Insula sub-regions across the psychosis spectrum: morphology and clinical correlates

ABSTRACT The insula is a heterogeneous cortical region, comprised of three cytoarchitecturally distinct sub-regions (agranular, dysgranular, and granular), which traverse the

anterior-posterior axis and are differentially involved in affective, cognitive, and somatosensory processing. Smaller insula volume is consistently reported in psychosis-spectrum disorders

and is hypothesized to result, in part, from abnormal neurodevelopment. To better understand the regional and diagnostic specificity of insula abnormalities in psychosis, their developmental

etiology, and clinical correlates, we characterized insula volume and morphology in a large group of adults with a psychotic disorder (schizophrenia spectrum, psychotic bipolar disorder)

and a community-ascertained cohort of psychosis-spectrum youth (age 8–21). Insula volume and morphology (cortical thickness, gyrification, sulcal depth) were quantified from T1-weighted

structural brain images using the Computational Anatomy Toolbox (CAT12). Healthy adults (_n_ = 196), people with a psychotic disorder (_n_ = 303), and 1368 individuals from the Philadelphia

Neurodevelopmental Cohort (PNC) (381 typically developing (TD), 381 psychosis-spectrum (PS) youth, 606 youth with other psychopathology (OP)), were investigated. Insula volume was

significantly reduced in adults with psychotic disorders and psychosis-spectrum youth, following an anterior-posterior gradient across granular sub-regions. Morphological abnormalities were

limited to lower gyrification in psychotic disorders, which was specific to schizophrenia and associated with cognitive ability. Insula volume and thickness were associated with cognition,

and positive and negative symptoms of psychosis. We conclude that smaller insula volume follows an anterior-posterior gradient in psychosis and confers a broad risk for psychosis-spectrum

disorders. Reduced gyrification is specific to schizophrenia and may reflect altered prenatal development that contributes to cognitive impairment. SIMILAR CONTENT BEING VIEWED BY OTHERS

MOLECULAR AND MICRO-ARCHITECTURAL MAPPING OF GRAY MATTER ALTERATIONS IN PSYCHOSIS Article Open access 12 September 2024 IN VIVO WHITE MATTER MICROSTRUCTURE IN ADOLESCENTS WITH EARLY-ONSET

PSYCHOSIS: A MULTI-SITE MEGA-ANALYSIS Article Open access 12 December 2022 CORTICAL PROFILES OF NUMEROUS PSYCHIATRIC DISORDERS AND NORMAL DEVELOPMENT SHARE A COMMON PATTERN Article 15

November 2022 INTRODUCTION The insula is a heterogeneous brain structure located deep in the Sylvian fissure and implicated in a vast array of human behaviors and mental experiences. Often

considered the fifth cerebral lobe, the insula is the first cortical region to develop and differentiate, starting at just six weeks of fetal life1. Due to its diverse role in a variety of

cognitive, affective, and regulatory functions, the insula is increasingly implicated in psychiatric disorders, with particular interest in its role in psychosis2. Insula abnormalities are

ubiquitous in psychotic disorders and include structural changes3, abnormal resting-state functional connectivity4,5, and altered task-based activation6,7. Of these, volume deficits are

among the most commonly reported. Despite apparent insula abnormalities in psychotic disorders, several questions remain. First, the anatomical specificity of these abnormalities remains

unclear. The insula is comprised of three granular sub-regions (agranular, dysgranular, and granular) distinguished by their cytoarchitectonic features, namely the presence/absence of

granular layer IV8,9. Yet, most studies in psychosis examine the whole insula10,11,12 or report relatively greater deficits in the anterior portion13,14. It is therefore unknown whether

granular sub-regions demonstrate distinct patterns of structural alterations in psychosis. Resolving this issue is important because, although strongly interconnected along the dorso-ventral

and rostro-caudal axes, these sub-regions support distinct functions across affective (agranular), cognitive (dysgranular), and somatosensory (granular) systems15. Therefore, investigating

sub-regions can help identify whether specific granular regions are more/less affected in psychosis, with implications for understanding structural changes underlying clinical phenotypes.

Relatedly, the diagnostic specificity of insula abnormalities is also unclear. Evidence of volume reductions in schizophrenia is fairly robust14 and some studies find abnormalities in

bipolar disorder, suggesting that insula dysfunction is transdiagnostic16. However, findings in bipolar disorder are mixed, possibly due to limited number of studies, small sample sizes, and

heterogeneous bipolar cohorts comprised of those with and without psychotic features17,18,19. In addition, the onset and trajectory of insula abnormalities are uncertain. Two distinct

approaches can be taken to address this question. First, examination of insula structure in high-risk or prodromal individuals as well as in adults with psychotic disorders ranging from the

first episode to chronic, can identify the presence of structural alterations at different stages of illness progression. Prior volumetric studies in prodromal and high-risk youth support

this developmental basis20,21. In high-risk youth, conversion to psychosis has been associated with smaller insula volume at baseline and a greater rate of insula volume loss over four

years20. Smaller insula volume is also observed in patients experiencing their first psychotic episode18,22. Meta-analyses of the whole brain identify smaller insula volume in high-risk,

first episode, and chronic psychosis compared to controls23,24, suggesting that insula volume changes observed in chronic schizophrenia patients are present early on in the disease course.

Yet, a comprehensive analysis of insula sub-regional structure at different illness stages has not before been investigated. Second, the developmental window of alteration is unknown;

characterization of different morphological features of the insula may point to perturbation of specific developmental stages, including prenatal brain development. The majority of prior

work on insula structure in psychosis has focused on volume, which is at its highest in childhood and declines throughout the lifespan25,26; however, parallel examination of multiple

structural measures may yield additional insights in the timing and etiology of insula abnormalities. Intriguingly, aspects of structural maturation, including thickness, gyrification, and

sulcal depth, appear to be driven by different biological processes27,28,29,30 with distinct developmental trajectories31 and genetic influences32, suggesting that insult at different

developmental stages could differentially impact structural metrics. Gyrification, for instance, develops prenatally (primarily third trimester)33 with relatively little change during

adolescence34. Sulcal depth increases considerably from birth to two years35, followed by relative stability or subtle decrease by adolescence36,37. Cortical thickness generally increases

over years one to six, in line with ongoing development of dendrites and glial cells in childhood, then begins to decrease by age eight in the context of synaptic pruning37,38. While

relatively understudied compared to volume, cortical thinning39,40,41, abnormal gyrification40,42, and reduced sulcal depth43 of the insula have been observed in small samples of psychotic

disorder participants using whole brain approaches. Few studies, however, have characterized these structural features in the same cohort, limiting conclusions about the specificity of

structural alterations in psychosis. Furthermore, whether morphometric measures of the insula (e.g., thickness, gyrification, and sulcal depth) are altered in psychosis-spectrum youth has

never before been examined. Therefore, in addition to providing a detailed account of insula structure across the psychosis spectrum, investigation of different structural metrics has

implications for understanding the developmental etiology of insula alterations. Finally, elucidating insula abnormalities across the psychosis spectrum is clinically relevant to psychosis

phenotypes2,11,44. As previously noted, insula sub-regions are distinctly involved in affective, cognitive, and somatosensory processing15,45, suggesting that sub-regional alterations may

differentially contribute to psychotic symptoms46. Each sub-region has distinct afferent and efferent connections47, recapitulated in their functional and structural connectivity

profiles48,49. Studies examining insula functioning have described large anterior and posterior clusters45 that can be further sub-divided based on their involvement in distinct behavioral

domains50. Most anterior, the agranular insula has reciprocal connections with limbic regions and is implicated in emotion processing and social cognition50. We, therefore, hypothesized it

