Abstract
Behavioural rigidity, the tendency to persist with inflexible patterns of thoughts or actions, is increasingly recognized as a transdiagnostic symptom across psychiatric, neurodevelopmental and neurodegenerative disorders. Empirical studies exploring the prevalence and underlying neural mechanisms of behavioural rigidity in dementia, however, are lacking.
This cross-sectional study sought to delineate the structural and functional neural correlates of behavioural rigidity using a transdiagnostic approach looking across the frontotemporal lobar degeneration (FTLD) spectrum and Alzheimer’s disease (AD). A total of 204 participants were recruited, including 110 frontotemporal dementia (FTD), 53 AD and 41 healthy older control participants. Within the FTD group, 66 cases were diagnosed with clinically probable behavioural variant FTD (bvFTD), 26 presented with semantic dementia (SD) and 18 cases had progressive non-fluent aphasia (PNFA). Behavioural rigidity was assessed using the Stereotypical and Motor Behaviour subscale of the Cambridge Behavioural Inventory-Revised. Voxel-based morphometry (VBM) was performed to identify grey matter regions associated with behavioural rigidity transdiagnostically, the results of which informed subsequent seed-based voxel-wise functional connectivity analyses. All imaging analyses were adjusted for relevant demographic and technical covariates. Statistical thresholds were set at voxel-level P < 0.001 (uncorrected) and cluster-level P < 0.05 (false discovery rate corrected).
Our main finding was that behavioural rigidity is pervasive across dementia subtypes, ranging from most pronounced in bvFTD to milder/relatively absent in PNFA, relative to controls. Whole-brain VBM across the entire patient sample revealed a significant negative association between behavioural rigidity and grey matter intensity exclusively in the bilateral nucleus accumbens. Using the bilateral nucleus accumbens as seeds, resting-state functional connectivity analysis showed that higher levels of behavioural rigidity were associated with stronger connectivity between the left nucleus accumbens and the left supplementary motor area, paracentral lobule and precuneus.
This is the first study, to our knowledge, to examine the neural substrates of behavioural rigidity across FTLD syndromes transdiagnostically using structural and functional neuroimaging approaches. Our findings reveal a gradation of rigid and repetitive behaviours, most apparent in bvFTD, which in turn reflects pathological disruption of the nucleus accumbens. Taken together, our findings highlight the need to consider repetitive and rigid behaviours as a transdiagnostic feature in neurodegenerative disorders, and one which indexes underlying nucleus accumbens pathology. More broadly, this study underscores the importance of screening for rigid and repetitive behaviours in the clinic and identifies the nucleus accumbens as a promising neural target to ameliorate these symptoms.
Keywords: frontotemporal dementia, reward processing, striatum, compulsivity, stereotypies, anhedonia
Chen et al. show that rigid and repetitive behaviours occur across dementia syndromes, most prominently in behavioural variant frontotemporal dementia. These symptoms are linked to structural and functional changes in the brain’s reward hub, the nucleus accumbens.
Introduction
We all exhibit a preference for routine from time to time, for example enjoying the same coffee from the same café every morning. Behavioural rigidity, however, represents an extreme form of entrenched behaviours, reflected as the tendency to persist with ongoing behaviour even when it is no longer appropriate for the given situation.1 Such repetitive behaviours are increasingly recognized as a feature of many neuropsychiatric [e.g. obsessive-compulsive disorder (OCD)], neurodevelopmental [autism spectrum disorder (ASD)] and other disorders of mental health (e.g. tobacco addiction),2-4 reflecting disruption of frontostriatal brain circuits that govern our capacity to respond to rewarding stimuli in an adaptive manner.5,6
Although behavioural rigidity has been documented in dementia syndromes, substantial variation in definitions, methods of assessment, and reporting practices has led to mixed and inconsistent reports. For example, a recent meta-analysis reported that 66.3% of individuals with Alzheimer’s disease (AD) exhibit repetitive or inflexible behaviours; however, many of these features appear to relate to cognitive difficulties (e.g. repetitive questioning, verbal repetition) and simple motor stereotypies rather than a pure behavioural rigidity complex.7 In contrast, rigid and repetitive behaviours are a core feature of frontotemporal dementia (FTD) syndromes, manifesting as stereotypies (e.g. excessive hand rubbing), compulsions (e.g. excessive hand washing), stereotyped language (e.g. catchphrases), impulsive behaviours (e.g. pathological gambling), hoarding, restricted interests, insistence on sameness and ritualistic actions.8-11 Early estimates suggest that 78% of patients with FTD display some form of repetitive or compulsive behaviour (e.g. stereotyped phrase repetition, compulsive bathing and dressing),12 which in turn takes a severe toll on carers and family members.13
Despite its high prevalence and implications for everyday functional outcomes, empirical evidence delineating the neurobiological mechanisms of behavioural rigidity in FTD is remarkably scarce. In a recent review, Restrepo-Martínez et al.3 proposed that transdiagnostic approaches developed for psychiatric disorders offer a valuable framework for elucidating the mechanisms and manifestations of repetitive and rigid behaviours in FTD. The term ‘repetitive behaviours’ was used in its widely accepted clinical sense; as an umbrella descriptor to denote behaviours that are inflexible, recurrent and difficult to modify, without implying that all such behaviours share identical underlying mechanisms.3 Notably, rigid behaviours across neuropsychiatric populations (e.g. OCD) have been conceptualized as emerging from large-scale brain network dysfunction converging on cortical-striatal-thalamic-cortical anatomical circuits.4,14 These circuits demonstrate remarkable overlap with frontostriatal regions that are vulnerable from early in the FTD disease course.3,15 Chief among such regions are the dorsal anterior cingulate cortex, anterior insula and the striatum.16
Mounting evidence suggests a link between alterations in frontostriatal regions specialized for reward processing and behavioural rigidity in FTD syndromes.17-19 Recent theoretical accounts have proposed that alterations in a striatally mediated reward processing system underlie inflexible behaviours, partly because diminished reward sensitivity increases the behavioural repetition required to exceed a heightened threshold for consummatory pleasure.5 This framework is supported by converging evidence linking behavioural rigidity in FTD to atrophy within reward-related striatal regions, including the nucleus accumbens (NAc), ventral putamen and pallidum.20-24 Indeed, a recent study demonstrated that behavioural rigidity severity scales as a function of atrophy to the right NAc in FTD, irrespective of clinical phenotype.21 Collectively, these studies point to a central role for the NAc in the genesis of rigid and repetitive behaviours in FTD syndromes.
One important consideration, however, is that FTD is a network disorder, characterized by the progressive deterioration of large-scale functional brain networks.25,26 Thus, while the NAc is implicated in behavioural rigidity in FTD, it likely forms part of a wider network of regions involved in higher-order cognitive processes and motor control.27 An early study comparing atrophy patterns of behavioural variant FTD (bvFTD) patients with and without stereotypies offers important clues in this regard. Josephs et al.28 demonstrated that stereotypies in bvFTD are associated with a greater proportion of striatal to cortical volume loss compared with those without. The authors concluded that striatal volume loss alone is unlikely to drive stereotypical behaviours; rather the imbalance between cortical and striatal volume loss in bvFTD is paramount.28 Evidence from neuropsychiatric disorders with rigid behaviours supports this proposal. Increased functional connectivity between the left NAc and right premotor cortex is associated with severity of rigid behaviours in ASD and OCD, suggesting that aberrant reward-related processing in the NAc drives excessive coupling with motor circuits responsible for implementing habitual action patterns.29 Moreover, regions supporting self-referential processing within the default mode network,30 such as the medial prefrontal cortex and medial temporal lobe (e.g. hippocampus), have been shown to interact with the NAc in the context of behavioural rigidity in OCD.31 Notably, the orbitofrontal cortex (OFC) plays a central role in the top-down regulation of the NAc, supporting adaptive behavioural selection and flexible adjustment to changing environmental demands.4 Prior work has shown that behavioural rigidity correlates with atrophy in the ventromedial/orbitofrontal prefrontal cortex as well as reward-related striatal regions, including the putamen and caudate,24 suggesting a frontostriatal mechanistic basis for inflexible behaviour. Taken together, these findings suggest that increased functional coupling between the NAc and cortical regions implicated in higher-order cognitive and motor control functions, represent a potential neurobiological substrate of behavioural rigidity.
