Overall cognitive status was evaluated using the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005), a brief screening tool designed to capture subtle cognitive changes. The MoCA was preferred over the MMSE because of its higher sensitivity to early deficits, particularly in executive domains that are often impaired in the initial stages of bvFTD. The validated Greek version was administered (Poptsi et al., 2019), and total scores (range: 0–30) were used in the analyses, with higher values indicating better cognitive performance.

Perceptual organization, visuospatial processing, and visuoconstructional skills were assessed using the Taylor Complex Figure Test (Taylor, 1959). Participants were instructed to reproduce a complex geometric figure as accurately as possible during the copy condition, within a 5-min time limit. Performance was scored based on the accuracy and spatial organization of 18 design elements, yielding a total score ranging from 0 to 36, which was included in the statistical analyses.

To assess long-term episodic verbal memory, the Story Memory Test from the “Neuropsychological Battery” (Kosmidis et al., 2011), developed by the Laboratory of Cognitive Neuroscience at the School of Psychology, AUTh, was administered. The Story Memory Test evaluates the capacity to encode and retrieve semantically related information. Participants heard a short story containing 16 informational units and were asked to recall it immediately and again after a 30-min delay filled with non-verbal tasks. For the purposes of this study, only the delayed recall score (range: 0–16) was analyzed, providing an index of verbal episodic long-term memory.

Executive abilities were assessed using both traditional and computerized measures. The Greek version of the Trail Making Test - Part B (TMT-B) (Reitan, 1958; Zalonis et al., 2008) was administered to evaluate cognitive flexibility and set-shifting abilities. Participants were instructed to connect numbers and letters in alternating ascending order (1-A-2-B, etc.) as quickly and accurately as possible. The completion time in seconds served as the performance score, with shorter times reflecting better executive control.

Additionally, a computerized assessment of executive functioning was conducted using the REMEDES for Alzheimer-Revised (R4Alz-R) battery (Poptsi et al., 2024a; Poptsi et al., 2024b), a digital tool designed to evaluate core components of cognitive control. The R4Alz-R was selected for its ecological validity, sensitivity to executive dysfunction, and minimal cultural or educational bias. Three modules were administered: (i) Working Memory Capacity, including short-term storage, processing, and updating subtests; (ii) Attentional Control, assessing sustained, selective, and divided attention; and (iii) Executive Functioning, including inhibition and task/rule switching subtests. Performance scores from each module were standardized (z-scores) and combined to yield a total composite score, with higher values indicating poorer performance.

Theory of mind and social inferencing abilities were measured using the Greek adaptation of the Awareness of Social Inference Test - Short (TASIT-S) (McDonald et al., 2017; Tsentidou et al., 2021). Specifically, Part 2 of the test, the Test of Social Inference—Minimal (SI-M), was administered to evaluate participants’ capacity to infer others’ mental states and communicative intentions. The SI-M consists of brief video vignettes depicting both sincere and sarcastic social exchanges, requiring participants to interpret subtle verbal and nonverbal cues such as facial expressions, gestures, and tone of voice. After viewing each of the nine scenarios, participants answered “yes/no” questions regarding the characters’ actions, statements, thoughts, and feelings. Total scores ranged from 0 to 36, reflecting the number of correct responses, with higher values indicating better social inferencing abilities.

Personality characteristics were evaluated using the Greek adaptation of Goldberg’s International Personality Item Pool (IPIP) Big-Five Questionnaire (Goldberg et al., 2005; Ypofanti et al., 2015). This 50-item measure assesses five major personality dimensions in accordance with the Big Five theory of personality (Costa and McCrae, 1992): Extraversion, Agreeableness, Conscientiousness, Emotional Stability-Neuroticism, and Intellect/Openness to Experience, rated on a 5-point Likert scale (1 = very inaccurate to 5 = very accurate). Given the reduced insight typical of bvFTD (Bracca et al., 2024), all questionnaire items were rephrased in third-person form (“he/she”) and completed by knowledgeable informants familiar with the patient’s long-term behavior and personality. Caregivers completed the questionnaire twice: first by retrospectively rating the patient’s premorbid personality (approximately 20 years prior to symptom onset), and second by rating the patient’s current personality traits following disease onset. Composite scores were calculated for each personality dimension (range: 10–50), and difference scores between premorbid and current ratings were computed for all five traits. These difference scores, reflecting the magnitude of disease-related longitudinal personality change, were included in the analyses of the present study.