would contribute to negative symptoms, as recently demonstrated using functional connectivity4. The middle dysgranular sub-region receives projections from the cingulate cortex and striatum

and projects to parietal and prefrontal cortices, supporting its involvement in higher-order cognition (e.g., memory, attention)4,51. Relationships with overall cognitive ability were

therefore investigated. Finally, the most posterior granular sub-region is considered a “primary interoceptive cortex” with somatotopic representations of objective physiological changes47

and involvement in perception. We, therefore, expected structural abnormalities of this region to show relationships with positive symptoms. The current study quantified volume and

morphological characteristics of insula sub-regions in two independent cohorts: adults with psychotic disorders and a community-ascertained developmental cohort52 that includes youth with

psychosis-spectrum symptoms. We addressed the following aims: (1) characterize volume and surface-based morphology of insula sub-regions across the psychosis spectrum, (2) determine

diagnostic specificity of these alterations, and (3) examine contributions to clinical phenotypes. METHODS AND MATERIALS PARTICIPANTS Two independent cohorts were used to examine insula

structure in adults with a psychotic disorder (Psychosis Cohort) and psychosis-spectrum youth (PNC). _Psychosis Cohort:_ The Psychosis Cohort came from a repository study of 593 individuals

that participated in one of three neuroimaging projects conducted in the Department of Psychiatry and Behavioral Sciences at Vanderbilt University Medical Center (VUMC). After excluding

individuals that did not meet our study criteria, including neuroimaging quality assurance, our final sample (Table 1) included 196 healthy individuals and 303 individuals with a primary

psychotic disorder (schizophrenia spectrum = 211; bipolar disorder with psychotic features = 92). Psychosis symptoms and cognitive ability were measured using the Positive and Negative

Syndrome Scale (PANSS)53 and the Screen for Cognitive Impairment in Psychiatry (SCIP)54, respectively. The SCIP includes measures of verbal memory (immediate and delayed), working memory,

verbal fluency, and processing speed and has been shown to be a reliable and valid measure of cognitive ability in psychotic disorders55. SCIP subtest raw scores were converted to z-scores

using normative data and averaged to create a composite z-score54. Symptom severity was assessed using the PANSS, which rates positive, negative, and general psychopathology symptoms over

the past two weeks56. Average positive and negative scale scores were used to measure positive and negative symptom severity. _Philadelphia Neurodevelopmental Cohort (PNC):_ The PNC was

obtained from the database of Genotypes and Phenotypes (dbGaP: Study Accession phs00607.v3.p2), and consists of 9498 youth aged 8–21. Of the 1601 participants that completed a neuroimaging

session57, 1368 were included in the current study (Table 1) after excluding participants that did not meet our inclusion criteria, including neuroimaging quality assurance. Participants

were classified as typically developing (TD) (_n_ = 381), psychosis-spectrum (PS) (_n_ = 381), or other psychopathology (OP) (_n_ = 606) using similar procedures as prior PNC studies58. OP

participants had suprathreshold psychopathology symptoms but did not meet PS criteria. Positive and negative symptoms were assessed using the PRIME Screen-Revised and Scale for Prodromal

Syndromes52. Cognition was assessed using the Penn Computerized Neurocognitive Battery (CNB), which includes measures of executive-control, episodic memory, complex cognition, and social

cognition59. Overall cognitive ability (average z-score of accuracy on each cognitive test) was used as a measure of cognition. Details on study procedures and group assignment are included

in the Supplement. All four studies (three neuroimaging projects at VUMC and the PNC study) were approved by the Institutional Review Board (IRB) at each site and all participants provided

signed written informed consent prior to completing study procedures. NEUROIMAGING DATA ACQUISITION Imaging data for the Psychosis Cohort were collected on one of two identical 3.0-T Philips

Intera Achieva MRI scanners (Philips Healthcare, Andover, MA), located at the Vanderbilt University Institute of Imaging Science. A T1-weighted anatomical image was collected on all

subjects and visually inspected for quality, blinded to diagnosis. The Psychosis Cohort was collected across three neuroimaging studies with slightly different scanning parameters (e.g.,

minor differences in TR/TE), and somewhat different inclusion criteria (e.g., one study emphasized recruitment of early psychosis patients while another included a large proportion of the

bipolar participants). Therefore ‘study’ was included as an additional covariate in all neuroimaging analyses of the Psychosis Cohort. Imaging data for the PNC was collected on a Siemens TIM

Trio scanner located at the Hospital of the University of Pennsylvania. A T1-weighted scan was obtained using a magnetization-prepared rapid acquisition gradient echo (MPRAGE) sequence.

Additional information on neuroimaging data acquisition and quality assurance are available in the Supplement. INSULA VOLUME AND SURFACE FEATURES Brain volumes and cortical surface features

(i.e., thickness, gyrification, sulcal depth) were quantified using voxel-based morphometry (VBM) and surface-based morphometry (SBM), respectively, as implemented in the Computational

Anatomy Toolbox 12 (CAT12) for Statistical Parametric Mapping version 12 (SPM12)60. _Voxel-Based Morphometry_: T1 images were pre-processed in CAT12 using recommended processing procedures

which include: correction for bias-field inhomogeneities, spatial registration to a reference brain using linear (12-parameter affine) and non-linear transformations, segmentation into gray

matter, white matter, and cerebrospinal fluid tissue classes, and spatial normalization to MNI-space using the DARTEL algorithm. The normalized, modulated gray matter segmentations were used

in the statistical analyses described below. _Surface-Based Morphometry_: The CAT12 toolbox uses the projection-based thickness (PBT) method to automatically reconstruct the cortical

surface61. Following surface reconstruction, morphometric indices were calculated, including cortical thickness, gyrification, and sulcal depth. These metrics provide complimentary and

non-overlapping features of insula structure. Cortical thickness estimates the thickness of the gray matter surface using segmentation of the distance between inner and outer surfaces of

white and gray matter boundaries62. Gyrification was estimated using a curvature-based approach to measure absolute mean curvature, with greater cortical folding contributing to greater

local gyrification, as described by Luders and colleagues63. Sulcal depth was estimated based on the Euclidean distance between the central surface and its convex hull, using square-root

transformed values. As recommended by the developers, cortical thickness data was resampled and smoothed at 15 mm FWHM, whereas gyrification and sulcal depth data were resampled and smoothed

at 20 mm FWHM. The distance between sulci and gyri is approximately 20–30 mm, so a 20 mm smoothing kernel enhances the features within this range63. NEUROIMAGING STATISTICAL ANALYSIS Volume

and surface features of the insula were analyzed using complimentary region-of-interest (ROI) and voxel/vertex-wise approaches. _ROI Analysis:_ A-priori masks of agranular, dysgranular, and

granular sub-regions, as determined by Farb and colleagues64, were used to extract insula sub-region volumes from each individuals normalized, modulated gray matter segmentation65. The Farb

atlas was chosen due to its use of well-characterized cytoarchitectonic divisions9 to identify granular sub-regions. This allowed, not only for the examination of structurally defined

sub-regions, but also limited the use of maps based on functional activation or connectivity, which could be less reliable in schizophrenia5. The HCP-MMP1 atlas66, a surface-based atlas

which includes insula sub-regions was used to extract average thickness, gyrification, and sulcal depth from the surface meshes for insula sub-regions. Volume and surface features were

analyzed separately in SPSS v.26. Repeated-measures ANCOVA was conducted in each cohort with insula sub-region as a repeated factor and diagnostic group as a between-groups factor. Age,

gender, total intracranial volume (TIV), study (for the psychosis cohort only), and topography defects (SBM analyses only) were included as covariates. Significant main effects of the group

were followed-up with analysis of group differences for each sub-region using multivariate ANCOVA. Significance was Bonferroni-corrected for three sub-regions (critical alpha = 0.017).

