NEW research (2026): Striatal and frontal signatures of social context and cost-benefit decision making in developmental stuttering

NEW research (2026): Striatal and frontal signatures of social context and cost-benefit decision making in developmental stuttering **doi:** [10.64898/2026.03.02.707906](https://www.biorxiv.org/content/10.64898/2026.03.02.707906v2.full) # Abstract Developmental stuttering is usually framed as a sensorimotor disorder, yet it manifests in communicative situations that engage motivational, self-referential, and regulatory processes. In this case-control study, we combined a socio-economic decision task outside the scanner with a socially modulated speech task during fMRI to test how listener presence and self-referential speech shape neural activity in 34 adults who stutter and 32 controls. Both groups valued talking with another person about themselves more than talking with another person about someone else or talking to themselves. On the neural level, listener-directed (vs. private) speech and self-disclosure (vs. guessing the preferences of a famous other) elicited stronger responses in the ventral striatum and medial prefrontal cortex, extending social valuation effects previously reported in fluent speakers to adults who stutter. Within the stuttering group, individual differences revealed a systematic reweighting of socially modulated activation as a function of symptom burden: higher anticipation of stuttering and greater overall impact were associated with stronger engagement of motivational circuitry, greater recruitment of frontal evaluative-control regions, and reduced contextual differentiation within speech-language cortex. Stuttering anticipation and lived experience gradually shift the balance between control, language, and motivational salience-processing systems, contributing to the disorder’s marked heterogeneity and context sensitivity. These findings indicate system-level signatures of the interaction between social context and symptom severity, rather than isolated motor deficits, in developmental stuttering. More generally, they reveal how recurrent experiences shape brain activity through the interplay between language, motivational and control systems that governs human social interactions. *Conclusion:* This multi-system perspective may help account for the context sensitivity and heterogeneity of stuttering and underscores the contribution of social and emotional factors to the neural organization of speech processing. Our findings do not directly speak to treatment efficacy but suggest that clinical assessment may benefit from explicitly considering how social exposure and stuttering anticipation modulate activity in salience, control, and speech-language systems, alongside traditional measures of overt fluency. This may help explain why the same individual can show very different fluency profiles across situations and support approaches that focus on reducing social and emotional load in challenging contexts rather than targeting speech motor behavior in isolation. Individual variability in neural modulation was systematically related to stuttering anticipation and overall impact, indicating that social context shapes brain activity in AWS along clinically relevant dimensions. Our findings support the view that social communicative experience contributes to experience-dependent reorganization of cortico-basal ganglia loops over time. It also implies that the neural effects we observed are unlikely to be driven simply by increased dysfluency under social exposure, but instead reflect changes in the motivational and regulatory state in which speech is produced. Given the established role of the nucleus accumbens in encoding motivational relevance and subjective value in socially meaningful contexts (51–57), this pattern suggest that socially directed speech acquires greater motivational salience as anticipatory load increases, rather than simply becoming more rewarding. In multi-system terms, anticipatory load appears to amplify automatic motivational signals in ventral striatum, consistent with how affective cues bias approach-avoid decisions via striatal circuitry (33). Structural work has identified altered morphology of the nucleus accumbens in persistent developmental stuttering, with reduced volume in children but increased right-lateralized volume in adults (22,32) , suggesting a non-linear developmental trajectory of ventral striatal circuitry. Higher stuttering anticipation was also associated with reduced listener-related modulation in cortical speech-language regions, including inferior frontal and posterior temporal areas that support speech planning and perception, as well as the lateral orbitofrontal cortex (BA 47), implicated in controlled semantic processing and in integration of value and context into language and decision-making. # Content: Speaking is a motor act embedded in socially motivated behavior that engages valuation and control systems. A reward-related circuitry, including the nucleus accumbens and ventral striatum, supports the valuation of actions (24) and is reliably engaged by social rewards (25,26). Beyond responses to primary reward, ventral striatum is sensitive to being observed, receiving social feedback (27,28), and communicating self-relevant information (29,30). Consistent with a role in assigning motivational salience to social acts, both adults and children who