Search Abstracts | Symposia | Slide Sessions | Poster Sessions
A preregistered MEG study on semantic ambiguity resolution in multilinguals: effects of context.
Poster Session C, Thursday, October 1, 10:45 am - 12:45 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Abigail E. Licata1, Alexis Hervais-Adelman2, Valentina Borghesani1; 1Faculty of Psychology and Educational Sciences, University of Geneva, Geneva, Switzerland, 2Faculty of Medicine, University of Geneva, Geneva, Switzerland
This ongoing project examines how semantic ambiguity affects language processing in native (L1) and non-native speakers, particularly when ambiguity patterns are either preserved or mismatched across languages (https://osf.io/7ychv/). We distinguish parallel ambiguity (PA), where two languages share the same ambiguity structure for a given word (e.g., French "ciel" and German "Himmel", which both describe concepts SKY and HEAVEN) from translation ambiguity (TA), where an ambiguous word in one language corresponds to multiple translation equivalents in another (e.g., French "échelle", which maps to German words "Skala"-SCALE and "Leiter"-LADDER). Building on prior work on semantic control (Jackson, 2021; Jefferies, 2013) and TA (Jouravlev & Jared, 2020), we will test how these types of ambiguous words are processed by L1-French speakers and by L1-German speakers who also speak French (L1-DE). We will examine, in both groups, the effects of ambiguous word processing relative to unambiguous words, and test whether TA imposes greater processing demands than PA in L1-DE participants. Moreover, we will examine whether this potential difference in TA and PA depends on context, semantic relatedness of the concepts underlying an ambiguous word, and individual language experience. Participants will complete two French tasks in MEG, during which ambiguous words will be presented without (single-word reading) or with context (sentence reading) (MacGregor et al., 2020). To ensure attention during single-word reading, participants will be instructed to make a lexical decision, with pseudorandomly presented pseudowords to ensure attention (6.25% of trials); during sentence reading, participants will be instructed to decide whether isolated words, presented pseudorandomly in a subset of sentences (7% of trials), are semantically related or unrelated to the sentence they just read. Source-localized (Gross et al., 2001) time-frequency data will be analyzed to compare neural responses to ambiguous and unambiguous words and sentences. For the L1-DE group, language experience measures including proficiency, age of acquisition, exposure, and dominance will be examined as potential moderators of ambiguity-related effects. If TA elicits stronger neural signatures of control or integration than PA in L1-DE, this would suggest that cross-language ambiguity structure mismatch increases the processing demands associated with semantic ambiguity. If PA and TA produce similar patterns, this would instead suggest that ambiguity processing in the second language is driven more by within-language semantic structure than by cross-language mapping. Finally, individual differences in language experience may help clarify whether ambiguity processing in a second language, particularly any observed effect of TA, is modulated by proficiency, exposure, or dominance. With a goal of 60 participants total (30 per group), data collection is ongoing and projected to reach 50% completion by September 2026.
Topic Areas: Multilingualism, Meaning: Lexical Semantics