Within the Psychosis Cohort, effects of diagnosis (schizophrenia-spectrum and bipolar disorder) and illness stage (early and chronic psychosis) were also examined using repeated-measures

ANCOVA as described above. Finally, group differences in gyrification and sulcal depth using the sub-region approach were supplemented by examination of the whole insula using the

Desikan-Killiany (DK40) atlas parcellation67. This was done to ensure sufficient area to measure cortical folding, in order to help validate sub-regional analyses. _Voxel/vertex-wise

analyses_: Voxel-wise analysis of insula volume was performed in SPM12 by entering the modulated gray matter images (smoothed with a 4 mm kernel) into an ANCOVA, masked to include only

voxels in the a-priori defined whole insula mask64. Similarly, vertex-based analysis of cortical thickness, gyrification, and sulcal depth was analyzed using one-way ANCOVAs for each surface

metric within CAT12, using the HCP-MMP1 whole insula as an explicit mask. Age, gender, TIV, study (Psychosis cohort only), and topological defects (SBM analysis only) were included as

covariates. The resultant statistical parametric maps were thresholded at cluster-level corrected _p_FWE = 0.05 for voxel-wise _p_ < 0.001 uncorrected. ASSOCIATIONS WITH COGNITIVE AND

CLINICAL PHENOTYPES In the psychosis cohort, relationships between structure and cognition were assessed in the whole sample controlling for the group (control, schizophrenia, and bipolar),

whereas relationships with symptoms were assessed in the psychosis patients only. In the PNC, all correlations were conducted across the entire sample, controlling for group. A-prior

associations were first tested: agranular insula and negative symptom severity, dysgranular insula and cognitive ability, granular insula and positive symptom severity. To examine

specificity, two-tailed a-priori correlations were followed-up with examination of all pairwise brain-behavior relationships. Associations were tested using partial correlation with group,

age, gender, TIV, and topography defects (for SBM measures) as covariates. A-priori correlations were Bonferroni-corrected for four metrics (critical alpha = 0.013). All other correlations

are Bonferroni-corrected for the additional 8 tests (critical alpha = 0.006). RESULTS GROUP DIFFERENCES Structural differences across sub-regions (main effect of sub-region) are presented in

Table S1. PSYCHOSIS COHORT _Volume:_ In the ROI-analysis, we observed a significant main effect of group (F(1,493) = 16.94, _p_ < 0.001; 4.28% smaller) and group by sub-region

interaction (F(2,986) = 3.66, _p_ = 0.03) (Fig. 1). Follow-up analyses indicated that while all insula sub-regions were smaller in psychosis (all _p_-values ≤ 0.001), there was an

anterior–posterior gradient. Specifically, reduced insula volume compared to healthy controls was most pronounced in the agranular (5.22% smaller), then dysgranular (4.48% smaller) then

granular sub-regions (3.56% smaller). These findings were recapitulated in our voxel-wise analysis, which revealed large bilateral clusters of reduced insula volume in psychosis spanning all

sub-regions (Fig. 2; Table S2). _Cortical Thickness_: There was no significant main effect of group (F(1,492) = 0.40, _p_ = 0.53) or group by sub-region interaction (F(2,984) = 2.37, _p_ = 

0.09) in the ROI-analysis. Vertex-based analysis revealed a cluster in the agranular insula [−34 31 −2] with significantly reduced cortical thickness in psychosis (Fig. 2; Table S2).

_Gyrification:_ In the ROI-analysis, gyrification was significantly reduced in psychosis (F(1,492) = 6.51, _p_ = 0.01) (Fig. 1) with a non-significant group by sub-region interaction

(F(2,984) = 2.46, _p_ = 0.09). Using the DK40 atlas, whole insula gyrification was still significantly reduced in psychosis (F(1,492) = 8.81, _p_ = 0.003). Vertex-based analysis identified

two clusters with reduced gyrification in psychosis, with one peak in the right dysgranular insula [36 −21 3] and the other peak in the left granular insula [−35 −9 11] (Fig. 2; Table S2).

_Sulcal Depth:_ No differences were detected between psychosis and healthy participants in either the ROI-analysis (F(1,492) = 0.00, _p_ = 0.99) or vertex-based analysis (no significant

clusters). When using the DK40 atlas, whole insula sulcal depth did not differ between groups (F(1,492) = 0.03, _p_ = 0.854). In the sub-regional ROI-approach, there was a significant group

by region interaction (F(2,984) = 4.82, _p_ = 0.008). Groups did not differ for any insula sub-region (_p_’s > 0.26), however, healthy subjects had relatively greater granular sulcal

depth while psychosis participants had relatively smaller granular sulcal depth (Fig. 1). In both groups, all sub-regions significantly differed from one another (_p_’s < 0.001), with

granular insula having the greatest sulcal depth and agranular insula having the smallest sulcal depth in both groups. DIAGNOSTIC EFFECTS (SCHIZOPHRENIA-SPECTRUM AND PSYCHOTIC BIPOLAR

DISORDER) Repeated-measured ANCOVA revealed that the structural abnormalities reported in the psychosis sample were specific to schizophrenia and not observed in psychotic bipolar disorder

(Full results are presented in Table S3). Regarding volume, all three sub-regions were smaller in schizophrenia compared to bipolar participants (_p_’s < 0.04). Bipolar participants did

not significantly differ from controls in any sub-region. Analysis of morphometry revealed that lower dysgranular gyrification was also specific to schizophrenia. When compared with

controls, schizophrenia (1.87% reduced, _p_ = 0.002) but not bipolar (0.72% reduced, _p_ = 0.37) participants showed significant reductions. The same was true for DK40 whole insula

gyrification, which was significantly reduced in schizophrenia (_p_ = 0.001), but not bipolar disorder (_p_ = 0.43) (bipolar patients were intermediate to schizophrenia and healthy

controls). Cortical thickness did not differ across the three groups (F(2,491) = 2.06, _p_ = 0.13). Sulcal depth was slightly increased in bipolar participants as compared with controls

(0.71% greater, _p_ = 0.03) and schizophrenia participants (1.01% greater, _p_ = 0.002), which was true also when considering the DK40 whole insula (bipolar vs. controls: _p_ = 0.05; bipolar

vs. schizophrenia: _p_ = 0.003). PNC COHORT _Volume:_ We observed a significant main effect of group (F(2,1362) = 4.23, _p_ = 0.015) and a non-significant group by region interaction

(F(4,2724) = 1.11, _p_ = 0.35). Similar to psychosis patients, PS youth demonstrated an anterior-posterior gradient with the largest volume reduction in the agranular (1.46% smaller), then

dysgranular (1.38% smaller), then granular (1.03% smaller) sub-regions when compared with TD youth (Fig. 1). The same pattern was seen when PS youth were compared with OP (agranular: 1.6%

smaller; dysgranular: 1.07% smaller; granular: 0.86% smaller). TD and OP youth had similar overall insula volumes (_p_ = 0.81). In the voxel-wise analysis, a cluster localized to the

agranular sub-region [38 15−10] (33 voxels) was reduced in PS youth; however, it fell short of our conservative correction for multiple comparisons (_p_FWE = 0.06; Supplementary Fig. 1;

