.
Potvin S, Giguere CE, Mendrek A. Functional Connectivity During Visuospatial Processing in Schizophrenia: A Classification Study Using Lasso Regression. Neuropsychiatr Dis Treat. 2021;17:1077–87.
Andre J, Picchioni M, Zhang R, Toulopoulou T. Working memory circuit as a function of increasing age in healthy adolescence: A systematic review and meta-analyses. Neuroimage Clin. 2016;12:940–8.
Roy DS, Zhang Y, Halassa MM, Feng G. Thalamic subnetworks as units of function. Nat Neurosci. 2022;25(2):140–53.
Zhu T, Wang Z, Zhou C, Fang X, Huang C, Xie C, Ge H, Yan Z, Zhang X, Chen J. Meta-analysis of structural and functional brain abnormalities in schizophrenia with persistent negative symptoms using activation likelihood estimation. Front Psychiatry. 2022;13:957685.
Cao H, Wei X, Hu N, Zhang W, Xiao Y, Zeng J, Sweeney JA, Lencer R, Lui S, Gong Q. Cerebello-Thalamo-Cortical Hyperconnectivity Classifies Patients and Predicts Long-Term Treatment Outcome in First-Episode Schizophrenia. Schizophr Bull. 2022;48(2):505–13.
Cheng W, Palaniyappan L, Li M, Kendrick KM, Zhang J, Luo Q, Liu Z, Yu R, Deng W, Wang Q, et al. Voxel-based, brain-wide association study of aberrant functional connectivity in schizophrenia implicates thalamocortical circuitry. NPJ Schizophr. 2015;1:15016.
Pauly K, Seiferth NY, Kellermann T, Backes V, Vloet TD, Shah NJ, Schneider F, Habel U, Kircher TT. Cerebral dysfunctions of emotion-cognition interactions in adolescent-onset schizophrenia. J Am Acad Child Adolesc Psychiatry. 2008;47(11):1299–310.
Pergola G, Selvaggi P, Trizio S, Bertolino A, Blasi G. The role of the thalamus in schizophrenia from a neuroimaging perspective. Neurosci Biobehav Rev. 2015;54:57–75.
Knobloch S, Leiding D, Wagels L, Regenbogen C, Kellermann T, Mathiak K, Schneider F, Derntl B, Habel U. Empathy in schizophrenia: neural alterations during emotion recognition and affective sharing. Front Psychiatry. 2024;15:1288028.
Duggirala SX, Schwartze M, Pinheiro AP, Kotz SA. Interaction of emotion and cognitive control along the psychosis continuum: A critical review. Int J Psychophysiol. 2020;147:156–75.
Anticevic A, Cole MW, Repovs G, Murray JD, Brumbaugh MS, Winkler AM, Savic A, Krystal JH, Pearlson GD, Glahn DC. Characterizing thalamo-cortical disturbances in schizophrenia and bipolar illness. Cereb Cortex. 2014;24(12):3116–30.

No competing interests reported.

Download PDF

Editor invited by journal
20 Sep, 2024
Reviews received at journal
11 Sep, 2024
Reviewers agreed at journal
11 Sep, 2024
Reviewers invited by journal
06 Sep, 2024
Editor assigned by journal
05 Aug, 2024
Submission checks completed at journal
05 Aug, 2024
First submitted to journal
01 Aug, 2024

You are reading this latest preprint version

Resting-state degree centrality and Granger causality analysis of facial working memory in patients with first-episode schizophrenia

Status:

Version 1

Abstract

Figures

Background

Materials and methods

Participants

Clinical evaluation

Evaluation of WM abilities

Rs-fMRI data acquisition

Rs-fMRI data preprocessing

Statistical analysis

Results

Demographic and clinical characteristics of the subjects

Behavioural performance on the WM task

WM results for the facial and house stimulus

FSA results

Correlation analysis between DC and facial WM in the FSZ group

GCA of the right MFG in FSZ patients

Correlation analysis between MRI signals in the ROIs and clinical symptoms

Discussion

Declarations

References

Additional Declarations

Status:

Version 1