Behavioral symptoms characteristic of bvFTD were quantified using the Greek version of the Frontal Behavioral Inventory (FBI) (Kertesz et al., 1997; Konstantinopoulou et al., 2012), a 24-item informant-based questionnaire specifically designed to capture behavioral changes associated with FTD. The FBI includes two subscales assessing (i) negative symptoms associated with executive and motivational deficits (e.g., apathy, inflexibility, loss of insight), and (ii) positive symptoms reflecting behavioral disinhibition (e.g., impulsivity, social inappropriateness, hyperorality). Each item is rated on a 4-point scale (0 = none to 3 = severe), resulting in two subscale scores ranging from 0 to 36. Higher scores indicate more pronounced behavioral disturbances.

Disease stage and overall functional impairment were assessed with the Greek version of the Frontotemporal Dementia Rating Scale (FRS) (Mioshi et al., 2010; Maiovis et al., 2016) a staging instrument specifically developed for FTD. The FRS includes 30 items evaluating both behavioral and functional domains. After scoring, the raw total is divided by the number of applicable items and multiplied by 100 to derive a percentage score. Higher FRS percentages reflect milder disease severity and better overall functioning.

Disease severity was also assessed using the Clinical Dementia Rating (CDR) scale (Morris, 1993), which evaluates cognitive and functional performance across six domains: memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal care. The CDR yields both a Global Score (CDR-GS; 0–3) and a Sum of Boxes score (CDR-SB; 0–18). To enhance sensitivity to FTLD-specific symptomatology, the FTLD-Modified CDR (FTLD-CDR) (Knopman et al., 2008) was also employed, adding two additional domains - language and behavior - and providing a composite Sum of Boxes score (0–24). Higher scores indicate greater overall impairment.

Of all measured lobes and BAs, only those demonstrating hypoperfusion greater than 3 standard deviations below the mean, compared to normative data, were retained for further analysis. These areas included: BA 5 R, BA 6 L, BA 8 R, BA 11 R, BA 28 L, BA 31 L, BA 32 R, BA 37 L, BA 40 R, and BA 46 R.

Statistically significant negative correlations were identified between FBI negative symptoms and rCBF in the right BA 8 (r = −0.469, p = 0.002). Additionally, FBI positive symptoms showed statistically significant negative correlations with rCBF in the right BA 46 (r = −0.506, p = 0.002).

Correlational analyses were performed to examine the relationship between brain perfusion and personality changes in individuals with bvFTD. Statistically significant negative correlations were identified between the magnitude of decline in conscientiousness and rCBF in the left BA 6 (r = −0.541, p = 0.001) and in the left BA 31 (r = −0.571, p = 0.001). Additionally, statistically significant negative correlations were identified between the magnitude of decline in extraversion and rCBF in the right BA 8 (r = −0.532, p = 0.002), in the right BA 32 (r = −0.643, p = 0.001), and in the left BA 37 (r = −0.443, p = 0.002). No other relationships between personality traits and rCBF survived Bonferroni correction for multiple comparisons.

The MoCA test did not demonstrate any statistically significant correlations with rCBF in any of the brain lobes or specific BAs.

The copy condition of the Taylor Complex Figure Test demonstrated a significant positive correlation with rCBF in the right BA 5 (r = 0.408, p = 0.002).

The delayed recall of the Story Memory test showed significant positive correlations with rCBF in the left BA 28 (r = 0.521, p = 0.001) and left BA 31 (r = 0.475, p = 0.002).

Performance on the TMT-Part B showed significant negative correlations with rCBF in the left BA 6 (r = −0.527, p = 0.001) and left BA 31 (r = −0.531, p = 0.001).

Additionally, the R4Alz-R Battery total score showed significant negative correlations with rCBF in the right BA 5 (r = −0.472, p = 0.002) and the right BA 8 (r = −0.565, p = 0.001).

Finally, the TASIT-S Part 2 demonstrated significant positive correlations with rCBF in the right BA 11 (r = 0.461, p = 0.002), right BA 32 (r = 0.431, p = 0.002), left BA 37 (r = 0.561, p = 0.001), right BA 40 (r = 0.471, p = 0.002), and right BA 46 (r = 0.442, p = 0.002).

Correlational analyses were performed to examine the relationship between rCBF and the FRS scale. Significant positive associations were identified between the FRS percentage and rCBF in the right BA 8 (r = 0.490, p = 0.002).