To date, no empirical study has delineated the neural mechanisms of rigid and repetitive behaviours across FTD syndromes. The objectives of this study were, therefore, twofold. Building on a recent comprehensive review of rigid and repetitive behaviours in FTD,3 we first sought to establish the prevalence of rigid behaviours across a large, well-characterized cohort of patients clinically diagnosed with FTD in comparison with a non-FTD disease control group of patients diagnosed with AD. We used the Stereotypical and Motor Behaviour subscale of the Cambridge Behavioural Inventory-Revised (CBI-R) as a proxy for core features of rigid behaviour expressed by the patient in their daily lives.32 Based on previous studies, we hypothesized that rigid behaviours would be most pronounced in the bvFTD and semantic dementia (SD), and less prominent or absent entirely in progressive non-fluent aphasia and AD. Second, we sought to explicate the neural substrates of rigid behaviours using a transdiagnostic approach across the entire patient sample, combining structural and functional resting-state brain imaging. Here, we hypothesized that, irrespective of clinical diagnosis, behavioural rigidity would scale with the magnitude of atrophy to the NAc and its functional coupling with key nodes of sensorimotor and higher-order cognitive networks (e.g. default mode network).
Materials and methods
Participant recruitment
A total of 204 participants took part in this cross-sectional study, comprising 163 patients with dementia and 41 cognitively healthy older controls. Within the dementia sample, 66 individuals were diagnosed with clinically probable behavioural variant frontotemporal dementia (bvFTD), 53 with typical AD, 26 with SD and 18 with progressive non-fluent aphasia (PNFA). Participants were recruited between 2013 and 2023 through the FRONTIER research clinic at the Brain and Mind Centre, University of Sydney, Australia. All participants underwent comprehensive clinical investigation, neuropsychological assessment, and 3T structural and functional brain imaging. Dementia diagnoses were determined in accordance with internationally recognized criteria and conferred by a multidisciplinary team, including a neuropsychologist, senior neurologist and speech pathologist.
Briefly, clinically probable bvFTD was diagnosed based on the presence of prominent behavioural and personality changes including apathy, emotional blunting, impulsivity and eating changes, accompanied by significant medial and lateral prefrontal atrophy.33 Diagnosis of SD proceeded in line with current criteria for the semantic variant of primary progressive aphasia (svPPA)34 and was characterized by progressive deterioration in conceptual knowledge (e.g. impaired comprehension and naming) with relatively preserved phonology, fluency, visuospatial abilities and episodic memory. We constrained our focus to SD patients with predominantly left-lateralized anterior temporal lobe atrophy, given that the right temporal variant of FTD has only recently been formally characterized.35 PNFA was diagnosed in accordance with criteria for the non-fluent variant of PPA,34 based on the presence of expressive language deficits, including dysarthria, agrammatism, hesitant speech and phonemic errors, with relatively preserved non-linguistic cognitive functions. Finally, clinically probable typical AD was diagnosed according to the McKhann et al.36 criteria, characterized by episodic memory impairment in conjunction with visuospatial and executive dysfunction. Patients with atypical language (e.g. logopenic progressive aphasia) or visuospatial (e.g. posterior cortical atrophy) presentations of AD were excluded. All dementia patients were required to achieve a minimum score of 40/100 on the Addenbrooke’s Cognitive Examination III (ACE-III) to guard against severe cognitive impairment.
Forty-one cognitively healthy older controls were recruited from the FRONTIER volunteer database and local community groups. Control participants were required to score above 88/100 on the ACE-III and within normal limits on all cognitive tasks. Exclusion criteria for all participants included a history of psychiatric illness, significant head trauma resulting in loss of consciousness, alcohol or substance abuse, or limited English proficiency.
Ethics statement
The South Eastern Sydney Local Health District, University of New South Wales, and University of Sydney ethics committees approved the study (HREC 10/126 and 13/177). Informed consent, in accordance with the Declaration of Helsinki, was obtained from all participants or their person responsible.
Clinical and neuropsychological assessment
Global cognitive function was assessed using the Addenbrooke’s Cognitive Examination-Revised (ACE-R)37 and the ACE-III38; general cognitive screening tools which evaluate five cognitive domains: Attention and Orientation, Memory, Fluency, Language and Visuospatial abilities. ACE-R scores were converted to their ACE-III equivalent using established algorithms39 to ensure consistency across assessments. Behavioural symptoms were evaluated using the CBI-R,32 which assesses changes in domains such as motivation, eating, abnormal behaviours and inhibition. Disease duration was calculated as the number of years from the first reported symptom onset to the date of cognitive assessment. Level of functional impairment was quantified using the Frontotemporal Dementia Functional Rating Scale (FRS)40 with lower scores representing more severe levels of decline.
Assessment of behavioural rigidity
The Stereotypical and Motor Behaviour subscale of the Cambridge Behavioural Inventory-Revised (CBI-R)32 was used as a proxy for behavioural rigidity as observed by the carer in the daily life of the patient. This subscale comprises four items evaluating rigid and repetitive behaviours: (i) is rigid and fixed in their ideas and opinions; (ii) develops routines from which they cannot easily be discouraged (e.g. insisting on eating or going for walks at fixed times); (iii) clock-watches or appears preoccupied with time; and (iv) repeatedly uses the same expression or catchphrase. Carers rated the frequency and severity of these behaviours using a 4-point Likert scale, while controls provided self-ratings. For this study, we focused on the frequency score, which ranges from 0 (never) to 4 (constantly). As such, the total Stereotypical and Motor Behaviour subscale score ranges from 0 to 16, with higher scores reflecting greater behavioural rigidity. The subscale showed good internal consistency in the present patient sample (Cronbach’s α = 0.76; McDonald’s ω = 0.87), based on the available item responses. Inspection of CBI-R data revealed that the Stereotypical and Motor Behaviour subscale was not complete for 10 participants, while a further 30 participants were missing items across various other components of the CBI-R. A Wilcoxon rank-sum test revealed no significant difference in functional disease severity between participants with and without rigidity scores (FRS Rasch score: W = 881, P = 0.426), indicating no systematic group differences in disease severity. As such, missingness in rigidity scores is unlikely to be related to patients’ clinical capacity or cognitive performance. Missing data and how they were treated are described in full in the Supplementary material.
Statistical analyses
Demographic, cognitive and clinical data analyses were conducted using Jeffreys’s Amazing Statistics Program (JASP).41 To ensure a complete dataset, multiple imputation was applied to the clinical and cognitive data prior to analyses using the multiple imputation by chained equations (MICE) package42 (see the Supplementary material for full details). Group differences in continuous demographic, cognitive and clinical variables were assessed using one-way ANOVAs, with Holm–Bonferroni-corrected post hoc tests applied to examine significant main effects of Group (controls, bvFTD, SD, AD, PNFA). Partial eta-squared values (ηp2) accompany all ANOVAs as measures of corresponding effect sizes, while Cohen’s d is included for all post hoc between group tests. Categorical variables, such as sex, were analysed using chi-square tests.
