Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Social pragmatic communication difficulties predict altered pitch use and serial dependence in prosodic stress judgments
Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai
Peter Bang1, Lea-Marie Tewald1, Lisa Bauknecht1, Dirk Wildgruber2, Yulia Oganian1; 1Centre for Integrative Neuroscience, University of Tübingen, Tübingen, Germany, 2Department of Psychiatry and Psychotherapy, University of Tübingen, Tübingen, Germany
Social pragmatic communication (SPC) difficulties are a core component of the autism phenotype. SPC abilities are tightly linked to prosody processing as listeners must interpret intonation in speech to infer intent and to anticipate how a conversation will unfold. When prosodic cues are difficult to interpret, intentions become harder to track, making interaction effortful and socially disconnecting. A central aspect of prosodic processing is the ability to identify where stress is placed within a word (e.g., to preSENT vs a PREsent) and within a sentence (“WHO bought cupcakes?” vs. “who bought CUPCAKES?”). A key open question is whether individual differences in SPC are associated with prosodic judgments through altered sensitivity to acoustic cues or altered flexibility in updating perception as evidence changes. Work on serial dependence highlights that consecutive perceptual decisions can be biased by recent history through at least two separable mechanisms: a repulsive influence of prior stimuli reflective of sensory adaptation, and an attractive pull toward prior choices reflective of decision-level integration. This dissociation provides a principled framework for testing whether SPC differences are linked to sensory encoding, serial dependence, or both. Participants (n = 172, 8.1% diagnosed autistic) completed a German word-level and a sentence-level stress judgment task in two separate order-randomized blocks. On each trial, listeners used a slider to judge which syllable or which word they perceived as being more strongly stressed. Acoustic evidence for stress was manipulated along either pitch- or duration continua. SPC difficulties were measured with the Conversation Questionnaire (CQ). We fit linear mixed-effects models predicting trial-wise prosodic judgments from current acoustic evidence (pitch and duration), as well as stimulus history (previous trial’s acoustic evidence and prosodic judgement). CQ was included as fixed effect and in interaction with both current acoustic evidence and history terms, allowing us to test whether SPC-related differences were better explained by reduced sensitivity to pitch and duration cues or by stronger dependence on prior stimuli and responses during perceptual updating. As expected, current acoustic evidence predicted perceptual judgements. Beyond these immediate acoustic effects, responses showed robust stimulus repulsion and choice attraction. Importantly, SPC-related differences were not well characterized as a general sensory reduction. Instead, higher SPC was associated with weaker pitch sensitivity in sentence-level stress judgments, but stronger pitch sensitivity in word-level judgments. This divergence suggests that pitch may support compact syllable-level contrasts when the relevant evidence is local, while becoming harder to integrate when stress must be interpreted across richer sentence-level contexts. SPC also strengthened repulsive carryover from prior pitch and attractive carryover from the preceding response, with choice attraction most pronounced in sentence-level judgments. Together, these findings suggest that SPC-linked prosodic differences arise not from reduced acoustic sensitivity alone, but from altered context-dependent cue weighting combined with stronger carryover from recent perceptual decisions. In conversation, such carryover may reduce flexibility when listeners must continuously update prosodic interpretations across longer, socially meaningful utterances.
Topic Areas: Prosody, Meaning: Discourse and Pragmatics