An overview of the statistically significant associations between rCBF and clinical variables is presented in Table 2. Results for non-significant relationships between lobar rCBF and clinical variables are reported in Supplementary material.

In addition, a conceptual schematic brain map illustrating the BAs showing statistically significant correlations with specific clinical variables is presented in Figure 1, offering a visual overview of the key brain-behavior associations identified in the study.

The absence of significant correlations between global cognitive performance and rCBF is consistent with previous SPECT and PET studies indicating that global indices such as the MMSE or MoCA are relatively insensitive to the anteriorly distributed dysfunction typical of bvFTD (Borroni et al., 2010; Deutsch et al., 2016). These screening measures predominantly assess posterior cortical functions - orientation, visuospatial ability, and language - that tend to be relatively preserved in the early stages of bvFTD. The lack of association in the current results therefore reflects both the selectivity of bvFTD-related frontal dysfunction and the limitations of global tests for capturing the subtle yet clinically significant impairments in executive and socioemotional domains.

A positive correlation between visuoconstructional abilities and rCBF in a right parietal region (BA 5) highlights the essential contribution of this area to spatial integration and visuomotor organization in bvFTD. BA 5, located in the superior parietal lobule, is crucial for coordinating spatial and motor representations required for complex figure construction (Vandenberghe and Gillebert, 2009). This region also maintains functional connections with frontal executive areas within the frontoparietal control network, which supports higher-order planning, organization, and strategy use (Vandenberghe and Gillebert, 2009). The observed hypoperfusion in BA 5 may therefore reflect both posterior cortical involvement and disrupted frontoparietal interactions, leading to impaired integration of spatial and executive components necessary for effective visuoconstructional processing.

Memory performance, as assessed by delayed recall of a Story Memory test, was positively associated with rCBF in left BAs 28 and 31, corresponding, respectively, to the parahippocampal region and posterior cingulate cortex. These structures are integral to the retrieval and contextualization of episodic memories (Aminoff et al., 2013; Leech and Sharp, 2013). While bvFTD is not classically characterized by primary memory deficits, dysfunction in medial temporal and limbic regions has been reported to contribute to episodic memory impairments (Perry et al., 2017). The parahippocampal cortex is critically involved in encoding and retrieval of contextual and spatial aspects of episodic memory, supporting the integration of information into coherent memory traces (Aminoff et al., 2013). Additionally, the posterior cingulate cortex has been linked to self-referential processing and autobiographical recall (Leech and Sharp, 2013), and its involvement here may partly reflect the nature of the memory assessment used in this study - a story-based test that might have unconsciously led participants to relate the material to their personal experiences.

Executive performance correlated with perfusion in both left and right frontal and parietal regions, including the bilateral frontal cortex (BAs 6 and 8), left posterior cingulate (BA 31), and right parietal (BA 5). These areas are critically implicated in planning, task switching, and response monitoring (Stuss and Alexander, 2007; Dadario et al., 2023). Notably, hypoperfusion in the right BA 8, located in the superior prefrontal gyrus just anterior to the premotor cortices, has been associated with decision-making under uncertainty, oculomotor control, motor planning, selection among competing stimuli, and adaptive cognitive control (Dadario et al., 2023). Additionally, reduced perfusion in the left premotor region (BA 6), situated anterior to the primary motor cortex, may reflect impairments in the generation, sequencing, and execution of goal-directed actions, a deficit frequently observed in bvFTD. The associations involving BA 31 further underscore the role of cingulate regions in sustaining attentional control and performance monitoring during complex cognitive tasks (Leech and Sharp, 2013). Interestingly, although involvement of BA 32 - a key node of the cingulo-opercular action-mode network implicated in executive control (Dosenbach et al., 2025)—swas anticipated, we did not observe statistically significant associations in our cohort. This absence may reflect the modest sample size, heterogeneity in perfusion patterns, or task-specific functional engagement, and does not preclude a role for BA 32 in supporting sustained goal-directed attention and executive control. Taken together, these results indicate that hypoperfusion in distributed but regionally specific frontal and parietal cortices is linked to poorer executive performance in bvFTD. These findings are also in line with prior SPECT studies reporting correlations between executive measures and rCBF in frontal and parietal regions (Borroni et al., 2010), reinforcing the view that executive dysfunction in bvFTD reflects regionally localized frontal and parietal hypoperfusion, but also potential disruption of frontoparietal functional systems supporting higher-order cognitive control.