Neuroimaging analyses
Participants assessed until December 2016 underwent MRI scanning on a 3 T Philips scanner (n = 103), whereas those assessed from January 2017 onwards were scanned using a 3 T GE Discovery MR750 scanner (n = 101). For dementia patients, MRI scans were acquired within 6 months of clinical assessment, while for healthy control subjects, scans were obtained within 12 months of assessment. All images were visually inspected by an experienced rater for image quality, head motion artefacts and the presence of white matter hyperintensities.
Structural imaging
Participants underwent whole-brain T1-weighted structural MRI on a 3 T scanner equipped with a standard 8-channel head coil. Imaging parameters were as follows, irrespective of scanner type: coronal orientation, 200 slices, 256 × 256 matrix, in-plane resolution of 1 × 1 mm, echo time/repetition time (TE/TR) of 2.6/5.8 ms, slice thickness of 1 mm and flip angle of 8°. A standard preprocessing pipeline and whole-brain voxel-based morphometry (VBM) protocol was employed. Full details of imaging preprocessing are provided in the Supplementary material.
Grey matter intensity-behaviour correlation analyses
Next, we explored whole-brain correlation analyses between grey matter intensity and the CBI-R Behavioural rigidity score, in the combined patient group (n = 163). A covariate-only statistical model was run with age, sex, education, scanning site and total intracranial volume included as nuisance variables. A negative t-contrast was used to explore associations between higher levels of behavioural rigidity and grey matter intensity decrease across the entire brain. Statistical maps were extracted voxel wise (P < 0.001, uncorrected) and thresholded at the cluster level using a false discovery rate of P < 0.05. Anatomical locations of statistical significance were overlaid on the Montreal Neurological Institute (MNI) standard brain with maximum coordinates provided in MNI stereotaxic space. Furthermore, to rule out potential confounding effects of diagnosis, we conducted additional hypothesis-driven region of interest (ROI) analyses in which diagnosis (AD, bvFTD, PNFA, SD) was explicitly included as a factor alongside age, sex, years of education, scanning site and total intracranial volume. To minimize potential confounding due to image quality, the weighted interquartile range (IQR) score was incorporated into the regression model in addition to age, sex, education, scanning site and total intracranial volume. All results were robust to this adjustment (Supplementary Table 2).
Resting-state functional MRI
Resting-state functionalMRI data were preprocessed using a standard pipeline. Full details of acquisition parameters and preprocessing steps are provided in the Supplementary material. Functional connectivity analysis was performed using Statistical Parametric Mapping (SPM12; http://www.fil.ion.ucl.ac.uk/spm) and the Data Processing & Analysis for Brain Imaging (DPABI 5.1; http://www.rfmri.org/dpabi) toolbox,43 implemented in MATLAB R2019b (The MathWorks, Inc., Natick, MA, USA). The significant clusters identified in the VBM analysis were extracted as binary masks and used as seed regions for the functional connectivity analysis. Functional connectivity maps were generated by computing Pearson correlation coefficients between the average time series of each seed region and whole-brain voxel-wise time series. Correlation coefficients were then transformed into normally distributed Z-scores using Fisher’s r-to-Z transformation.
Functional connectivity-behaviour correlation analyses
Adjusting for age, sex, years of education and framewise displacement, whole-brain correlation analyses were conducted to identify functional connectivity patterns significantly associated with behavioural rigidity across the entire dementia group (n = 163). Statistical significance was determined using a voxel-level threshold of P < 0.001 (uncorrected) and a cluster-level threshold of P < 0.05, corrected for multiple comparisons using the false discovery rate. To rule out potential confounding effects of diagnosis, we conducted additional hypothesis-driven ROI analyses in which diagnosis (AD, bvFTD, PNFA, SD) was explicitly included as a factor alongside framewise displacement, age, sex and years of education.
Validation analysis
To assess the robustness of the brain–behaviour correlation analyses, we conducted an additional validation analysis in which the same analytic procedures were applied after excluding the 10 participants who were missing rigidity scores on the CBI-R.
Results
Demographic, clinical information
Table 1 presents the demographic, clinical and cognitive data for study participants. Groups were comparable in terms of age [F(4,199) = 1.56, P = 0.187, ηp2 = 0.03], years of education [F(4,199) = 0.37, P = 0.833, ηp2 = 0.01] and sex distribution (χ2 = 7.41, P = 0.116). Disease duration (i.e. years elapsed since symptom onset) was largely comparable between the groups, except for a significantly longer disease duration in bvFTD relative to the AD group (P = 0.032). In terms of cognitive functioning, significant group differences were found on the ACE-III screening tool, with all patient groups showing significant impairments relative to controls (all P-values < 0.001; PNFA, bvFTD > AD, SD) and comparable levels of impairment in PNFA relative to bvFTD (P = 0.645) and in SD relative to AD (P = 0.125). Further group differences were evident on the FRS [F(3,199) = 27.80, P < 0.001, ηp2 = 0.34], whereby bvFTD showed the greatest functional impairment compared with all other patient groups (all P-values < 0.001), while PNFA and SD patients showed milder impairment relative to bvFTD. All dementia groups were classified as ‘moderate’ in terms of disease severity using the FRS, except PNFA, which was classified as ‘mild’. Total carer-rated behavioural changes on the CBI-R were significantly higher in all patient groups [F(4,199) = 51.32, P < 0.001, ηp2 = 0.51] relative to controls (all P-values < 0.001), and followed a graded profile: PNFA < AD < bvFTD and SD < bvFTD (all P-values < 0.001). Behavioural changes were comparable in SD and AD (P = 0.098) and between SD and PNFA (P = 0.072).
Table 1.
Demographic, cognitive and clinical characteristics of study participants
| bvFTD (n = 66) | SD (n = 26) | PNFA (n = 18) | AD (n = 53) | Control (n = 41) | Test statistic | Post hoc comparisons | |
|---|---|---|---|---|---|---|---|
| Age, years | 63.18 (8.49) | 64.93 (6.54) | 68.25 (10.08) | 64.91 (7.73) | 65.29 (7.27) | F = 1.56 | n/s |
| Education, years | 12.62 (2.91) | 12.33 (3.03) | 12.82 (2.82) | 12.72 (3.27) | 13.15 (2.33) | F = 0.37 | n/s |
| Sex, male:femalea | 45:21 | 14:12 | 8:10 | 25:28 | 20:21 | χ2 = 7.41 | n/s |
| Disease duration, years | 5.66 (3.81) | 4.56 (2.62) | 4.87 (2.32) | 4.10 (2.03) | – | F = 2.80 | AD < bvFTD |
| ACE-III [100] | 76.07 (13.51) | 60.58 (10.88) | 77.45 (10.32) | 65.61 (12.41) | 95.32 (3.09) | F = 54.04*** | AD, SD < PNFA, bvFTD < HC |
| FRS logit score [−6.66, 5.39]b | −0.69 (1.4) | 1.66 (1.70) | 2.42 (1.54) | 0.70 (1.68) | – | F = 27.80*** | bvFTD < AD < PNFA, SD |
| FRS stage | Moderate | Moderate | Mild | Moderate | – | – | – |
| CBI-R Total [180]c | 71.69 (27.74) | 41.56 (23.59) | 26.15 (15.68) | 51.04 (29.03) | 6.50 (4.83) | F = 51.32*** | HC < PNFA < AD < bvFTD HC < SD < bvFTD |
Means with standard deviations in parentheses. Maximum test score or score range of variables provided in square brackets, where appropriate. ACE-III = Addenbrooke’s Cognitive Examination-III; AD = Alzheimer’s disease; bvFTD = behavioural variant of frontotemporal dementia; CBI-R = Cambridge Behavioural Inventory-Revised; FRS = Frontotemporal Dementia Functional Rating Scale; HC = healthy controls; n/s = not significant; PNFA = progressive non-fluent aphasia; SD = semantic dementia.
bvFTD = behavioural variant of frontotemporal dementia; SD = semantic dementia; PNFA = progressive non-fluent aphasia; AD = Alzheimer’s disease; HC = Healthy Controls; FRS = Frontotemporal Dementia Functional Rating Scale; ACE-III = Addenbrooke’s Cognitive Examination-III. CBI-R = Cambridge Behavioural Inventory-Revised; n/s = not significant.
aChi-square test.
bLower logit scores, as derived using Rasch analyses, denote greater levels of functional impairment on the FRS.
cHigher scores on the CBI-R denote greater behavioural disturbances.
