Adapting linguistic representations during social interactions: a dynamic view of spoken human communication – REaDY-SPOK

The original feature of this work will be to develop a new methodological approach in which cutting-edge behavioral and EEG measures will be used within the context of spontaneous dialogue settings. Our methodological approach will begin by a phase during which two partners will either interact together freely. We will use microphones to measure the partners’ acoustic cues during the interaction and we will analyze the degree of lexical convergence (that is, the extent to which speakers converge on the words they use) and the expression of verbal feedback during the interaction. In a new manner, the dialogic settings will be followed immediately or one day later after by experimental designs in word recognition and in spoken language comprehension to probe the persistence of the adaptation of linguistic representations at the semantic and phonological levels and of the adaptation of prediction.

We developed a Tangrams pictures database with norms for name agreement, image agreement, familiarity, visual complexity, image variability, age of acquisition and concreteness. This database was constructed so that it could be adapted to studies in social intreraction while looking at the properties usually described in databases with concrete pictures.

Studies on the adaptation of semantic and phonological representations after a social interaction between two interlocutors are underway.

Fasquel, A., Brunellière, A., Knutsen, D. (soumis) Naming 332 Tangram pictures in French: norms for name agreement, image agreement, familiarity, visual complexity, image variability, age of acquisition and concreteness.

Stinkeste, C., Vincent, M., Delrue, L., & Brunellière, A. (2021). Alpha and high-gamma oscillations support the interface between language prediction and speakers’ communicative intentions in spoken-language comprehension. Society for the neurobiology of language, 5-8 october 2021, oral communication for virtual conference.

The purpose of the project is to examine how spontaneous interactive cues conveyed during social interactions contribute to learning and knowledge co-construction. The project seeks to address the question of how communication shapes people’s mental representations, and whether people’s states of mind become increasingly similar as they interact due to linguistic representation adaptation. Even though convergence and feedback can be described as reflecting the speakers’ attempts to improve mutual comprehension, it is still unknown whether speakers adapt their linguistic representations after an interaction, that is, whether spoken human communication has a direct influence on the content and organization of such representations. This proposal, which combines psychological and neuroscience approaches, addresses the question of whether feedback, a dialogic marker used to ensure mutual comprehension, contributes to the adaptation of linguistic representations after an interaction between two speakers. It also seeks to determine whether feedback facilitates prediction in comprehension. Two types of linguistic representations will be examined. Using behavioral and electrophysiological measures, we will focus on the conceptual level of word meaning (i.e., the semantic level) and on the level of word form (i.e., phonological level). Prediction in comprehension (i.e., predicting of what the interlocutor wants to say) is seen as a key mechanism of the adaptation of mental representations. By using cutting-edge behavioral and electrophysiological measures in the context of spontaneous dialogue settings, this project will examine both the reorganization of semantic and phonological representations after the interaction and the mechanisms underlying this phenomenon and persisting after the interaction. The consortium of the project will be coordinated by a young researcher, who works on spoken-word recognition, semantic memory and the role of prediction in spoken-language comprehension by using behavioural and electrophysiological measures. Thanks to complementary skills of the consortium at the methodological and theoretical levels, the project will be based on solid scientific knowledge about cognitive models in spoken word recognition and semantic memory. It will also integrate notions such as concept sharedness in dialogue and prediction in comprehension. This fundamental research proposes a new theoretical view of spoken language communication accounting for the adaptation of linguistic representations in the short and the long term after a dialogue and it is centered on an innovative dynamic view of spoken human communication in which linguistic representations are conceptualised as flexible. It also has direct implications for education, as learning settings are typical situations in which one person attempts to modify another person’s mental representations through communication. Thanks to the novelty and the originality of project, the findings will be presented at international conferences and journals. This project will contribute to create a European network on spoken human communication and linguistic representations in the fields of psychology, linguistics, and cognitive neuroscience.