Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Functional Activation and Connectivity During Phonological Conflict Processing: An fMRI Study Based on the Consonant Same–Different Judgment Task
Poster Session F, Friday, October 2, 2:45 - 4:45 pm, Wangari Maathai
Zirui ZOU1, Xavreila Sheung-Wa NG1, Jiaxin CHEN1, Nga-Yan HUI1, Matthew King-Hang MA1, Min Ney WONG1, Manson Cheuk-Man FONG1; 1The Hong Kong Polytechnic University
Previous research on phonological processing has largely overlooked the role of cognitive-control network. The conflict monitoring theory, built largely upon domain-general, non-linguistic tasks (e.g., Simon/Flanker tasks), predicts that canonical conflict-related regions (medial and lateral prefrontal cortex, PFC) are similarly implicated in resolving phonological conflicts, but empirical evidence has remained scarce. Moreover, while functional connectivity studies revealed that general speech/phonological processing relies on extensive fronto-temporal connectivity (particularly, posterior-temporal to inferior-frontal) within the language network, it remains unclear whether similar fronto-temporal coupling supports dual linguistic and cognitive-control demands in phonological conflict processing. Twenty-nine young Hong Kong Cantonese adults completed an fMRI consonant same–different judgment task, judging whether the initial consonants of sequentially-presented Cantonese syllable-pairs were matched or mismatched. A 2×2 design (CONFLICT: high/low; RESPONSE: same/different) was employed, with different levels of conflict being induced by manipulating the task-irrelevant lexical tone. Imaging data were preprocessed using a surface-based pipeline via CONN and FreeSurfer. Surface-based activation and gPPI functional connectivity analyses were conducted using AFNI/SUMA and CONN. To examine conflict-related brain–behaviour associations, subject-level ROI activation and ROI-to-ROI connectivity conflict effects were tested against individual differences in perceptual sensitivity (d’), using LME models. Ten a priori ROIs were defined using HCP-MMP parcels: for conflict control, bilateral rostral anterior cingulate cortex, rACC (a24, s32), bilateral dorsolateral prefrontal cortex, DLPFC (46), and right inferior frontal cortex, rIFC (IFJp); for phonemic processing, the left posterior middle temporal gyrus, pMTG (PHT), left supramarginal gyrus (PFcm), and left anterior superior temporal sulcus, laSTS (STSva). We hypothesized that for activation, ACC, DLPFC, rIFC, and temporoparietal regions would be sensitive to degree of phonological conflict; for connectivity, phonological conflict would recruit fronto-temporal connectivity between frontal-control and temporal-phonological regions (pMTG, STS); and that participants’ behavioral sensitivity would modulate conflict effects. Behavioral results showed a significant conflict effect on reaction time, p = .015. For activation, whole-brain analysis revealed a main effect of CONFLICT (high>low) in left rACC, dorsomedial PFC, and temporal pole, and right pMTG, vertex-level uncorrected p < .001, cluster > 10 vertices. ROI analysis showed CONFLICT effects in bilateral a24, left s32, STSva, PFcm, and PHT, and right IFJp, FDR-corrected p < .05. Consistent with the conflict monitoring theory, these findings indicate conflict-control regions (ACC, rIFC) are sensitive to phonological conflict. While DLPFC showed no group-level effect, individual-difference analyses showed a bilateral conflict effect negatively associated with d’: left, p = .045; right, p = .023, suggesting that DLPFC recruitment reflects individual sensitivity variations. For connectivity, ROI-to-ROI analysis revealed a conflict-modulated functional connectivity network (PHT-to-a24, a24-to-46), TFCE-FWE-corrected p < .05. Furthermore, gPPI connectivity conflict effects between the laSTS and rIFC were positively associated with d’, ps = .006–.009. These findings suggest that phonological conflict modulates the coupling between left pMTG/STS and medial/inferior frontal-control regions. Taken together, our results showed that phonological conflict engages an auditory selective-attention network comprising temporoparietal language regions and fronto-cingulate control regions, confirming that the conflict monitoring theory applies similarly to phonological/linguistic modalities. Furthermore, conflict-modulated, fronto-temporal connectivity suggested interactions between language and control networks during phonological conflict processing.
Topic Areas: Control, Selection, and Executive Processes, Phonology