Table S2). _Structure-Based Morphometry:_ There were no differences between groups in insula cortical thickness (F(2,1361) = 0.04, _p_ = 0.97), gyrification (F(2,1361) = 0.29, _p_ = 0.75),

or sulcal depth (F(2,1361) = 1.48, _p_ = 0.23). There were also no significant region by group interactions (thickness: F(4,2722 = 0.32, _p_ = 0.87); gyrification: F(4,2722) = 2.03, _p_ = 

0.09); sulcal depth: F(4,2722) = 1.62, _p_ = 0.17)). No group differences were observed for DK40 whole insula gyrification (F(2,1361) = 0.18, _p_ = 0.836) or sulcal depth (F(2,1361) = 1.81,

_p_ = 0.16). Results were similar for the vertex-based analysis (Fig. 2). ASSOCIATIONS WITH COGNITIVE AND CLINICAL PHENOTYPES Group differences between the cognitive and clinical phenotypes

are presented in Table 1. All brain–behavior relationships are presented in Table S4 and summarized below. Within the psychosis cohort, we also examined group differences between bipolar and

schizophrenia-spectrum participants. Schizophrenia participants had significantly worse cognitive ability (F(1, 262) = 9.12, _p_ = 0.003), positive (F(1, 265) = 6.70, _p_ = 0.01) and

negative (F(1, 265) = 36.98, _p_ < 0.001) symptoms as compared with psychotic bipolar disorder participants. AGRANULAR STRUCTURE-NEGATIVE SYMPTOMS Agranular insula volume (_r_ = −0.24,

_p_ < 0.001) and thickness (_r_ = −0.23, _p_ < 0.001) were significantly associated with negative symptoms in psychotic disorders, but not in the PNC (_p_’s > 0.05). DYSGRANULAR

STRUCTURE-COGNITIVE ABILITY Overall cognitive ability was significantly associated with volume of the dysgranular insula in both the Psychosis Cohort (_r_ = 0.17, _p_ < 0.001) and PNC

(_r_ = 0.12, _p_ < 0.001) (Fig. 3). In the Psychosis Cohort, dysgranular thickness (_r_ = 0.19, _p_ < 0.001) and gyrification (_r_ = 0.12, _p_ = 0.011) were also associated with

cognition. In the PNC, greater dysgranular sulcal depth was associated with better cognitive ability (_r_ = 0.13, _p_ < 0.001) but this relationship was not seen in the Psychosis Cohort

(_r_ = 0.014, _p_ = 0.76). GRANULAR STRUCTURE-POSITIVE SYMPTOMS Positive symptom severity was associated with granular insula thickness (_r_ = −0.19, _p_ = 0.001) in psychosis patients, but

not in PS youth (_r_ = 0.01, _p_ = 0.83). No other significant relationships were observed. SUB-REGION AND BEHAVIOR SPECIFICITY In addition to the above-reported relationships, we also found

that cognitive ability was associated with agranular volume in both the Psychosis Cohort (_r_ = 0.22, _p_ < 0.001) and the PNC (_r_ = 0.11, _p_ < 0.001). In the Psychosis Cohort only,

cognition was also related to agranular thickness (_r_ = 0.19, _p_ < 0.001), granular volume (_r_ = 0.21, _p_ < 0.001) and granular thickness (_r_ = 0.15, _p_ < 0.001) indicating

that volume and thickness of the entire insula was associated with cognition. Negative symptoms were also associated with volume of the dysgranular (_r_ = −0.18, _p_ = 0.002) and granular

(_r_ = −0.18, _p_ = 0.002) sub-regions, indicating relationships between whole insula volume and negative symptoms. Positive symptoms were also associated with agranular thickness (_r_ = 

−0.17, _p_ = 0.006). In the PNC, cognitive ability was related to sulcal depth of the agranular (_r_ = 0.15, _p_ < 0.001) and granular (_r_ = 0.09, _p_ < 0.001) sub-regions. Positive

symptoms were related to agranular gyrification (_r_ = 0.08, _p_ = 0.004). SUPPLEMENTAL ANALYSES Supplemental analyses were conducted to examine associations between insula structure,

antipsychotic medication, and illness stage in psychosis, as well as interactions with age and sex. These findings are presented in the Supplement (see Supplementary Information, including

Figs. S2 and S3 and Tables S5–S8). In brief, both early and chronic patients demonstrated smaller insula volumes and lower gyrification as compared with healthy controls. Volume reductions

followed an anterior-posterior gradient in both groups. Furthermore, age and sex effects were detected and are described in detail in the Supplement. Briefly, in both cohorts, volume and

sulcal depth were negatively associated with age for all insula sub-regions. Group by age interactions were non-significant except for a group by age interaction for dysgranular gyrification

(_r_ = 0.14, _p_ = 0.002) in the Psychosis Cohort, driven by a slightly negative relationship in the psychosis patients but a zero-slope relationship in the healthy controls. Sex effects

were largely non-significant in both cohorts, except for slightly greater granular gyrification in men within the Psychosis Cohort (_p_ = 0.003) and slightly greater dysgranular volume (_p_ 

= 0.007), dysgranular thickness (_p_ < 0.001), and agranular gyrification (_p_ = 0.001) in boys in the PNC. In addition, within the Psychosis Cohort, chlorpromazine (CPZ)-equivalence

values did not significantly correlate with any structure metrics (Table S9) suggesting that group differences were unlikely to be driven by medication use. Finally, to mitigate the

potential impact of demographic variables (age, gender, race) on the PNC findings, a matched sample of TD and PS youth were identified (Table S10). Insula volume was lower in the PS youth to

a similar degree as was observed in the larger sample, but the findings in this smaller sample were not significant (_p_ = 0.22). DISCUSSION We examined insula sub-regional volume,

morphology, and clinical correlates in adults with psychotic disorders (schizophrenia-spectrum and psychotic bipolar disorder) and youth with psychosis-spectrum symptoms. We found that

insula volume is smaller in psychotic disorder patients and psychosis-spectrum youth, and is associated with cognitive impairment in both cohorts. Investigation of insula sub-regions

revealed that smaller insula volume follows an anterior-posterior gradient across the psychosis spectrum, with smaller volume being more pronounced in anterior insula. Parallel analysis of

surface-based morphometry identified lower gyrification in psychotic disorders, with limited evidence of reduced cortical thickness or sulcal depth. Volume and gyrification reductions were

observed only in schizophrenia, not psychotic bipolar disorder. Gyrification was primarily reduced in the dysgranular sub-region, correlating with cognitive deficits. Finally, positive and

negative symptoms were associated with smaller volume and thickness of multiple granular sub-regions in psychotic disorders but not psychosis-spectrum youth. Together, these findings suggest

insula volume loss as a neurodevelopmental vulnerability that confers broad risk for psychosis-spectrum disorders and is associated with clinical phenotypes, especially cognitive

impairment. Insula morphometry, on the other hand, is more specific to reduced gyrification in schizophrenia, pointing to a possibly early neurodevelopmental abnormality in this population.