stutter show altered grey-matter volume of the ventral striatum/nucleus accumbens, when compared to fluent speakers and typically developing peers (22,31,32). Interestingly, adults show increased right ventral striatum/nucleus accumbens morphology compared with fluent speakers, while children show reduced nucleus accumbens volume compared to their peers. This pattern points to an atypical developmental trajectory of striatal circuitry in stuttering. Rather than a uniform increase or decrease, it suggests that the motivational salience system for social communication is reorganized over time when language development is accompanied by a communication disorder such as stuttering. In this context, speech directed toward a listener is likely to recruit motivational circuits differently in people who stutter than in fluent speakers, with downstream effects on speech-motor and control systems in a listener-dependent manner. Our research analyses suggested that AWS assigned higher value than AWNS to sharing information about themselves and to speaking about others in private. The self > other contrast revealed group differences in regions implicated in self-referential and evaluative processing. Relative to AWS, AWNS showed greater engagement of the subcallosal cortex, often linked to affective valuation and self-related appraisal (74,75), whereas AWS exhibited stronger activity in the medial inferior parietal cortex, including angular gyrus, associated with perspective-taking and integration of self-related information within a broader social context (76,77). Although these differences require replication, the pattern suggests that self-related speech may rely on partially distinct self-referential circuits in AWS and AWNS, potentially reflecting differences in how self-relevant information is evaluated or contextualized during speaking. On current accounts, ventromedial prefrontal cortex/subcallosal regions encode cognitive ‘maps’ for self-relevant value, whereas parietal regions support perspective-taking and contextual integration (33). Within this view, self-related speech in AWS may depend relatively more on distributed perspective-taking networks and relatively less on ventromedial prefrontal-centered self-valuation schemas, possibly reflecting experience-dependent adaptations to chronic social evaluative contexts. Together with the OASES correlations, these group differences indicate that both the evaluation of self-related speech and its modulation by listener presence vary systematically with the perceived impact of stuttering. This dovetails with evidence that, after fluency-shaping therapy, greater relief from the social-emotional burden of stuttering is linked to lower fractional anisotropy in the right frontal aslant tract, suggesting that both frontal gray- and white-matter systems contribute to how stuttering burden is experienced and reduced. Consistent reports of altered frontotemporal and cerebellar-orbitofrontal connectivity that scale with stuttering severity (49) further suggest that individuals with higher burden recruit evaluative and regulatory hubs within a broader fronto-cerebellar and temporo-parietal network. In our task, this is reflected in stronger context-dependent engagement of medial prefrontal and ventrolateral frontal regions when speaking under observation. Prior work using naturalistic interpersonal communication paradigms has shown that disfluent speech in adults who stutter is accompanied by right amygdala activity and reduced prefrontal engagement (18), highlighting a prominent role for limbic circuits during socially demanding speech. Our results complement these findings by showing that, even in a relatively low-stakes setting, higher stuttering-related burden shifts listener-related modulation toward medial prefrontal und lateral orbitofrontal valuation circuits. Together, these studies indicate that limbic, cognitive and sensorimotor systems jointly shape how social context is mapped onto brain networks in developmental stuttering. Overall, our findings argue against a uniform neural deficit in developmental stuttering and instead support a shift in the balance between salience, control, and language processing systems that tracks anticipation and burden. This system-level perspective helps explain why stuttering is highly context-sensitive and heterogeneous and suggests that targeting motivational and evaluative processes, alongside speech motor mechanisms, may be important for understanding and treating the disorder. Greater engagement for directed relative to private speech suggests that speaking under observation places additional demands on evaluative and monitoring processes. Within AWS, higher OASES scores were associated with stronger listener-related modulation in medial prefrontal and ventrolateral frontal regions, indicating that individuals who experience greater stuttering-related burden recruit this evaluative-regulatory circuitry more strongly under listener presence. OASES and PAiS were not correlated (Spearman r(31) = 0.27, P = 0.13, two-tailed), supporting the view that these neural associations track distinct aspects of the stuttering experience and phenotype. Within a multi-system framework (33), dorsal anterior cingulate cortex and medial prefrontal cortex may act as a hub, where learned negative consequences of speaking and current contextual cues jointly shape the value of speaking under observation.