*** P < 0.001.
Profiles of behavioural rigidity in dementia
Figure 1 displays levels of behavioural rigidity on the CBI-R for patients (carer-rated) and controls (self-rated). A one-way ANOVA revealed a significant main effect of group for severity of behavioural rigidity on the CBI-R [F(4,199) = 30.97, P < 0.001, ηp2 = 0.38]. Post hoc tests with Holm–Bonferroni correction indicated that, relative to controls, all dementia subtypes exhibited higher levels of behavioural rigidity (all P-values < 0.001), with the exception of the PNFA group who were not found to differ from controls (P = 0.117, Cohen’s d = −0.59). Behavioural rigidity was most pronounced in bvFTD compared with all other patient groups (all P-values < 0.05). In contrast, the level of behavioural rigidity did not differ between AD, SD and PNFA groups (AD versus SD: P = 0.149, Cohen’s d = −0.43; AD versus PNFA: P = 0.229, Cohen’s d = 0.33; SD versus PNFA: P = 0.057, Cohen’s d = 0.76).
Figure 1.

Profiles of behavioural rigidity across dementia syndromes as indexed by the Stereotypical and Motor Behaviour subscale of the CBI-R. Carers provided ratings for dementia patients while controls provided self-ratings. Each dot represents an individual participant’s behavioural rigidity score. The box indicates the interquartile range (IQR), the horizontal line within the box marks the median and the whiskers extend to 1.5 × IQR. Higher scores represent greater severity of behavioural rigidity (maximum score of 16). AD = Alzheimer’s disease; bvFTD = behavioural variant of frontotemporal dementia; CBI-R = Cambridge Behavioural Inventory-Revised; PNFA = progressive non-fluent aphasia; SD = semantic dementia. **P < 0.01, ***P < 0.001.
Profiles of grey matter atrophy in dementia
As shown in Fig. 2 and Supplementary Table 1, each dementia group exhibited patterns of atrophy consistent with their established clinical phenotype, relative to controls. Briefly, bvFTD patients displayed widespread bilateral grey matter atrophy in the orbitofrontal and medial prefrontal cortices, extending to insular and temporopolar regions, as well as the cerebellum. Subcortical atrophy encompassed the hippocampus, amygdala and striatal nuclei, including the putamen, caudate and NAc, bilaterally. Patients with SD exhibited a characteristic pattern of asymmetric anterior temporal lobe atrophy, most pronounced on the left-hand side, including medial temporal structures (hippocampus and amygdala), and extending anteriorly to the bilateral insular and orbitofrontal cortices. Striatal involvement, particularly of the putamen, was also evident in the SD group. For PNFA, atrophy was primarily left-lateralized, involving the inferior frontal gyrus (including Broca’s area), supplementary motor area and insula. Finally, the AD group displayed grey matter volume loss primarily in the bilateral medial temporal lobes, including the hippocampus and amygdala, spreading to lateral temporal and prefrontal cortices. Parietal atrophy was also present, predominantly in the left hemisphere, involving the angular gyrus, supramarginal gyrus and inferior parietal lobule.
Figure 2.

Regions showing significant reductions in grey matter intensity in patient groups relative to healthy controls (HC): (A) behavioural-variant frontotemporal dementia (bvFTD), (B) semantic dementia (SD), (C) progressive non-fluent aphasia (PNFA) and (D) Alzheimer’s disease (AD). Age, sex, years of education, scanning site and total intracranial volume were included as covariates. Statistical significance was defined using a voxel-wise threshold of P < 0.001 (uncorrected) and a cluster-level threshold of P < 0.05, corrected for multiple comparisons using the false discovery rate. L = left; R = right.
Grey matter correlates of behavioural rigidity in dementia
Figure 3 displays the results from the VBM analyses exploring associations between carer ratings of behavioural rigidity on the Stereotypical and Motor Behaviour subscale of the CBI-R across the entire patient cohort (n = 163), where higher scores on the CBI-R indicate greater behavioural rigidity. The only region to emerge as significant in the analysis was the bilateral NAc, where higher levels of behavioural rigidity were associated with decreased grey matter intensity in the NAc across the dementia sample.
Figure 3.

Negative correlation between grey matter intensity of the nucleus accumbens and behavioural rigidity in the combined dementia sample (n = 163). Behavioural rigidity measured using the Stereotypical and Motor subscale of the CBI-R. Age, sex, years of education, scanning site and total intracranial volume were included as covariates. Statistical thresholds were set at P < 0.001 at the voxel level and P < 0.05 at the cluster level, corrected for multiple comparisons using the false discovery rate. Significant clusters were located in the left nucleus accumbens (cluster size = 161, peak MNI coordinates: x = −9, y = 15, z = −9, t = 3.89) and the right nucleus accumbens (cluster size = 195, peak MNI coordinates: x = 9, y = 7.5, z = −7.5, t = 4.90). CBI-R = Cambridge Behavioural Inventory-Revised; L = left; R = right.
We further extracted the mean grey matter intensity values from the left and right NAc and ran two separate multiple regression models predicting behavioural rigidity across the dementia cohort. After adjusting for diagnosis, site, total intracranial volume, age, sex and education, left-sided NAc intensity remained a significant negative predictor of behavioural rigidity (standardized β = −0.26, 95% confidence interval (CI) = −0.43 to −0.09, P = 0.003). A similar significant association was observed for right-sided NAc intensity (standardized β = −0.27, 95% CI = −0.44 to −0.10, P = 0.002). These results demonstrate that the observed VBM findings are robust to diagnostic adjustment and are not driven by group differences. As shown in Supplementary Fig. 7, the negative association was directionally consistent across all dementia subtypes, with the clearest pattern observed for the left NAc.
Resting-state functional connectivity analyses
We next sought to explore how NAc dysfunction relates to wider brain-network level changes as indexed by resting-state functional connectivity, and its association with behavioural rigidity in dementia. Using the left and right NAc as seed regions, we explored associations between whole-brain resting-state functional connectivity and behavioural rigidity across the entire dementia cohort. Figure 4 displays the results of the resting-state functional connectivity analyses. Overall, we found a significant positive correlation between behavioural rigidity and functional connectivity strength between the left NAc, supplementary motor area and paracentral lobule/precuneus. No significant associations were found when seeding from the right NAc.
Figure 4.

Left nucleus accumbens-based resting-state functional connectivity strength positively correlates with behavioural rigidity across dementia (n = 163). Behavioural rigidity measured using the Stereotypical and Motor subscale of the CBI-R. Framewise displacement, age, sex and years of education were included as covariates. Statistical thresholds were set at P < 0.001 at the voxel level and P < 0.05 at the cluster level, corrected for multiple comparisons using the false discovery rate. Significant clusters were identified in the left supplementary motor area (cluster size = 157, peak MNI coordinates: x = −6, y = 18, z = 54, t = 4.61) and the left paracentral lobule extending into the left precuneus (cluster size = 113, peak MNI coordinates: x = 0, y = −45, z = 63, t = 3.90). CBI-R = Cambridge Behavioural Inventory-Revised; L = left; R = right.
