Investigating Cortical Hierarchical Organization and its Role in Psychosis


The human brain is a remarkably complex and intricately organized organ, comprised of a multitude of cortical areas responsible for various cognitive functions. Converging evidence from tract-tracing, lesion, and physiological studies indicates that these cortical systems are organized along a hierarchical axis.

This hierarchical axis is anchored at one end by unimodal sensorimotor areas and at the other by transmodal association areas, creating a continuum that enables the integration of sensory signals and multimodal interoceptive and exteroceptive inputs. Disruptions to this cortical hierarchy have been linked to adverse cognitive and clinical outcomes, shedding light on its pivotal role in brain function.

Psychosis, a debilitating mental health condition characterized by a disconnection from reality, is thought to arise from impaired brain connectivity and a breakdown of higher-order cognitive processes. This article delves into the concept that disturbances in the hierarchical organization of cortical systems contribute to the pathophysiology of psychosis. Specifically, we will explore how impaired hierarchical signaling between cortical systems manifests in the symptoms of psychosis, focusing on sensory deficits and impairments in executive control and attention.

Unpacking the Hierarchical Organization

The hierarchical organization of cortical function can be effectively investigated using various dimensionality reduction techniques, which allow researchers to dissect inter-regional functional coupling (FC). Resting-state functional magnetic resonance imaging (fMRI) is a powerful tool for this purpose, providing insights into the dynamic interactions between different brain regions. Dimensionality reduction techniques, such as independent component analysis, can break down these complex networks into orthogonal axes of variation. These axes define smoothly varying gradients of regional FC profiles, ordered by the amount of variance they explain.

The Dominant Sensory-Fugal Gradient

The primary gradient, the most influential in this hierarchy, represents an axis extending from visual and other sensorimotor areas to the default mode network (DMN) and other transmodal regions.

This gradient aligns with classical descriptions of the sensory-fugal axis of cortical organization, supporting the integration of sensory signals with higher-order cognitive processes.

The Secondary Sensorimotor-Visual Gradient

The secondary gradient, on the other hand, spans from sensorimotor areas to visual areas. It serves to differentiate unimodal areas situated at approximately the same level in the cortical hierarchy. This gradient provides insights into how the brain distinguishes between different types of sensorimotor information and processes them accordingly.

Psychosis and Cortical Hierarchical Disruptions

Recent research has brought to light the fact that patients with schizophrenia exhibit a contracted primary sensory-fugal gradient. This suggests that there is a reduced functional differentiation between sensorimotor and transmodal association areas in these patients.

However, it remains unclear whether the secondary gradient, which delineates sensorimotor and visual areas, is similarly affected. Moreover, there is uncertainty regarding when these changes in hierarchical cortical organization manifest – whether they are present from the earliest stages of psychosis or emerge only after prolonged illness.

To address these questions, researchers have embarked on a comprehensive investigation of the cortical hierarchy in two independent cohorts of patients. One cohort comprises individuals in the early stages of psychosis, while the other includes patients with established schizophrenia. By comparing the primary and secondary FC gradients in these cohorts, researchers aim to determine whether disrupted hierarchical function persists across different stages of the illness.

Hypotheses and Expectations

Given the significant role ascribed to disruptions of hierarchical processing in the onset of psychotic symptoms, the researchers hypothesize that both cohorts, early psychosis, and established schizophrenia, will exhibit a disrupted organization of the dominant sensory-fugal gradient. This disruption would be characterized by a reduced functional differentiation between sensorimotor and transmodal association areas.


Disturbed connectivity between higher-order and sensorimotor systems is a recurring theme in the understanding of psychotic symptoms (13, 14, 24). In this study, diffusion map embedding was employed to characterize the gradients of sensory-fugal and visual-to-sensorimotor organization in the cerebral cortex of patients with early psychosis and established schizophrenia. The primary hypothesis, that the primary sensory-fugal gradient would be disrupted in both early psychosis and schizophrenia, was not supported. However, the findings did reveal disruptions in the secondary visual-to-sensorimotor gradient in individuals with schizophrenia. Visual regions along this gradient demonstrated reduced differentiation from opposing sensorimotor areas. This distinction was not observed in the early psychosis cohort. These results suggest that schizophrenia may be linked to a blurring of functional boundaries between different unimodal regions, a phenomenon that becomes evident only after the illness has progressed.

Primary and Secondary Axes of Cortical Organization in Schizophrenia

The evaluation of average gradients within each cohort uncovered broad similarities in the spatial patterning of the primary and secondary gradients. This consistency implies that neither early psychosis nor schizophrenia is associated with large-scale reorganizations of cortical function. This finding aligns with other clinical research that has demonstrated a consistent architecture of the primary and secondary gradients in both patient and control groups (43, 44). Consequently, it suggests that organizational disturbances in schizophrenia are subtle and do not induce significant changes in the primary modes of function, even when there are notable group differences along the gradient.