While prior work has reported on insula volume loss in psychosis, few studies have considered the heterogeneity of insula structure. Cytoarchitecture of the insula changes along the

anterior-posterior axis, transitioning from an agranular (anterior) to a granular (posterior) neocortex with a large intermediate dysgranular region68. Integration along this axis is

believed to underlie the neurobiology of subjective feelings69. Although volumetric studies point to whole insula reductions in psychotic disorders, meta-analysis shows relatively smaller

volume in anterior than posterior regions14. We extend these findings by demonstrating an anterior-posterior gradient of smaller insula volume in psychotic disorders, with the most

pronounced reductions (relative to healthy controls) in the agranular insula, followed by the dysgranular and granular sub-regions. We also demonstrate that this pattern of volume

differences is observed in youth with psychosis-spectrum symptoms, early and chronic psychosis. These findings suggest, in a cross-sectional sample, that smaller insula volume has origins in

early adolescence that confers risk for psychosis-spectrum disorders. Intriguingly, volume deficits were not observed in youth with other psychopathology (e.g., depression, anxiety

disorders) but were instead specific to psychosis-spectrum youth. Prior work has demonstrated smaller insula volume in adults with non-psychotic psychiatric disorders3, and insula

dysfunction is implicated in various psychiatric conditions70. Our findings suggest that, during adolescence, insula volume abnormalities may be specific, or at least most pronounced, in

individuals with psychosis-spectrum symptoms. Insula structural alterations were also more pronounced in schizophrenia-spectrum than psychotic bipolar disorder participants. Prior research

on insula pathology in bipolar disorders has been mixed12,16,17,71. Unlike prior work in heterogeneous bipolar samples, all bipolar participants in our cohort had experienced psychosis.

Therefore, we provide additional evidence that those with schizophrenia-spectrum disorders have notable alterations to insula structure that may be distinct from individuals with affective

psychosis. This finding also speaks to the weaker structural differences observed in PS youth, many of whom may never convert to schizophrenia-spectrum disorder or develop bipolar

disorder72. On-going longitudinal follow-up studies of the PNC will further clarify the specificity of insula abnormalities. Consistent with previous reports, we also observed reduced insula

gyrification in psychotic disorders, particularly schizophrenia17,73,74. Follow-up analyses revealed reduced gyrification in both early and chronic psychosis, with the most marked deficits

in the dysgranular sub-region. Gyrification develops prenatally75 and is relatively stable over childhood and adolescence76. In youth at clinical-high risk for psychosis, gyrification has

been shown to be stable over one year, suggesting that abnormal gyrification reflects very early developmental changes77. Gyrification is particularly intriguing in the context of

schizophrenia, as cortical folding is closely linked with the development of neural connectivity patterns78. In fact, altered gyrification has been put forth as foundational to the global

connectivity abnormalities in schizophrenia thought to sub-serve clinical phenotypes79. Recently, we have shown altered functional connectivity of insula sub-regions during resting-state in

an overlapping sample of schizophrenia participants4. Functional differentiation of insula connectivity in schizophrenia is also altered along the anterior-posterior axis5. While

speculative, our findings support a neurodevelopmental aberrant connectivity model of schizophrenia80 with early, possibly prenatal, origins. Our data further suggests that reduced

dysgranular gyrification may contribute to the very early manifestation of cognitive deficits observed in those who develop schizophrenia81. Cortical thickness, on the other hand, was not

widely reduced. Cortical thickness captures dendritic arborization, myelination, and pruning82,83 which increases during the first two years of life84, followed by a decrease37, similar to

trajectories observed for volume. Prior work in schizophrenia and bipolar disorder has shown cortical thinning across many regions of the brain, particularly in schizophrenia, including in

the insula17. Our vertex-based approach revealed a cluster in the agranular insula with cortical thinning in psychosis, but this was not observed in PS youth. Discrepancies between the

widespread volume deficits and relatively small changes in cortical thickness could be explained (at least in part) by differences in surface area. Both thickness and surface area contribute

to estimates of volume but are themselves independent85. Surface area is highly correlated with our gyrification index63, and surface area of the insula has previously been shown to be

reduced in schizophrenia86. Insula sulcal depth, which reflects cortical folding related to both early radial (vertical) and later tangential (horizontal) brain growth87, did not demonstrate

group differences in psychosis. Sulcal depth is calculated based on the Euclidian distance between the top and bottom of a sulcus, indicating that psychosis is not associated with shallower

insula sulci. This is in contrast with lower gyrification, which reflects the degree of folding, but not the depth of the folds. Our data suggest that folding was reduced in psychotic

disorders but the depth of the sulci were similar between groups. Previous literature on sulcal depth in psychosis is limited and mixed43,77. In fact, the current data may represent the most

comprehensive analysis of insula sulcal depth in psychotic disorders to-date and suggests it is intact in schizophrenia and may be slightly increased in bipolar disorder. Further studies in

independent cohorts are needed to replicate these findings. Finally, sub-regional structure is associated with clinical phenotypes across the psychosis spectrum. In both cohorts,

dysgranular and agranular insula volume were significantly associated with overall cognition, suggesting that smaller volume of these sub-regions contributes to greater cognitive impairment

across the psychosis-spectrum – a core feature of psychosis-risk88. Furthermore, worse negative symptoms were associated with lower insula volume in all three sub-regions, while positive

symptoms were associated with lower thickness. Insula sub-regions have distinct whole-brain connectivity patterns, reflecting their involvement in different functional systems9,47,48,89.

Despite this, we saw limited evidence of specific brain-behavior relationships depending on insula sub-region, instead findings that volume and thickness were predictive of many of the

phenotypes, particularly in the psychotic disorders cohort. This suggests that abnormal insula structure (particularly a smaller and thinner insula) contributes to overall clinical symptom

severity in psychotic disorders. We do note, however, that one intriguing area of specificity was dysgranular gyrification, which was only related to cognitive ability. While requiring

replication, this finding suggests that altered cortical folding driven by local connectivity contributes to the cognitive deficits observed in schizophrenia-spectrum disorders. While these

effect sizes are small, in the context of large cohorts and correlational psychological research, they are still meaningful90 and should motivate continued research into the clinical

implications of altered insula structure in psychosis. Strengths of the current study include large sample sizes, comparison across multiple cohorts at varying stages along the

psychosis-spectrum, analysis of insula sub-regions based on segmentation of granular layers, and analysis of structural integrity using both volume and surface-based measures. Limitations

include different measures of cognition and symptoms collected across the cohorts, limiting direct comparison of brain-behavior relationships. Furthermore, this is a cross-sectional sample

and we therefore can only infer progressive changes from psychosis-spectrum youth to psychotic disorders. In sum, insula volume is reduced in psychotic disorders and psychosis-spectrum youth

and correlates with cognitive ability in both cohorts, suggesting insula volume deficits confer broad risk for psychosis. Reduced gyrification of the insula is also observed in psychotic

disorders, but is most pronounced in schizophrenia and contributes to cognitive impairment. Negative and positive symptoms are also related to insula sub-regional structure, suggesting that

abnormal pathophysiology of the insula contributes to the clinical phenotypes. Future directions include longitudinal analysis of the PNC cohort to determine whether those who convert to

schizophrenia demonstrate more pronounced gyrification abnormalities than the PS youth as a whole. Examination of gyrification and volume abnormalities in relationship with polygenic risk

scores for schizophrenia may also reveal their contribution to cognitive and developmental subtyping of psychosis91. Finally, molecular analysis of granular layers in psychotic disorders can

be used to assess how alterations picked up by MRI map onto the cytoarchitecture, informing biological models of psychosis. REFERENCES * Streeter, G. L. The development of the nervous

system. In: _Manual of human embryology_, _vol II_. Lippincott, Philadelphia. * Wylie, K. P. & Tregellas, J. R. The role of the insula in schizophrenia. _Schizophr. Res._ 123, 93–104

(2010). Article  PubMed  PubMed Central  Google Scholar  * Goodkind, M. et al. Identification of a common neurobiological substrate for mental illness. _JAMA Psychiatry_ 72, 305–315 (2015).

Article  PubMed  PubMed Central  Google Scholar  * Sheffield, J. M., Rogers, B. P., Blackford, J. U., Heckers, S. & Woodward, N. D. Insula functional connectivity in schizophrenia.