We further implemented hypothesis-driven ROI analyses to ensure that the functional connectivity findings were not driven by diagnostic differences. Specifically, connectivity values were extracted from the two peak clusters (left paracentral lobule/precuneus and left supplementary motor area), and two separate multiple regression models were performed with each ROI entered as the predictor of behavioural rigidity. After adjusting for diagnosis, framewise displacement, age, sex and education years, connectivity between the left NAc and the left paracentral lobule/precuneus remained a significant positive predictor of behavioural rigidity (standardized β = 0.31, 95% CI = 0.18 to 0.45, P < 0.001). A comparable effect was observed for left NAc–supplementary motor area connectivity (standardized β = 0.32, 95% CI = 0.18 to 0.46, P < 0.001). These results demonstrate that the functional connectivity associations are robust to the inclusion of diagnosis alongside other covariates. As shown in Supplementary Fig. 8, the positive association pattern was broadly consistent across individual dementia syndromes.
Validation analysis
In total, 10 participants did not have complete rigidity scores on the CBI-R, while 40 participants did not have a complete CBI-R total score due to missing data in one or more domains. The diagnostic distribution of the 10 participants with missing behavioural rigidity data is shown in Supplementary Fig. 2. The mean frequency of each behavioural rigidity item across dementia groups, calculated based on the available item responses, is presented in Supplementary Table 3.
Following exclusion of the 10 participants who were missing behavioural rigidity scores, the voxel-level grey matter–behaviour associations were fully replicated (Supplementary Fig. 3). ROI analyses further confirmed that grey matter intensity in both the left (standardized β = −0.26, 95% CI = −0.44 to −0.09, P = 0.004) and right (standardized β = −0.27, 95% CI = −0.44 to −0.09, P = 0.003) NAc significantly predicted behavioural rigidity, consistent with the VBM findings across dementia subtypes (Supplementary Fig. 4).
For resting-state analyses, excluding the same 10 participants also yielded fully replicated voxel-level connectivity results (Supplementary Fig. 5). Functional connectivity between the left NAc and the left paracentral lobule/precuneus (standardized β = 0.33, 95% CI = 0.20 to 0.47, P < 0.001) and left supplementary motor area (standardized β = 0.36, 95% CI = 0.22 to 0.50, P < 0.001) remained significant, with positive associations consistently observed across diagnostic groups (Supplementary Fig. 6).
Discussion
Classically considered the remit of neuropsychiatry, repetitive and rigid behaviours are gaining increasing attention in the field of neurodegenerative disorders. Using a symptom-centred approach, we explored the prevalence of rigid behaviours across the FTLD spectrum in comparison with AD and employed multimodal neuroimaging to delineate the neural mechanisms of these behaviours. Overall, we show that rigid behaviours are common in dementia yet vary systematically according to diagnostic subtype, ranging from most severe in bvFTD, to relatively absent in PNFA. Applying a transdiagnostic approach, we further demonstrate that the emergence of behavioural rigidity relates to dysfunction of a cortico-striatal network centred on the NAc and its respective functional connectivity with regions implicated in habitual motor actions.
Previous studies in neuropsychiatric and neurodevelopmental populations have emphasized the importance of frontostriatal structures in driving repetitive and rigid behaviours, the premise being that goal-directed behaviours are regulated by circuits involving the OFC and striatum, and that dysfunction within these pathways drives rigid behaviours.4,14 Using whole-brain VBM, we found an exclusive role for the NAc in the origin of rigid behaviours transdiagnostically in dementia, whereby decreased NAc grey matter intensity was associated with increased levels of behavioural rigidity. This finding is consistent with previous reports of associations between NAc volume loss and compulsive behaviours in FTD.20,21 Notable parallels are evident in the neuropsychiatric literature. For example, both ASD and OCD have been conceptualized as disorders of aberrant reward processing manifesting in ‘behavioural dependency’ on restricted and repetitive behaviours and interests.44-46 Strikingly, deep brain stimulation targeting the NAc has been shown to significantly diminish compulsive symptoms in OCD.31 Collectively, these lines of evidence suggest that an imbalance in brain circuits specialized for reward processing—particularly the NAc—may bias individuals towards rigid thoughts and actions over neutral or adaptive alternatives.
Moving beyond structural associations, we explored how disruption of cortico-striatal functional pathways underlies inflexible behaviours in dementia. Using the bilateral NAc as seeds, we found that resting-state functional connectivity between the left NAc and supplementary motor area, paracentral lobule and precuneus was positively correlated with severity of behavioural rigidity across the dementia sample. The supplementary motor area plays a central role in the planning and initiation of internally generated, automatic motor sequences47 while the paracentral lobule is posited to support motor and sensory functions of the lower limbs.48 Our findings converge remarkably well with a study,29 which showed that rigid behaviours in a mixed sample of ASD and OCD are associated with increased functional connectivity between the left NAc and premotor regions. Similarly, restricted and repetitive behaviours in ASD have been linked to both the temporal dynamics of functional connectivity and the structural integrity of motor regions, including the supplementary motor area.49,50 It may be that aberrant reward-related processing in the NAc drives excessive coupling with motor circuits responsible for implementing habitual action patterns, resulting in disruption of the balance between goal-directed and habitual behavioural control, as is typically observed in OCD.51
Moreover, functional connectivity between the NAc and the dorsal-anterior portion of the precuneus was positively associated with behavioural rigidity across dementia syndromes. As a hub of the default mode network, the precuneus is critically involved in self-referential processing, episodic memory, visuospatial imagery and internal mentation.52-54 Notably, the dorsal anterior portion of the precuneus is suggested to influence behaviour by integrating sensory, motor and cognitive information in preparation for executing goal-directed actions.55 As such, increased connectivity between the NAc and precuneus may reflect enhanced incentive salience towards internally generated thoughts or beliefs, which in turn influences motor control and repetitive approach behaviour.
Interestingly, we did not find significant associations between structural integrity of the OFC and rigid behaviours, nor did any prefrontal regions emerge as functionally correlated with the NAc in our resting-state analyses. This lack of association runs somewhat counter to existing theoretical positions which emphasize a prominent role for frontal regions such as the OFC in the top-down regulation of the NAc, facilitating adaptive and flexible behavioural patterns.4,56 Previous research has linked OFC atrophy to impaired value updating in reversal learning tasks in bvFTD, suggesting a mechanistic basis for inflexible behaviour.57 Similarly, theoretical frameworks focusing on the limbic cortical-striatal-thalamic-cortical circuit have implicated disruption of the limbic loop linking the OFC and NAc in compulsive behaviours across clinical populations such as OCD.3,4,56 Our failure to find significant associations between OFC degeneration and rigidity in this study may therefore reflect the inherent limitations of using the CBI-R Stereotypical and Motor behaviour subscale as a proxy measure for behavioural rigidity. Whereas psychiatric syndromes such as ASD and OCD typically show internalizing emotional and anxiety-related symptoms that drive compulsive behaviours, the dementia syndromes of interest here do not typically show such features. Future research adopting multidimensional assessment frameworks may help to clarify this issue enabling us to situate dementia populations within a truly transdiagnostic perspective that encompasses broader behavioural and affective domains. In this context, incorporating measures designed to capture internalizing-driven forms of rigidity, such as anxiety-related repetitive thoughts, compulsions and affective inflexibility, may be particularly informative. The feasibility and validity of such instruments in dementia populations remains unclear and thus represents an important area for future investigation.