The contracted visual-to-sensorimotor gradient observed in schizophrenia patients implies that the disorder primarily affects this secondary gradient, rather than the primary sensory-fugal axis. While the primary sensory-fugal gradient is typically linked to interactions between unimodal and higher-order transmodal associative processes, deviations along the secondary visual-to-sensorimotor gradient might be associated with schizophrenia symptoms related to perceptual disturbances, such as hallucinations. This deviation in the gradient could be understood within the context of Mesulam’s hierarchical framework, as it blurs the boundaries between different sensory modalities within the same hierarchical level.

Impact on Macroscale Network Organization

These changes in gradient architecture in schizophrenia have a significant impact on macroscale network organization. Reduced between-network dispersion distinguishes the Visual network from other networks. There was also reduced within-network dispersion in the Visual, DAN (Dorsal Attention Network), Limbic, and Frontoparietal networks. Notably, the most pronounced effect was observed in the Visual network and DAN, suggesting that these networks have more homogenous functional connectivity profiles.

Previous neuroimaging studies exploring edge-level functional connectivity and resting-state activity in schizophrenia have identified similar changes in primary sensory regions, which are often linked to impaired visual and auditory processing (45, 46, 47). However, individuals with schizophrenia frequently struggle to integrate information from multiple opposing sensory modalities. These deficits in multisensory processing are believed to be due to dysfunction in both top-down and bottom-up processing, stemming from diminished cross-modal interactions across the primary sensory cortices (52). Reduced activity in both auditory and visual regions in individuals with schizophrenia could lead to shifts in group-level gradients, predominantly affecting the secondary visual-to-motor gradient, as functional connectivity becomes less differentiated among these primary sensory cortices.

Contrasting Findings and Possible Explanations

The findings of this study contrast with a prior study that reported disruptions in the first principal gradient in schizophrenia patients (24). This discrepancy might be attributed to differences in clinical stage and severity between the samples. Schizophrenia patients in this study scored higher on domains assessing positive symptoms than negative symptoms compared to the previous study. Given the association between disrupted cognition and negative symptoms, it is possible that such symptoms predict deviations along the sensory-fugal hierarchy, as higher-order thought propagation is closely linked to cortical hierarchical organization. Methodological differences in gradient construction may also contribute to this inconsistency, as the previous study applied Procrustes rotation to align subject-level gradients, while the current study used joint embedding. However, joint embedding preserves individual variations in gradient data better than Procrustes analysis, making it unlikely that joint embedding would mask any significant effects along the first gradient.

Gradient Architecture in Early Psychosis

The absence of gradient differences in early psychosis implies that abnormalities along the visual-to-sensorimotor axis may only emerge as the illness progresses. This may be due to heterogeneity within the early-phase psychosis cohort and the greater severity of brain changes associated with later stages of schizophrenia. Previous studies investigating early psychosis have found reduced functional connectivity within visual regions, often accompanied by perceptual anomalies. This finding appears to contradict the present study’s results. However, one possible explanation is that local edge-level differences are more sensitive to the current patient state and identify brain activity patterns related to present symptoms, whereas gradient-based approaches reflect trait-level global differences that are not readily modified by active symptoms.

Clinical Correlates of Disrupted Cortical Organization

In the CNP cohort, more severe schizophrenia symptoms were associated with atypical gradient architecture. Specifically, reduced network dispersion between most network pairs was associated with higher scores on clinical measures assessing general and positive symptoms, such as delusions. In contrast, elevated between-network dispersion was associated with higher scores on measures assessing negative symptoms, such as affective flattening and blunting.

Interestingly, the between-network dispersion values between pairs of higher-order and lower-order networks showed the strongest loadings of gradient metrics within the Canonical Correlation Analysis (CCA). These pairings mirrored the regional anchors of the primary sensory-fugal gradient, rather than the secondary visual-to-sensorimotor gradient. Although the present group-level analyses suggest the most significant shift in gradient architecture between healthy controls and individuals with schizophrenia occurs along the secondary gradient, dispersion metrics assess gradient architecture within a multidimensional space, incorporating information from both the primary and secondary gradients. This underscores the complexity of cognitive and behavioral processes in schizophrenia, which may be influenced by deviations along multiple axes of organization.

In conclusion, this study provides valuable insights into the cortical hierarchy and its relationship with psychosis, shedding light on the subtleties of gradient disruptions in schizophrenia. While the primary sensory-fugal gradient appears relatively unaffected, the secondary visual-to-sensorimotor gradient demonstrates significant changes, with potential implications for perceptual disturbances in the disorder. The findings also emphasize the role of gradient architecture in macroscale network organization and highlight the complexity of symptoms associated with different axes of organization in schizophrenia. This knowledge paves the way for further research and potential interventions in understanding and treating psychotic disorders.

reference link :


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.