_Schizophr. Res._ 220, 69–77 (2020). Article  PubMed  PubMed Central  Google Scholar  * Tian, Y., Zalesky, A., Bousman, C., Everall, I. & Pantelis, C. Insula functional connectivity in

schizophrenia: subregions, gradients, and symptoms. _Biol. Psychiatry Cogn. Neurosci. Neuroimaging_ 4, 399–408 (2019). PubMed  Google Scholar  * Moran, L. V. et al. Disruption of anterior

insula modulation of large-scale brain networks in schizophrenia. _Biol. Psychiatry_ 74, 467–474 (2013). Article  PubMed  PubMed Central  Google Scholar  * Palaniyappan, L., Simmonite, M.,

White, T. P., Liddle, E. B. & Liddle, P. F. Neural primacy of the salience processing system in schizophrenia. _Neuron_ 79, 814–828 (2013). Article  CAS  PubMed  PubMed Central  Google

Scholar  * Nieuwenhuys, R. The insular cortex. A review. In: _Progress in Brain Research_. Elsevier B.V., 2012, pp 123–163. * Mesulam, M. & Mufson, E. J. Insula of the Old World Monkey.

I: Architectonics in the Insulo-orbito-temporal Component of the Paralimbic Brain. * McDonald, C. et al. Regional volume deviations of brain structure in schizophrenia and psychotic bipolar

disorder: computational morphometry study. _Br. J. Psychiatry_ 186, 369–377 (2005). Article  PubMed  Google Scholar  * Crespo-Facorro, B. et al. Insular cortex abnormalities in

schizophrenia: a structural magnetic resonance imaging study of first-episode patients. _Schizophr. Res_ 46, 35–43 (2000). Article  CAS  PubMed  Google Scholar  * Takahashi, T. et al.

Diagnostic specificity of the insular cortex abnormalities in first-episode psychotic disorders. _Prog. Neuro-Psychopharmacol. Biol. Psychiatry_ 33, 651–657 (2009). Article  Google Scholar 

* Makris, N. et al. Decreased volume of left and total anterior insular lobule in schizophrenia. _Schizophr. Res._ 83, 155–171 (2006). Article  PubMed  Google Scholar  * Shepherd, A. M.,

Matheson, S. L., Laurens, K. R., Carr, V. J. & Green, M. J. Systematic meta-analysis of insula volume in Schizophrenia. _Biol. Psychiatry_ 72, 775–784 (2012). Article  PubMed  Google

Scholar  * Uddin, L. Q., Kinnison, J., Pessoa, L. & Anderson, M. L. Beyond the tripartite cognition–emotion–interoception model of the human insular cortex. 2013.

https://doi.org/10.1162/JOCN_A_00462. * Ellison-Wright, I. & Bullmore, E. Anatomy of bipolar disorder and schizophrenia: a meta-analysis. _Schizophr. Res._ 117, 1–12 (2010). Article 

PubMed  Google Scholar  * Rimol, L. M. et al. Cortical volume, surface area, and thickness in schizophrenia and bipolar disorder. _Biol. Psychiatry_ 71, 552–560 (2012). Article  PubMed 

Google Scholar  * Lee, S.-H. et al. Initial and progressive gray matter abnormalities in insular gyrus and temporal pole in first-episode schizophrenia contrasted with first-episode

affective psychosis. _Schizophr. Bull._ 42, 790–801 (2016). Article  PubMed  Google Scholar  * Bechdolf, A. et al. Amygdala and insula volumes prior to illness onset in bipolar disorder: a

magnetic resonance imaging study. _Psychiatry Res. Neuroimaging_ 201, 34–39 (2012). Article  Google Scholar  * Takahashi, T. et al. Insular cortex gray matter changes in individuals at

ultra-high-risk of developing psychosis. _Schizophr. Res._ 111, 94–102 (2009). Article  PubMed  Google Scholar  * Smieskova, R. et al. Insular volume abnormalities associated with different

transition probabilities to psychosis. _Psychol. Med._ 42, 1613–1625 (2012). Article  CAS  PubMed  Google Scholar  * Borgwardt, S. J. et al. Regional gray matter volume abnormalities in the

at risk mental state. _Biol. Psychiatry_ 61, 1148–1156 (2007). Article  PubMed  Google Scholar  * Ellison-Wright, I., Glahn, D. C., Laird, A. R., Thelen, S. M. & Bullmore, E. The anatomy

of first-episode and chronic schizophrenia: an anatomical likelihood estimation meta-analysis. _Am. J. Psychiatry_ 165, 1015–1023 (2008). Article  PubMed  PubMed Central  Google Scholar  *

Chan, R. C. K., Di, X., McAlonan, G. M., & Gong, Q. -Y. Brain anatomical abnormalities in high-risk individuals, first-episode, and chronic schizophrenia: an activation likelihood

estimation meta-analysis of illness progression. _Schizophr. Bull._ 37, 177–188 (2011). Article  PubMed  Google Scholar  * Mills, K. L. et al. Structural brain development between childhood

and adulthood: convergence across four longitudinal samples. _Neuroimage_ 141, 273–281 (2016). Article  PubMed  Google Scholar  * Giedd, J. N. & Rapoport, J. L. Structural MRI of

pediatric brain development: what have we learned and where are we going? _Neuron_ 67, 728–734 (2010). Article  CAS  PubMed  PubMed Central  Google Scholar  * Shaw, P. et al.

Neurodevelopmental Trajectories of the Human Cerebral Cortex. 2008. https://doi.org/10.1523/JNEUROSCI.5309-07.2008. * Shaw, P. et al. Intellectual ability and cortical development in

children and adolescents. _Nature_ 440, 676–679 (2006). Article  CAS  PubMed  Google Scholar  * Giedd, J. N. et al. Anatomical brain magnetic resonance imaging of typically developing

children and adolescents. _J. Am. Acad. Child Adolesc. Psychiatry_ 48, 465–470 (2009). Article  PubMed  PubMed Central  Google Scholar  * Rakic, P. Specification of cerebral cortical areas.

_Science_ 241, 170–176 (1988). Article  CAS  PubMed  Google Scholar  * Norbom, L. B. et al. New insights into the dynamic development of the cerebral cortex in childhood and adolescence:

integrating macro- and microstructural MRI findings. https://doi.org/10.31234/OSF.IO/F384C. * Hofer, E. et al. Genetic correlations and genome-wide associations of cortical structure in

general population samples of 22,824 adults. _Nat. Commun._ 11, 1–16 (2020). Article  Google Scholar  * Armstrong, E., Schleicher, A., Omran, H., Curtis, M. & Zilles, K. The ontogeny of

human gyrification. _Cereb. Cortex_ 5, 56–63 (1995). Article  CAS  PubMed  Google Scholar  * Zilles, K., Palomero-Gallagher, N. & Amunts, K. Development of cortical folding during

evolution and ontogeny. _Trends Neurosci._ 36, 275–284 (2013). Article  CAS  PubMed  Google Scholar  * Meng, Y., Li, G., Lin, W., Gilmore, J. H. & Shen, D. Spatial distribution and

longitudinal development of deep cortical sulcal landmarks in infants. _Neuroimage_ 100, 206–218 (2014). Article  PubMed  Google Scholar  * Im, K. & Grant, P. E. Sulcal pits and patterns

in developing human brains. _Neuroimage_ 185, 881–890 (2019). Article  PubMed  Google Scholar  * Ducharme, S. et al. Trajectories of cortical thickness maturation in normal brain

development - The importance of quality control procedures. _Neuroimage_ 125, 267–279 (2016). Article  PubMed  Google Scholar  * Remer, J. et al. Quantifying cortical development in

typically developing toddlers and young children, 1–6 years of age. _Neuroimage_ 153, 246–261 (2017). Article  PubMed  Google Scholar  * Roiz-Santiáñez, R. et al. Insular cortex thinning in

first episode schizophrenia patients. _Psychiatry Res. Neuroimaging_ 182, 216–222 (2010). Article  Google Scholar  * Spalthoff, R., Gaser, C. & Nenadić, I. Altered gyrification in

schizophrenia and its relation to other morphometric markers. _Schizophr. Res._ 202, 195–202 (2018). Article  PubMed  Google Scholar  * Nesvåg, R. et al. Regional thinning of the cerebral

cortex in schizophrenia: effects of diagnosis, age and antipsychotic medication. _Schizophr. Res._ 98, 16–28 (2008). Article  PubMed  Google Scholar  * Palaniyappan, L. & Liddle, P. F.

