Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Linguistic repertoire size and language experience increase auditory verbal imagery vividness: evidence from the Multilingual Inner Speech Survey (MISS)
Poster Session B, Wednesday, September 30, 4:30 - 6:30 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Sophie Wieser1, Alan Chauvin1, Hélène Lœvenbruck1; 1Univ. Grenoble Alpes, Univ. Savoie Mont Blanc, CNRS, LPNC, 38000, Grenoble, France
Cognitive neuroscience has deepened our understanding of inner speech (or endophasia), the internalised production of language without articulation, gesture, or sound, by uncovering its neural and functional link to overt speech. Yet, most neurocognitive models of inner speech overlook its multilingual nature, despite sociolinguistic evidence that its form and function are shaped by linguistic diversity, proficiency, age of acquisition, and social context (Pavlenko, 2014). To address this gap, we introduce the Multilingual Inner Speech Survey (MISS; Wieser et al., 2025), a comprehensive questionnaire designed to characterize language-specific inner speech experience across multilingual repertoires. The MISS builds upon preliminary findings from the Modalities of Thought Questionnaire (MOT; Huson et al., 2021), which focuses on auditory verbal imagery (AVI), the subjective experience of an "inner voice" that often accompanies inner speech. An initial analysis of a large MOT dataset (N = 2025), using participants’ self-reported number of languages spoken as a proxy for multilingual status, revealed a significant positive association between the number of languages and AVI vividness (χ²(3) = 31.37, p < .001, ε² = .014), suggesting that multilingual individuals tend to report more vivid inner voice experiences. The MISS extends these findings by measuring self-reported phenomenological characteristics of endophasia, focusing on three key dimensions: condensation (ranging from fully expanded to condensed forms), dialogality (monologue vs. dialogue), and intentionality (deliberate vs. spontaneous) (Grandchamp et al., 2019). To capture the multidimensional nature of multilingualism, the MISS incorporates detailed linguistic profiles for up to eight languages, including age of acquisition, proficiency, and frequency of use. Critically, it assesses the frequency and vividness of AVI across all languages in a participant’s repertoire. Pilot data from 56 participants (monolingual to heptalingual) enabled an exploration of vividness differences across the languages each participant uses. Mixed-effects modeling on a restricted MISS dataset (N = 42; repertoire size ≥ 3), focusing on participants’ first three languages, revealed that vividness for each language in their repertoire was significantly predicted by an experience factor, a composite of proficiency and frequency of use (b = 0.49, p < .001), as well as by repertoire size (b = 0.31, p = .031). Results demonstrated systematic differences: first languages (L1s), acquired between ages 0–3, exhibited significantly higher vividness than later-acquired languages (L2 vs. L1: b = −0.32, p = .047; L3 vs. L1: b = −0.47, p = .021). Together, these findings indicate that earlier acquisition, greater proficiency, and higher frequency of use are associated with increased vividness ratings, while larger linguistic repertoires amplify overall AVI vividness. From the perspective of predictive control models of speech production, these findings suggest that the habitual co-activation of multiple languages and phonological systems may enhance the functional necessity and awareness of sensory predictions during inner speech production, potentially amplifying AVI vividness. We further hypothesize that phonological diversity within a repertoire, rather than repertoire size alone, drives this effect. Ultimately, these results warrant an extension of current neurocognitive frameworks by integrating linguistic repertoire diversity as a structural variable in endophasia research.
Topic Areas: Language Production, Multilingualism