The most extensive brain-behavior associations were observed for social cognition, which correlated with rCBF in several regions including the right orbitofrontal cortex (BA 11), right anterior cingulate (BA 32), left inferior temporal area (BA 37), right inferior parietal cortex (BA 40), and right dorsolateral prefrontal cortex (BA 46). Each of these regions has been independently implicated in core social cognitive processes such as theory of mind, perspective taking, sarcasm perception, emotional evaluation, and empathic processing (Frith and Frith, 2006; Rankin et al., 2009; Ibañez and Manes, 2012). The orbitofrontal cortex (BA 11) plays a central role in evaluating the emotional and reward-related value of social stimuli and in guiding socially appropriate decision-making (Rolls, 2004). Hypoperfusion in this region has been consistently associated with impaired emotional regulation, altered social judgment, and disinhibited behavior in bvFTD (Rolls, 2004). Moreover, the involvement of the right dorsolateral prefrontal cortex (BA 46) highlights the contribution of executive control mechanisms to social cognition, including the regulation of social responses, maintenance of social rules, and flexible adjustment of behavior in dynamic interpersonal contexts (Miller and Cohen, 2001; Rankin et al., 2009). Similarly, the anterior cingulate (BA 32) is critically involved in affective monitoring and modulation, motivational drive, and the integration of emotional signals with goal-directed behavior (Seeley et al., 2007; Njomboro et al., 2012; Leech and Sharp, 2013). Dysfunction within this region has been linked to apathy, reduced empathy, and diminished social engagement in FTLD syndromes (Seeley et al., 2007; Njomboro et al., 2012; Leech and Sharp, 2013). Beyond frontal and cingulate regions, significant associations were also identified in temporal and parietal cortices. The inferior temporal cortex (BA 37) contributes to high-level visual and semantic processing of socially relevant stimuli, including facial identity and emotional expressions, while the inferior parietal cortex (BA 40) supports the integration of multimodal contextual and somatosensory information necessary for inferring others’ mental states (Saxe and Kanwisher, 2003; Olson et al., 2007). The observed correlations between TASIT-S performance and rCBF in these regions suggests that effective social inference relies on intact perceptual, semantic, executive, and contextual processing, which may become compromised as perfusion declines. Although causal interpretations cannot be drawn from the present correlational design, the observed pattern indicates that reduced perfusion across discrete but interconnected frontal, temporal, limbic, and parietal regions may underlie the complex social cognitive deficits that represent a core and predictive clinical hallmark of bvFTD (Chatzidimitriou et al., 2025b).

Behavioral disturbances in bvFTD were significantly associated with lower perfusion in right prefrontal areas, specifically BAs 8 and 46. This pattern is consistent with prior findings linking right-sided frontal hypoperfusion to behavioral disinhibition, apathy, and loss of social insight (Warriner and Velikonja, 2006; Konstantinopoulou et al., 2024). Negative behavioral symptoms, such as apathy and emotional flatness, correlated with perfusion in BA 8, a region implicated in sustained attention, initiation of goal-directed behavior, and adaptive cognitive control (Dadario et al., 2023). Hypoperfusion in this area may reflect a diminished capacity to initiate and maintain goal-directed activity, contributing to the passivity and motivational deficits frequently observed in bvFTD (Chatzidimitriou et al., 2025a). In contrast, positive behavioral symptoms, including impulsivity and disinhibition, correlated with perfusion in BA 46, a region encompassing the dorsolateral prefrontal cortex, which supports inhibitory control and behavioral modulation. These findings suggest that different aspects of behavioral dysregulation in bvFTD may be associated with dysfunction in distinct prefrontal regions, with right BA 8 contributing primarily to motivational drive and initiation of goal-directed actions, and right BA 46 supporting behavioral control. Overall, the current results align with previous SPECT and PET studies identifying right frontal dysfunction as a prominent neural correlate of both apathic and disinhibited behavioral phenotypes in bvFTD (Torregrossa et al., 2008; Valotassiou et al., 2020; Konstantinopoulou et al., 2024).

A particularly novel contribution of this study is the identification of specific rCBF correlates of disease-related personality changes in bvFTD, a domain rarely examined in brain perfusion imaging studies. In the present cohort, changes in conscientiousness showed positive associations with perfusion in the left premotor cortex (BA 6) and left dorsal posterior cingulate cortex (BA 31), whereas changes in extraversion correlated with perfusion in a right prefrontal region (BA 8), right dorsal anterior cingulate (BA 32), and left temporal cortex (BA 37).