Instead, our findings provide evidence of an alternative pathway whereby coupling between NAc and supplementary motor areas drive behavioural rigidity in dementia. We tentatively suggest that differential patterns of NAc connectivity result in distinct dissociations on the behavioural level. For example, while the NAc-motor pathway appears particularly important in driving rigid and repetitive behaviours, alterations in the classic frontostriatal pathway might give rise to a more compulsive profile of behaviours (e.g. hoarding). Importantly, recent perspectives emphasize the functional role of motor cortices in shaping the output of reward-related behaviour.58 In this light, it will be important to delineate how pathological alterations to key nodes of the brain’s reward system and regions in the motor control system impact the initiation and maintenance of goal-directed actions. While frontal cortical degeneration represents a plausible candidate mechanism driving rigid behaviours, we demonstrate here that behavioural manifestations of rigidity are also closely associated with functional dysregulation of cortico-striatal networks implicated in habitual motor control. It is noteworthy that the association between behavioural rigidity and NAc–motor coupling was confined to the left NAc. Although direct evidence for lateralization of the NAc remains limited, our finding raises the possibility that the left NAc plays a preferential role in regulating motor circuits underlying rigid behaviour. Supporting this view, one study in OCD and ASD reported that increased functional connectivity between the left NAc and premotor/motor regions correlates with the severity of rigid behaviours.29 Future studies are required to address this issue.
Behavioural rigidity is increasingly recognized as a transdiagnostic feature across a range of neuropsychiatric conditions, including OCD, ASD, anorexia nervosa, Parkinson’s disease and Huntington’s disease, and tobacco addiction.2-4,59 Application of transdiagnostic frameworks therefore provides a useful foundation to arrive at a more precise characterization of rigid behaviours in FTD. For example, it has been suggested that rigid behaviours in ASD may reflect compensatory shifts to modulate sensory input (e.g. mitigating overstimulation from aversive sounds), manage anxiety and foster a sense of predictability in uncertain environments.60 In this light, individuals with SD might seek to preserve a sense of control and coherence in an increasingly incomprehensible environment by adhering rigidly to fixed and familiar routines, for example, engaging in repetitive behaviours such as clock-watching and insistence on sameness and routines.11 In contrast, individuals with bvFTD may display a pattern of compulsive engagement in behaviours driven by immediate, tangible rewards (e.g. repetitive hoarding).61 Articulating the underlying cognitive mechanisms driving distinct manifestations of behavioural rigidity will be important to inform the development of targeted interventions that can be personalized to the symptom profile of the individual. These mechanisms could be explored using paradigms that dissociate externally cued set-shifting from internally generated semantic control, alongside tasks probing habit learning versus goal-directed behaviour. Applying such approaches across SD and bvFTD, together with task-based or resting-state functional MRI, would clarify whether distinct cognitive mechanisms map onto anterior temporal involvement in SD versus frontostriatal and salience-network dysfunction in bvFTD. Notably, deep brain stimulation targeting the NAc has shown efficacy in alleviating compulsive symptoms in OCD,31 and represents an attractive candidate given our findings of NAc involvement in behavioural rigidity in this study. Selective serotonin reuptake inhibitors (SSRIs), enhancing serotonergic activity in the central nervous system, are well-established as the first-line pharmacological treatment for OCD due to their efficacy, tolerability, safety and low potential for abuse.62,63 Given the association between serotonin system deficiencies and the emergence of behavioural rigidity in FTD,64-66 it would be valuable to investigate whether SSRIs effectively reduce behavioural rigidity symptoms in FTD. This is particularly relevant considering that some studies have successfully used SSRIs to target other behavioural features of FTD, including disinhibition.67
Rigid and routinized behaviours are commonly reported in SD, yet in this study, levels of behavioural rigidity were more severe in bvFTD relative to the SD group. This finding may reflect the functional status of our SD group, who showed relatively preserved daily functioning potentially indicating a milder stage of disease. It is possible that behavioural rigidity emerges more floridly with disease progression in SD; however, it may also be the case that patients with SD show relatively preserved functioning for a reduced set of repeated activities or routines, suggesting the need for more sensitive and targeted assessments of rigidity. Encroachment of atrophy into the right temporal lobe likely plays a crucial role in the emergence of rigid behaviours in SD.5 Support for this account comes from evidence that rigid behaviours are more prevalent in the right temporal variant of FTD,18 suggesting a regionally specific driver of rigidity. Longitudinal studies exploring the emergence of rigid behaviours in SD in relation to right temporal lobe disruption will be particularly informative in this context.
Several limitations of this study should be acknowledged. First, the association between behavioural rigidity and functional connectivity between the NAc and other brain regions was assessed using correlation-based analyses within a cross-sectional design, which precludes causal inference. Future research would benefit from applying causal or interventional approaches to determine whether NAc activity exerts a directional influence on motor and default mode network regions. Building on this, future studies could use deep brain stimulation in dementia populations to directly test whether stimulation of the NAc reduces its functional connectivity with motor and default mode networks, and in turn, attenuates behavioural rigidity. Moreover, repetitive behaviours are inherently heterogeneous, differing in complexity, phenomenology, assessment methods, reporting approaches and underlying mechanisms. As such, superficially similar prevalence rates reported in previous studies of AD and FTD may not necessarily indicate comparable behavioural profiles. By this view, our use of a pragmatic behavioural rigidity score to aggregate distinct features into a composite score represents something of a conceptual limitation. In addition, the CBI-R measure captures externally observable behaviours but does not index internalizing emotional drivers (e.g. anxiety) implicated in rigidity in disorders such as OCD and ASD. Future studies should adopt more fine-grained subdomain distinctions to improve phenotypic precision and mechanistic specificity. In response to this need, we recently developed and validated the Flexibility in Daily Life (FIDL) scale68: a novel self-report tool assessing cognitive and behavioural flexibility across five dimensions in everyday contexts. This instrument is expected to enable more precise characterization of rigidity-related symptoms beyond the scope of current clinical rating scales and, in turn, support more refined phenotyping and targeted interventions. Finally, SSRIs may modulate the expression and severity of stereotypical and rigid behaviours, however, medication data were not systematically collected for the present cohort. Future work incorporating detailed treatment information will be necessary to clarify these effects.
Conclusions
In conclusion, this study shows that transdiagnostic behavioural rigidity in dementia is associated with grey matter atrophy of the NAc and its increased functional connectivity with motor control regions. Our findings not only localize the NAc as a crucial hub in the origin of rigid behaviours in dementia but situate it in terms of a broader network implicated in habitual motor actions. Importantly, this work offers novel insights into potential intervention strategies, suggesting that modulating NAc function may help to mitigate behavioural rigidity in individuals with dementia. Our findings highlight the pervasive nature of rigid behaviours in dementia syndromes, identify the NAc as part of a wider network implicated in the pathophysiology of such disturbances, and point to the NAc as a potential treatment target to tackle these challenging symptoms.
Supplementary Material
Acknowledgements
We thank all participants and their families for their time and contribution to this study.
Contributor Information
Tao Chen, Brain and Mind Centre, The University of Sydney, Sydney, NSW 2050, Australia; School of Psychology, The University of Sydney, Sydney, NSW 2006, Australia.
Rebekah M Ahmed, Brain and Mind Centre, The University of Sydney, Sydney, NSW 2050, Australia; Central Clinical School, The University of Sydney, Sydney, NSW 2006, Australia.
Olivier Piguet, Brain and Mind Centre, The University of Sydney, Sydney, NSW 2050, Australia; School of Psychology, The University of Sydney, Sydney, NSW 2006, Australia.
Muireann Irish, Brain and Mind Centre, The University of Sydney, Sydney, NSW 2050, Australia; School of Psychology, The University of Sydney, Sydney, NSW 2006, Australia.
Data availability
The ethical requirement to ensure patient confidentiality precludes public archiving of our data. Researchers who would like to access the raw data should contact the corresponding author who will liaise with the ethics committee that approved the study, and accordingly, as much data that are required to reproduce the results will be released to the individual researcher. No part of the study procedures or analyses were preregistered prior to the research being undertaken.
