Post-Sentence Responses to Syntax-Plausibility Conflict in Dative Sentences Produced With versus Without Stuttering

During sentence comprehension, listeners may base an eventual interpretation on syntactic structure, event plausibility, or both. Whether the relationship between these information sources differs for sentences produced with versus without stuttering has not been established.

To test whether sentence plausibility and dative structure were related to the probability of a post-sentence response consistent with literal syntax differently for listeners who heard sentences produced with versus without stuttering.

Two hundred twenty-four adults each heard one condition of 24 experimental items. Half heard sentences containing stuttering, and half heard sentences without stuttering. The items crossed plausible versus implausible meanings with double-object versus prepositional-dative structure. A yes/no question after each sentence elicited a response coded as literal or nonliteral. Individual binary responses were analysed using binomial logistic mixed-effects models that accounted for participant and item variability.

The primary analysis identified a strong Plausibility effect (b = 4.134, SE = 0.282, p < 0.001), a Group × Structure interaction (b = -0.712, SE = 0.292, p = 0.015), and a Plausibility × Structure interaction (b = -1.576, SE = 0.509, p = 0.002). The Group × Plausibility × Structure interaction was not statistically significant (b = 0.328, SE = 0.752, p = 0.663), and none of the four planned listener-group comparisons within individual Plausibility × Structure conditions was significant after correction for multiple comparisons. An exploratory analysis restricted to implausible sentences showed the same Group × Structure pattern but no significant listener-group difference within either dative structure.

Post-sentence responses reflected event plausibility and dative structure. The relative pattern of responses to double-object and prepositional-dative sentences differed between recording conditions, without a detected overall Group effect. Stuttering may contribute to this pattern, but the present design cannot determine whether its effects are distinctive or shared with other speech disruptions, or separate them from accompanying recording differences. Replication across speakers, stronger speech controls, and process-sensitive measures are needed to resolve these possibilities.

What is already known on this subject Speech containing stuttering can affect listener comprehension and measures collected during sentence processing. Post-sentence response paradigms also show that syntactic structure and event plausibility can jointly influence the interpretation selected. However, this relationship has not been compared for sentences produced with versus without stuttering. What this paper adds to existing knowledge Listeners' post-sentence responses reflected event plausibility and dative structure. The relative pattern of responses to double-object and prepositional-dative sentences differed between listeners who heard recordings with versus without stuttering, rather than showing an overall group difference in literal responding. The findings leave open whether stuttering has a distinctive effect on interpretation or whether the observed pattern reflects processes shared with other speech disruptions or accompanying recording characteristics. What are the potential or actual clinical implications of this work? The findings underscore the importance of considering listener-side outcomes when studying communication involving stuttering. Their clinical relevance will depend on replication across speakers and further research identifying when and how these response differences arise.