The link between conscientiousness and perfusion in premotor and cingulate regions is consistent with the role of these areas in the planning and regulation of purposeful behavior (Tanaka et al., 2005; Leech and Sharp, 2013). Conscientiousness involves self-discipline, orderliness, dutifulness and goal maintenance, characteristics that depend on intact premotor and cingulate function (Costa and McCrae, 1992; Wilson et al., 2015). Specifically, the premotor cortex (BA 6) is critical for planning, initiating, and executing purposeful actions, supporting structured, goal-directed behavior. In addition, BA 6 contributes to attentional control and the temporal maintenance of task-relevant information in working memory, integrating spatial and verbal representations necessary for achieving complex goals and guiding adaptive behavior (Tanaka et al., 2005). On the other hand, the dorsal posterior cingulate cortex (BA 31) is traditionally implicated in self-referential processing and monitoring ongoing actions, integrating internal goals with external demands, and aligning behavior with current tasks and long-term objectives (Leech and Sharp, 2013). Importantly, this region is a key node of the default mode network, showing increased activity during autobiographical memory retrieval, future planning, and internally-directed cognition, suggesting a central role in regulating attention toward internally relevant goals. Together, these regions facilitate goal-directed, adaptive self-regulation (Leech and Sharp, 2013). Prior studies have suggested that higher levels of conscientiousness may act as a protective factor associated with better cognitive and functional trajectories in later life (Wilson et al., 2015); the current findings provide preliminary evidence of possible regional substrates underlying this effect.

Extraversion change correlated positively with perfusion in right prefrontal cortex (BA 8), right dorsal anterior cingulate (BA 32), and left temporal cortex (BA 37). The right prefrontal region (BA 8) plays a critical role in executive control, attentional shifting, and the organization of goal-directed behavior (Dadario et al., 2023). Activity within this region supports the initiation and maintenance of actions in response to motivationally relevant environmental cues and facilitates adaptive decision-making in uncertain situations, such as socially complex contexts. The dorsal anterior cingulate cortex is involved in emotional engagement, reward processing, and motivational drive (Bush et al., 2002), processes that underlie sociability and approach behavior. The left temporal cortex, encompassing BA 37, includes the fusiform and inferior temporal gyri, which are critical for high-level visual processing (Ardila et al., 2015), including facial recognition and the interpretation of socially relevant visual cues. Adjacent parahippocampal regions also contribute to contextual and mnemonic aspects of social interactions (Aminoff et al., 2013), by linking social experiences to memory and emotional context. Hypoperfusion in these areas may therefore contribute to the blunted affect, social withdrawal, reduced motivation, and impaired interpersonal processing frequently observed in individuals with bvFTD.

Overall, these findings provide novel SPECT-based evidence linking regional brain perfusion to the magnitude of personality change in bvFTD, highlighting the functional neuroanatomical substrates underlying disease-related alterations in conscientiousness and extraversion. The integration of frontal executive, cingulate evaluative, and temporal mnemonic processes appears essential for maintaining adaptive personality features, and disruption of these regions may help explain the pronounced personality changes characteristic of bvFTD. It is recommended that future research examine personality-related perfusion correlates in other clinical populations as well as in healthy adults, to more clearly delineate the neural underpinnings of personality traits and clarify whether these associations are disease-specific or reflect broader brain-personality relationships.

Disease severity, as assessed by the FRS, was positively associated with perfusion in the right superior prefrontal cortex (BA 8). This result indicates that higher perfusion in this region corresponds to less functional impairment. The right superior frontal cortex has been implicated in strategic planning and sustained attention (Dadario et al., 2023), abilities essential for maintaining independence in daily life. Prior SPECT studies have similarly linked disease severity and functional decline in bvFTD to reduced perfusion in frontal areas (Borroni et al., 2009; Konstantinopoulou et al., 2024; Chatzidimitriou et al., 2025a). The current finding reinforces the importance of prefrontal integrity for global clinical functioning and suggests that perfusion in this region could serve as a potential biomarker of disease progression.

The present study employed a fine-grained, BA-based analysis of SPECT data in combination with a comprehensive clinical characterization, encompassing cognitive, behavioral, personality, and disease-severity measures. This integrative approach enabled a multidimensional examination of brain-behavior relationships in bvFTD and allowed for anatomically specific associations between rCBF and distinct clinical features. By moving beyond global cognitive indices and focusing on domain-specific clinical correlates, this study provides a more nuanced understanding of the functional neuroanatomy underlying the heterogeneous clinical presentation of bvFTD.