Funding
This work was supported in part by funding to ForeFront, a collaborative research group dedicated to the study of frontotemporal dementia and motor neuron disease, from the National Health and Medical Research Council (NHMRC) (GNT1037746) and the Australian Research Council (ARC) Centre of Excellence in Cognition and its Disorders Memory Program (CE11000102). M.I. is supported by a Medical Research Future Fund Dementia, Aging, and Aged Care Grant from the Australian Government Department of Health and Aged Care (GNT2024329), and an NHMRC Investigator Grant (GNT2025228). O.P. is supported by an NHMRC Investigator Grant (GNT2008020).
Competing interests
The authors report no competing interests.
Supplementary material
Supplementary material is available at Brain online.
References
- 1. Poljac E, Hoofs V, Princen MM, Poljac E. Understanding behavioural rigidity in autism spectrum conditions: The role of intentional control. J Autism Dev Disord. 2017;47:714–727. [DOI] [PubMed] [Google Scholar]
- 2. Uddin LQ. Cognitive and behavioural flexibility: Neural mechanisms and clinical considerations. Nat Rev Neurosci. 2021;22:167–179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Restrepo-Martínez M, Ramirez-Bermudez J, Chacon-Gonzalez J, Ruiz-Garcia R, Malik R, Finger E. Defining repetitive behaviours in frontotemporal dementia. Brain. 2024;147:1149–1165. [DOI] [PubMed] [Google Scholar]
- 4. Langen M, Durston S, Kas MJH, van Engeland H, Staal WG. The neurobiology of repetitive behavior: …and men. Neurosci Biobehav Rev. 2011;35:356–365. [DOI] [PubMed] [Google Scholar]
- 5. Horne K, Irish M. Reconciling profiles of reward-seeking versus reward-restricted behaviours in frontotemporal dementia. Brain Commun. 2023;5:fcad045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Voon V, Derbyshire K, Rück C, et al. Disorders of compulsivity: A common bias towards learning habits. Mol Psychiatry. 2015;20:345–352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Polin C, Gellé T, Auditeau E, Adou C, Clément J-P, Calvet B. Repetitive behaviors in Alzheimer’s disease: A systematic review and meta-analysis. J Alzheimers Dis. 2023;96:483–497. [DOI] [PubMed] [Google Scholar]
- 8. Nyatsanza S, Shetty T, Gregory C, Lough S, Dawson K, Hodges JR. A study of stereotypic behaviours in Alzheimer’s disease and frontal and temporal variant frontotemporal dementia. J Neurol Neurosurg Psychiatry. 2003;74:1398–1402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kamminga J, O'Callaghan C, Hodges JR, Irish M. Differential prospective memory profiles in frontotemporal dementia syndromes. J Alzheimers Dis. 2014;38:669–679. [DOI] [PubMed] [Google Scholar]
- 10. Snowden JS, Bathgate D, Varma A, Blackshaw A, Gibbons ZC, Neary D. Distinct behavioural profiles in frontotemporal dementia and semantic dementia. J Neurol Neurosurg Psychiatry. 2001;70:323–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Bozeat S, Gregory CA, Ralph MAL, Hodges JR. Which neuropsychiatric and behavioural features distinguish frontal and temporal variants of frontotemporal dementia from Alzheimer's disease? J Neurol Neurosurg Psychiatry. 2000;69:178–186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ames D, Cummings JL, Wirshing WC, Quinn B, Mahler M. Repetitive and compulsive behavior in frontal lobe degenerations. J Neuropsychiatry Clin Neurosci. 1994;6:100–113. [DOI] [PubMed] [Google Scholar]
- 13. Kumamoto K, Arai Y, Hashimoto N, Ikeda M, Mizuno Y, Washio M. Problems family caregivers encounter in home care of patients with frontotemporal lobar degeneration. Psychogeriatrics. 2004;4:33–39. [Google Scholar]
- 14. Robbins TW, Banca P, Belin D. From compulsivity to compulsion: The neural basis of compulsive disorders. Nat Rev Neurosci. 2024;25:313–333. [DOI] [PubMed] [Google Scholar]
- 15. Chen T, Ahmed RM, Narasimhan M, et al. Anterior insula drives progressive structural brain network atrophy in the behavioural variant of frontotemporal dementia. Hum Brain Mapp. 2025;46:e70374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bang J, Spina S, Miller BL. Frontotemporal dementia. Lancet. 2015;386:1672–1682. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Shaw SR, El-Omar H, Roquet D, et al. Uncovering the prevalence and neural substrates of anhedonia in frontotemporal dementia. Brain. 2021;144:1551–1564. [DOI] [PubMed] [Google Scholar]
- 18. Ulugut Erkoyun H, Groot C, Heilbron R, et al. A clinical-radiological framework of the right temporal variant of frontotemporal dementia. Brain. 2020;143:2831–2843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Chokesuwattanaskul A, Jiang H, Bond RL, et al. The architecture of abnormal reward behaviour in dementia: Multimodal hedonic phenotypes and brain substrate. Brain Commun. 2023;5:fcad027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Perry DC, Sturm VE, Seeley WW, Miller BL, Kramer JH, Rosen HJ. Anatomical correlates of reward-seeking behaviours in behavioural variant frontotemporal dementia. Brain. 2014;137:1621–1626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Horne K, Ahmed RM, Piguet O, Irish M. Establishing the link between motivational disturbances and behavioural rigidity in frontotemporal dementia. Eur J Neurol. 2024;31:e16132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Perry DC, Kramer JH. Reward processing in neurodegenerative disease. Neurocase. 2015;21:120–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Woolley JD, Gorno-Tempini M-L, Seeley WW, et al. Binge eating is associated with right orbitofrontal-insular-striatal atrophy in frontotemporal dementia. Neurology. 2007;69:1424–1433. [DOI] [PubMed] [Google Scholar]
- 24. Yi DS, Bertoux M, Mioshi E, Hodges JR, Hornberger M. Fronto-striatal atrophy correlates of neuropsychiatric dysfunction in frontotemporal dementia (FTD) and Alzheimer's disease (AD). Dement Neuropsychol. 2013;7:75–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Ahmed RM, Devenney EM, Irish M, et al. Neuronal network disintegration: Common pathways linking neurodegenerative diseases. J Neurol Neurosurg Psychiatry. 2016;87:1234–1241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Irish M, Piguet O, Hodges JR. Self-projection and the default network in frontotemporal dementia. Nat Rev Neurol. 2012;8:152–161. [DOI] [PubMed] [Google Scholar]
- 27. Yin HH, Knowlton BJ. The role of the basal ganglia in habit formation. Nat Rev Neurosci. 2006;7:464–476. [DOI] [PubMed] [Google Scholar]
- 28. Josephs KA, Whitwell JL, Jack CR, Jr. Anatomic correlates of stereotypies in frontotemporal lobar degeneration. Neurobiol Aging. 2008;29:1859–1863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Akkermans SEA, Rheinheimer N, Bruchhage MMK, et al. Frontostriatal functional connectivity correlates with repetitive behaviour across autism spectrum disorder and obsessive-compulsive disorder. Psychol Med. 2019;49:2247–2255. [DOI] [PubMed] [Google Scholar]
- 30. Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, Buckner RL. Functional-anatomic fractionation of the brain's default network. Neuron. 2010;65:550–562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Figee M, Luigjes J, Smolders R, et al. Deep brain stimulation restores frontostriatal network activity in obsessive-compulsive disorder. Nat Neurosci. 2013;16:386–387. [DOI] [PubMed] [Google Scholar]
- 32. Wedderburn C, Wear H, Brown J, et al. The utility of the Cambridge Behavioural Inventory in neurodegenerative disease. J Neurol Neurosurg Psychiatry. 2008;79:500–503. [DOI] [PubMed] [Google Scholar]
