Obsessive-compulsive disorder and abstract sequence task contributions shift prefrontal cortical connectivity

Abstract

Obsessive-compulsive disorder (OCD) is characterized, in part, by repetitive, sequential behaviors, such as cleaning rituals, yet underlying neural circuitry related to abstract sequencing in OCD remains poorly understood. Prior work has implicated a set of cortical regions activated during abstract sequences, which are defined by rules rather than specific stimulus features (Desrochers et al., 2022). These regions include the rostrolateral prefrontal cortex (RLPFC) that is necessary for performance on abstract sequence tasks (Desrochers et al., 2015), as well as the anterior cingulate cortex/dorsolateral prefrontal cortex (ACC/DLPFC), supplementary motor area (SMA), middle temporal gyrus (MTG), and temporo-occipital junction (TOJ) that are differentially activated in OCD compared to healthy participants during abstract sequencing (Doyle et al., 2026). It remains unclear, however, whether these regions form a coordinated circuit, and how their interactions may differ in OCD. In the present study, we examined task based functional and effective connectivity among these regions using a previously published dataset. We tested hypotheses that connectivity within this circuit would be altered in OCD relative to healthy controls (HCs), and that prefrontal regions (ACC/DLPFC and RLPFC) would direct information to downstream regions (SMA, MTG, and TOJ) during a sequential task. We found that connectivity within this circuit differed significantly between groups. HCs exhibited less negative connectivity from the ACC/DLPFC to the TOJ and stronger positive coupling between the MTG and TOJ, as well as stronger coordination between the RLPFC and DLPFC, suggesting coordinated prefrontal control. In contrast, individuals with OCD showed increased connectivity between the RLPFC and MTG, indicating a more direct influence of RLPFC on posterior regions. Effective connectivity analyses further indicated that, across participants, the ACC/DLPFC and MTG function as central hubs of information flow, with task-related inputs entering the circuit via the TOJ, propagating through the MTG to the RLPFC, and subsequently modulating ACC/DLPFC and downstream regions. These findings suggest a shared underlying circuit architecture in OCD and healthy participants despite differences in functional coupling, particularly involving prefrontal cortical regions. Overall, differences arise at the level of functional coordination within a preserved circuit for abstract sequential processing in OCD, adding to current neurobiological models of OCD and suggesting a novel circuit that supports abstract sequencing.