Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Do Visible Social Cues Modulate Syntactic Priming? An fMRI Study of Avatar-Based Sentence Production
Poster Session D, Thursday, October 1, 4:30 - 6:30 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Qiang Huang1, Jiaxin Yan1, Yuyong Liu2, Motoaki Sugiura3,4, Hyeonjeong Jeong1; 1Graduate School of International Cultural Studies, Tohoku University, 2Graduate School of Medicine, Tohoku University, 3Institute of Development, Aging and Cancer, Tohoku University, 4International Research Institute of Disaster Science, Tohoku University
Syntactic priming—the tendency to reuse recently processed sentence structures—has typically been explained as residual activation or implicit learning from prior linguistic experience. However, Pickering and Garrod’s (2004) Interactive Alignment Model proposes that syntactic repetition is also part of broader interpersonal alignment, through which interlocutors align linguistic representations to support communication. Consistent with this view, Schoot et al. (2019) showed stronger syntactic alignment in the presence of an interlocutor than in non-interactive settings, and Heyselaar et al. (2017) reported stronger priming for human-like than computer-like avatars. At the neural level, fMRI adaptation studies have shown repetition suppression for repeated syntax in language-related regions, including left inferior frontal gyrus (IFG) and left middle temporal gyrus (MTG) (Segaert et al., 2013). Yet previous neuroimaging work has not shown stronger syntactic repetition effects in communicative contexts (Schoot et al., 2014), leaving open whether visible social cues such as faces or avatars modulate the neural mechanisms of syntactic priming. This fMRI study tests whether visible social cues modulate syntactic repetition during Japanese sentence production. Japanese native speakers will perform an overt production task in the scanner. On each trial, participants first hear and view a prime sentence produced by either a human-like avatar or a mosaic version of the same speaker. They then describe a transitive-event picture using an active or passive Japanese sentence, guided by a stoplight cue that determines which character is named first. The design crosses Prime Source (avatar, mosaic) with Syntactic Repetition (target syntax repeated vs. novel relative to the prime). Because the target structure is experimentally elicited, behavioral priming will be measured as speech-onset facilitation, and neural priming as BOLD repetition suppression for repeated relative to novel syntax. The central prediction is an Avatar × Syntactic Repetition interaction. Repeated syntax should produce greater speech-onset facilitation and stronger repetition suppression after avatar primes than after mosaic primes, particularly in left IFG and left MTG. A simple avatar > mosaic contrast will not be interpreted as evidence for social modulation, because such effects could reflect face perception, audiovisual speech, or attention. Instead, the key test is whether avatar cues specifically modulate the syntactic repetition effect. To examine a social-alignment mechanism, participants will rate the avatar on humanness, likability, and familiarity. Preliminary pilot data motivate this individual-differences analysis. We predict that higher social-perception ratings will be associated with larger avatar-specific priming effects in both behavior and language-region repetition suppression. Exploratory connectivity analyses will test whether avatar primes increase task-dependent coupling between social-cognitive or social-perceptual regions, including medial prefrontal cortex, temporoparietal junction, superior temporal sulcus, and precuneus, and language-production regions during syntactic repetition. We will further test whether social–language coupling accounts for the association between social-perception ratings and priming magnitude. Evidence for such an interaction would suggest that visible social cues modulate syntactic alignment through social–language network coupling. Conversely, avatar-related activation without changes in syntactic repetition or connectivity would support a perceptual or attentional account rather than a social-cognitive modulation account.
Topic Areas: Language Production, Syntax and Combinatorial Semantics