Aberrant cortical gyrification in schizophrenia: a surface-based morphometry study. _J. Psychiatry Neurosci._ 37, 399–406 (2012). Article  PubMed  PubMed Central  Google Scholar  * Yan, J.

et al. Cortical thinning and flattening in schizophrenia and their unaffected parents. _Neuropsychiatr. Dis. Treat._ 15, 935–946 (2019). Article  PubMed  PubMed Central  Google Scholar  *

Palaniyappan, L., Mallikarjun, P., Joseph, V., White, T. P. & Liddle, P. F. Reality distortion is related to the structure of the salience network in schizophrenia. _Psychol. Med._ 41,

1701–1708 (2011). Article  CAS  PubMed  Google Scholar  * Cauda, F. et al. Meta-analytic clustering of the insular cortex: characterizing the meta-analytic connectivity of the insula when

involved in active tasks. _Neuroimage_ 62, 343–355 (2012). Article  PubMed  Google Scholar  * Howes, O. D. & Murray, R. M. Schizophrenia: an integrated sociodevelopmental-cognitive

model. _Lancet_ 383, 1677–1687 (2014). Article  PubMed  Google Scholar  * Evrard, H. C. The organization of the primate insular cortex. _Front. Neuroanat._ 13, 43 (2019). Article  CAS 

PubMed  PubMed Central  Google Scholar  * Deen, B., Pitskel, N. B. & Pelphrey, K. A. Three systems of insular functional connectivity identified with cluster analysis. _Cereb. Cortex_

21, 1498–1506 (2011). Article  PubMed  Google Scholar  * Chang, L. J., Yarkoni, T., Khaw, M. W. & Sanfey, A. G. Decoding the role of the insula in human cognition: functional

parcellation and large-scale reverse inference. _Cereb. Cortex_ 23, 739–749 (2013). Article  PubMed  Google Scholar  * Kelly, C. et al. A convergent functional architecture of the insula

emerges across imaging modalities. _Neuroimage_ 61, 1129–1142 (2012). Article  PubMed  Google Scholar  * Uddin, L. Q., Nomi, J. S., Hébert-Seropian, B., Ghaziri, J. & Boucher, O.

Structure and function of the human insula. _J. Clin. Neurophysiol._ 34, 300–306 (2017). Article  PubMed  PubMed Central  Google Scholar  * Calkins, M. E. et al. The Philadelphia

Neurodevelopmental Cohort: constructing a deep phenotyping collaborative. _J. Child Psychol. Psychiatry_ 56, 1356–1369 (2015). Article  PubMed  PubMed Central  Google Scholar  * Kay, S. R.,

Opler, L. A. & Lindenmayer, J. P. The Positive and Negative Syndrome Scale (PANSS): rationale and standardisation. _Br. J. Psychiatry_ 155, 59–67 (1989). * Purdon, S. E. _The Screen for

Cognitive Impairment in Psychiatry (SCIP): Administration Manual and Normative Data_. PNL Inc: Edmonton, Alberta, 2005. * Gomez-Benito, J. et al. The screen for cognitive impairment in

psychiatry: diagnostic-specific standardization in psychiatric ill patients. _BMC Psychiatry_ 13, 127 (2013). Article  PubMed  PubMed Central  Google Scholar  * Kay, S. R., Fiszbein, A.

& Opler, L. A. The positive and negative syndrome scale (PANSS) for schizophrenia. _Schizophr. Bull._ 13, 261–276 (1987). Article  CAS  PubMed  Google Scholar  * Satterthwaite, T. D. et

al. Connectome-wide network analysis of youth with Psychosis-Spectrum symptoms. _Mol. Psychiatry_ 20, 1508–1515 (2015). Article  CAS  PubMed  PubMed Central  Google Scholar  * Wolf, D. H. et

al. Functional neuroimaging abnormalities in youth with psychosis spectrum symptoms. _JAMA Psychiatry_ 72, 456–465 (2015). Article  PubMed  PubMed Central  Google Scholar  * Gur, R. C. et

al. Age group and sex differences in performance on a computerized neurocognitive battery in children age 8–21. _Neuropsychology_ 26, 251–265 (2012). Article  PubMed  PubMed Central  Google

Scholar  * Friston, K. J. et al. Statistical parametric maps in functional imaging: a general linear approach. _Hum Brain Mapp_ 1994. https://doi.org/10.1002/hbm.460020402. * Dahnke, R.,

Ziegler, G., & Gaser C. Local adaptive segmentation. Beijing, HBM. 2012. * Dahnke, R., Yotter, R. A. & Gaser, C. Cortical thickness and central surface estimation. _Neuroimage_ 65,

336–348 (2013). Article  PubMed  Google Scholar  * Luders, E. et al. A curvature-based approach to estimate local gyrification on the cortical surface. _Neuroimage_ 29, 1224–1230 (2006).

Article  CAS  PubMed  Google Scholar  * Farb, N. A. S., Segal, Z. V. & Anderson, A. K. Attentional modulation of primary interoceptive and exteroceptive cortices. _Cereb. Cortex_ 23,

114–126 (2013). Article  PubMed  Google Scholar  * Brett M., Anton J.-L., Valabregue, R. & Poline, J.-B. Region of interest analysis using an SPM toolbox. Sendai, Japan, (2002)

http://www.mrc-cbu.cam.ac.uk/Imaging/marsbar.html (Accessed 15 May 2020). * Glasser, M. F. et al. A multi-modal parcellation of human cerebral cortex. _Nature_ 536, 171–178 (2016). Article 

CAS  PubMed  PubMed Central  Google Scholar  * Desikan, R. S. et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest.

_Neuroimage_ 31, 968–980 (2006). Article  PubMed  Google Scholar  * Augustine, J. R. Circuitry and functional aspects of the insular lobe in primates including humans. _Brain Res. Rev_.

https://doi.org/10.1016/S0165-0173(96)00011-2 (1996). * Craig, A. D. How do you feel—now? The anterior insula and human awareness. Nat. Rev. Neurosci. 10, 59–70 (2009). * Uddin, L. Q.

Salience processing and insular cortical function and dysfunction. _Nat. Rev. Neurosci._ 16, 55–61 (2015). Article  CAS  PubMed  Google Scholar  * Wang, X. et al. Brain grey-matter volume

alteration in adult patients with bipolar disorder under different conditions: a voxel-based meta-analysis. _J. Psychiatry Neurosci._ 44, 89–101 (2019). Article  PubMed  Google Scholar  *

Fusar-Poli, P. et al. Predicting psychosis: meta-analysis of transition outcomes in individuals at high clinical risk. _Arch. Gen. Psychiatry_ 69, 220–229 (2012). Article  PubMed  Google

Scholar  * Palaniyappan, L. & Liddle, P. F. Diagnostic discontinuity in psychosis: a combined study of cortical gyrification and functional connectivity. _Schizophr. Bull._ 40, 675–684

(2014). Article  PubMed  Google Scholar  * Nesvåg, R. et al. Reduced brain cortical folding in schizophrenia revealed in two independent samples. _Schizophr. Res._ 152, 333–338 (2014).