- 33. Rascovsky K, Hodges JR, Knopman D, et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain. 2011;134:2456–2477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Gorno-Tempini ML, Hillis AE, Weintraub S, et al. Classification of primary progressive aphasia and its variants. Neurology. 2011;76:1006–1014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Ulugut H, Younes K, Montembeault M, et al. Clinical recognition of frontotemporal dementia with right temporal predominance: A consensus statement from the International Working Group. Commun Med (Lond). 2025;5:523. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement. 2011;7:263–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Mioshi E, Dawson K, Mitchell J, Arnold R, Hodges JR. The Addenbrooke's Cognitive Examination Revised (ACE-R): A brief cognitive test battery for dementia screening. Int J Geriatr Psychiatry. 2006;21:1078–1085. [DOI] [PubMed] [Google Scholar]
- 38. Hsieh S, Schubert S, Hoon C, Mioshi E, Hodges JR. Validation of the Addenbrooke's cognitive examination III in frontotemporal dementia and Alzheimer's disease. Dement Geriatr Cogn Disord. 2013;36:242–250. [DOI] [PubMed] [Google Scholar]
- 39. So M, Foxe D, Kumfor F, et al. Addenbrooke’s Cognitive Examination III: Psychometric characteristics and relations to functional ability in dementia. J Int Neuropsychol Soc. 2018;24:854–863. [DOI] [PubMed] [Google Scholar]
- 40. Mioshi E, Hsieh S, Savage S, Hornberger M, Hodges JR. Clinical staging and disease progression in frontotemporal dementia. Neurology. 2010;74:1591–1597. [DOI] [PubMed] [Google Scholar]
- 41. Love J, Selker R, Marsman M, et al. JASP: Graphical statistical software for common statistical designs. J Stat Softw. 2019;88:1–17. [Google Scholar]
- 42. van Buuren S, Groothuis-Oudshoorn K. Mice: Multivariate imputation by chained equations in R. J Stat Softw. 2011;45:1–67. [Google Scholar]
- 43. Yan C-G, Wang X-D, Zuo X-N, Zang Y-F. DPABI: Data processing & analysis for (resting-state) brain imaging. Neuroinformatics. 2016;14:339–351. [DOI] [PubMed] [Google Scholar]
- 44. Kohls G, Antezana L, Mosner MG, Schultz RT, Yerys BE. Altered reward system reactivity for personalized circumscribed interests in autism. Mol Autism. 2018;9:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Figee M, Vink M, de Geus F, et al. Dysfunctional reward circuitry in obsessive-compulsive disorder. Biol Psychiatry. 2011;69:867–874. [DOI] [PubMed] [Google Scholar]
- 46. Kohls G, Schulte-Rüther M, Nehrkorn B, et al. Reward system dysfunction in autism spectrum disorders. Soc Cogn Affect Neurosci. 2013;8:565–572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Nachev P, Kennard C, Husain M. Functional role of the supplementary and pre-supplementary motor areas. Nat Rev Neurosci. 2008;9:856–869. [DOI] [PubMed] [Google Scholar]
- 48. Patra A, Kaur H, Chaudhary P, Asghar A, Singal A. Morphology and morphometry of human paracentral lobule: An anatomical study with its application in neurosurgery. Asian J Neurosurg. 2021;16:349–354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Supekar K, Ryali S, Mistry P, Menon V. Aberrant dynamics of cognitive control and motor circuits predict distinct restricted and repetitive behaviors in children with autism. Nat Commun. 2021;12:3537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Hau J, Kohli JS, Shryock I, et al. Supplementary and premotor aspects of the corticospinal tract show links with restricted and repetitive behaviors in middle-aged adults with autism spectrum disorder. Cereb Cortex. 2021;31:3962–3972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Gillan CM, Papmeyer M, Morein-Zamir S, et al. Disruption in the balance between goal-directed behavior and habit learning in obsessive-compulsive disorder. Am J Psychiatry. 2011;168:718–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Cavanna AE, Trimble MR. The precuneus: A review of its functional anatomy and behavioural correlates. Brain. 2006;129:564–583. [DOI] [PubMed] [Google Scholar]
- 53. Utevsky AV, Smith DV, Huettel SA. Precuneus is a functional core of the default-mode network. J Neurosci. 2014;34:932–940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Dadario NB, Sughrue ME. The functional role of the precuneus. Brain. 2023;146:3598–3607. [DOI] [PubMed] [Google Scholar]
- 55. Zhang S, Li CS. Functional connectivity mapping of the human precuneus by resting state fMRI. NeuroImage. 2012;59:3548–3562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Langen M, Kas MJH, Staal WG, van Engeland H, Durston S. The neurobiology of repetitive behavior: Of mice…. Neurosci Biobehav Rev. 2011;35:345–355. [DOI] [PubMed] [Google Scholar]
- 57. Rahman S, Sahakian BJ, Hodges JR, Rogers RD, Robbins TW. Specific cognitive deficits in mild frontal variant frontotemporal dementia. Brain. 1999;122:1469–1493. [DOI] [PubMed] [Google Scholar]
- 58. Derosiere G, Shokur S, Vassiliadis P. Reward signals in the motor cortex: From biology to neurotechnology. Nat Commun. 2025;16:1307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Friederich H-C, Herzog W. Cognitive-behavioral flexibility in anorexia nervosa. In: Adan RAH, Kaye WH, eds. Behavioral neurobiology of eating disorders. Springer Berlin Heidelberg; 2011:111–123. [DOI] [PubMed] [Google Scholar]
- 60. Lung SLM, Picard È, Soulières I, Bertone A. Identifying the functions of restricted and repetitive behaviours and interests in autism: A scoping review. Res Autism Spectr Disord. 2024;117:102458. [Google Scholar]
- 61. Horne K, de Andrade Saraiva L, de Souza LC, Irish M. Social interaction as a unique form of reward – insights from healthy ageing and frontotemporal dementia. Neurosci Biobehav Rev. 2025;172:106128. [DOI] [PubMed] [Google Scholar]
- 62. Soomro GM, Altman DG, Rajagopal S, Oakley Browne M. Selective serotonin re-uptake inhibitors (SSRIs) versus placebo for obsessive compulsive disorder (OCD). Cochrane Database Syst Rev. 2008;2008:CD001765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63. Stein DJ, Costa DLC, Lochner C, et al. Obsessive–compulsive disorder. Nat Rev Dis Primers. 2019;5:52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64. Huber N, Korhonen S, Hoffmann D, et al. Deficient neurotransmitter systems and synaptic function in frontotemporal lobar degeneration—Insights into disease mechanisms and current therapeutic approaches. Mol Psychiatry. 2022;27:1300–1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65. Huey ED, Putnam KT, Grafman J. A systematic review of neurotransmitter deficits and treatments in frontotemporal dementia. Neurology. 2006;66:17–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66. Murley AG, Rowe JB. Neurotransmitter deficits from frontotemporal lobar degeneration. Brain. 2018;141:1263–1285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67. Hughes LE, Rittman T, Regenthal R, Robbins TW, Rowe JB. Improving response inhibition systems in frontotemporal dementia with citalopram. Brain. 2015;138:1961–1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68. Horne K, Chen T, Irish M. Development of the flexibility in daily life scale to measure multidimensional cognitive and behavioural flexibility in health and disease. Br J Clin Psychol. 2025;64:315–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The ethical requirement to ensure patient confidentiality precludes public archiving of our data. Researchers who would like to access the raw data should contact the corresponding author who will liaise with the ethics committee that approved the study, and accordingly, as much data that are required to reproduce the results will be released to the individual researcher. No part of the study procedures or analyses were preregistered prior to the research being undertaken.