However, several limitations should be acknowledged. First, the relatively small sample size, although comparable to that of other neuroimaging studies in bvFTD, significantly limits the generalizability of the findings and may reduce statistical power. Additionally, the cross-sectional and correlational nature of the design precludes causal inferences regarding the directionality of the observed associations; consequently, the reported relationships should be interpreted as indicative rather than explanatory. Longitudinal studies are required to determine whether regional perfusion alterations precede, accompany, or follow clinical deterioration and to clarify their potential role as biomarkers of disease progression.

Moreover, in this study, disease severity and early-stage characterization were determined exclusively based on caregiver-reported estimates of symptom duration, which are inherently prone to bias. Standardized clinical staging criteria, such as the CDR score, were not employed for eligibility, which limits the precision of classifying disease stage. Consequently, statements referring to “early-stage” bvFTD should be interpreted with caution, and future studies would benefit from incorporating formalized staging criteria to reduce subjectivity and improve reproducibility.

Additionally, although this study provides new insights into the neural correlates of personality changes in bvFTD, it is important to note that these changes were assessed retrospectively using caregiver reports referencing a 20-year premorbid interval, which may be subject to recall bias. There are currently no established paradigms for such assessments in bvFTD research, making this approach novel. The extended premorbid timeframe was chosen to minimize the likelihood that subtle, prodromal behavioral changes influenced caregiver ratings. Nevertheless, this retrospective method is inherently prone to memory inaccuracies (including potential halo-horn effects), which may artificially inflate observed correlations. Future studies would benefit from exploring different premorbid reference windows, implementing improved anchoring strategies, or incorporating complementary objective measures to enhance the reliability, validity, and objectivity of premorbid personality assessments in bvFTD.

Another considerable methodological limitation of this study is the selection of BAs based on hypoperfusion greater than 3 standard deviations in at least 50% of the sample. While this approach allowed us to focus on the most severely affected and clinically relevant regions and to ensure adequate statistical power, it may have excluded areas with subtler perfusion changes that could also contribute to cognitive, behavioral, or personality alterations. Future studies with larger samples could explore more liberal thresholds or alternative dimensionality reduction strategies.

Furthermore, an additional methodological limitation of the present study relates to the exclusive use of functional imaging data. While brain perfusion SPECT is sensitive to early functional changes and remains a widely accessible tool in clinical settings, it provides an indirect measure of neuronal activity that offers lower spatial resolution than other neuroimaging modalities, such as PET, which may constrain precise anatomical localization. In addition, partial volume correction was not applied in the imaging pipeline, and regional differences in cortical atrophy were not explicitly accounted for, which may complicate the interpretation of perfusion measures and potentially confound brain-behavior associations in a neurodegenerative population.

Moreover, the absence of complementary structural neuroimaging data, such as MRI-derived volumetric measures, has limited the ability to directly relate functional perfusion abnormalities to underlying patterns of cortical and subcortical atrophy. The integration of multimodal neuroimaging approaches - combining functional techniques such as SPECT with structural modalities like MRI - could have offered a more comprehensive characterization of the neurobiological substrates of bvFTD. Such approaches are particularly valuable for disentangling the relative contributions of structural degeneration and functional abnormalities to clinical symptomatology and for capturing complementary aspects of brain organization and pathology. Future studies would therefore benefit from leveraging multimodal imaging data to achieve a deeper and more integrative understanding of brain-behavior relationships in bvFTD.

Finally, the present study focused only on region-specific associations between rCBF and clinical measures. Beyond region-specific analyses, future research using brain SPECT imaging data may benefit from adopting network-level analytical approaches to better capture large-scale patterns of functional organization in bvFTD. Multivariate, data-driven techniques applied to rCBF measures - such as dimensionality reduction analyses - could help identify coordinated perfusion patterns across distributed brain regions, thereby providing valuable insights into the functional networks underlying cognitive, behavioral, and personality changes in bvFTD. Such approaches may be particularly informative in larger samples, where greater statistical power would facilitate the identification of stable network-level signatures. Leveraging SPECT imaging in this way could extend its utility beyond regional characterization, potentially offering a complementary perspective on the network-level functional alterations that characterize bvFTD.