Article  PubMed  Google Scholar  * Afif, A., Bouvier, R., Buenerd, A., Trouillas, J. & Mertens, P. Development of the human fetal insular cortex: study of the gyration from 13 to 28

gestational weeks. _Brain Struct. Funct._ 212, 335–346 (2007). Article  CAS  PubMed  Google Scholar  * White, T., Su, S., Schmidt, M., Kao, C. Y. & Sapiro, G. The development of

gyrification in childhood and adolescence. _Brain Cogn._ 72, 36–45 (2010). Article  PubMed  Google Scholar  * Damme, K. S. F. et al. Cortical morphometry in the psychosis risk period: a

comprehensive perspective of surface features. _Biol. Psychiatry Cogn. Neurosci. Neuroimaging_ 4, 434–443 (2019). PubMed  Google Scholar  * Van Essen, D. C. A tension-based theory of

morphogenesis and compact wiring in the central nervous system. _Nature_ 385, 313–318 (1997). Article  PubMed  Google Scholar  * White, T. & Hilgetag, C. C. Gyrification and neural

connectivity in schizophrenia https://doi.org/10.1017/S0954579410000842. (2006). * Khadka, S. et al. Is aberrant functional connectivity a psychosis endophenotype? A resting state functional

magnetic resonance imaging study. _Biol. Psychiatry_ 74, 458–466 (2013). Article  PubMed  PubMed Central  Google Scholar  * Sheffield, J. M., Karcher, N. R. & Barch, D. M. Cognitive

deficits in psychotic disorders: a lifespan perspective. _Neuropsychol. Rev._ 28 https://doi.org/10.1007/s11065-018-9388-2 (2018). * Huttenlocher, P. R. Morphometric study of human cerebral

cortex development. _Neuropsychologia_ 28, 517–527 (1990). Article  CAS  PubMed  Google Scholar  * Sowell, E. R. et al. Longitudinal mapping of cortical thickness and brain growth in normal

children. _J. Neurosci._ 24, 8223–8231 (2004). Article  CAS  PubMed  PubMed Central  Google Scholar  * Li, G., Lin, W., Gilmore, J. H. & Shen, D. Spatial patterns, longitudinal

development, and hemispheric asymmetries of cortical thickness in infants from birth to 2 years of age. _J. Neurosci._ 35, 9150–9162 (2015). Article  CAS  PubMed  PubMed Central  Google

Scholar  * Winkler, A. M. et al. Joint analysis of cortical area and thickness as a replacement for the analysis of the volume of the cerebral cortex. _Cereb. Cortex_ 28, 738–749 (2018).

Article  PubMed  Google Scholar  * Palaniyappan, L. & Liddle, P. F. Differential effects of surface area, gyrification and cortical thickness on voxel based morphometric deficits in

schizophrenia. _Neuroimage_ 60, 693–699 (2012). Article  PubMed  Google Scholar  * Im, K. et al. Reliable identification of deep sulcal pits: the effects of scan session, scanner, and

surface extraction tool. _PLoS ONE_ 8 https://doi.org/10.1371/journal.pone.0053678 (2013). * Reichenberg, A. Cognitive impairment as a risk factor for psychosis. _Dialogues Clin. Neurosci._

7, 31–38 (2005). Article  PubMed  PubMed Central  Google Scholar  * Namkung, H., Kim, S.-H. & Sawa, A. The insula: an underestimated brain area in clinical neuroscience, psychiatry, and

neurology. _Trends Neurosci._ 40, 200–207 (2017). Article  CAS  PubMed  PubMed Central  Google Scholar  * Funder, D. C. & Ozer, D. J. Evaluating effect size in psychological research:

sense and nonsense. _Adv. Methods Pr. Psychol. Sci._ 2, 156–168 (2019). Google Scholar  * Dickinson, D. et al. Distinct polygenic score profiles in schizophrenia subgroups with different

trajectories of cognitive development. _Am. J. Psychiatry_ 177, 298–307 (2020). Article  PubMed  Google Scholar  Download references FUNDING Dr. Sheffield, Dr. Rogers, Dr. Huang, Dr.

Blackford, Dr. Heckers, and Dr. Woodward all report that they have no financial interests or potential conflicts of interest. This work was supported by NIMH grants R01 MH102266 and R01

MH115000 (awarded to N.D.W.), R01 MH070560 (awarded to S.H.), the Charlotte and Donald Test Fund, the Jack Martin, MD, Research Professorship in Psychopharmacology (awarded to J.U.B.), and

the Vanderbilt Institute for Clinical and Translational Research (through grant 1-UL-1-TR000445 from the National Center for Research Resources/NIH). This work was conducted in part using

the resources of the Center for Computational Imaging at the Vanderbilt University Institute of Imaging Sciences and the Advanced Computing Center for Research and Education (ACCRE) at

Vanderbilt University, Nashville, TN. AUTHOR INFORMATION AUTHORS AND AFFILIATIONS * Department of Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, TN, USA

Julia M. Sheffield, Anna S. Huang, Stephan Heckers & Neil D. Woodward * Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA

Baxter P. Rogers * Institute of Imaging Science, Vanderbilt University, Nashville, TN, USA Baxter P. Rogers * Munroe-Meyer Institute, University of Nebraska Medical Center, Omaha, NE, USA

Jennifer Urbano Blackford Authors * Julia M. Sheffield View author publications You can also search for this author inPubMed Google Scholar * Anna S. Huang View author publications You can

also search for this author inPubMed Google Scholar * Baxter P. Rogers View author publications You can also search for this author inPubMed Google Scholar * Jennifer Urbano Blackford View

author publications You can also search for this author inPubMed Google Scholar * Stephan Heckers View author publications You can also search for this author inPubMed Google Scholar * Neil

D. Woodward View author publications You can also search for this author inPubMed Google Scholar CONTRIBUTIONS Conception of the work and study design were developed by J.M.S. and N.D.W.

J.M.S., A.S.H., B.R., and N.D.W. contributed to data analysis. S.H., J.U.B., N.D.W., and J.M.S. contributed to data interpretation. J.M.S. and N.D.W. drafted the manuscript and all authors

provided important revisions for intellectual content. CORRESPONDING AUTHOR Correspondence to Julia M. Sheffield. ETHICS DECLARATIONS CONFLICT OF INTEREST The authors declare no competing

interests. ADDITIONAL INFORMATION PUBLISHER’S NOTE Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. SUPPLEMENTARY

INFORMATION SUPPLEMENTAL MATERIAL RIGHTS AND PERMISSIONS OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing,

adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons

license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a

credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted

use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions ABOUT

THIS ARTICLE CITE THIS ARTICLE Sheffield, J.M., Huang, A.S., Rogers, B.P. _et al._ Insula sub-regions across the psychosis spectrum: morphology and clinical correlates. _Transl Psychiatry_

11, 346 (2021). https://doi.org/10.1038/s41398-021-01461-0 Download citation * Received: 23 March 2021 * Revised: 04 May 2021 * Accepted: 14 May 2021 * Published: 04 June 2021 * DOI:

https://doi.org/10.1038/s41398-021-01461-0 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not

currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative