An ongoing controversy in psycholinguistics concerns the representational nature of complex but compositional combinations in the lexicon. While some (e.g. Janssen & Barber, 2012) have proposed that all units of linguistic experience, including literal combinations like "red car", are stored holistically in the mental lexicon as unanalyzed wholes, others have pointed out that frequencies of many sizes influence language processing (e.g. Bannard & Matthews, 2008; Arnon & Cohen Priva, 2013). In this talk I will present evidence that joint contributions from individual component words within compound words and adjective-noun phrases still exert a strong influence on language processing. First, I will present a corpus phonetics analysis of American English compound words like "birthday" from a novel statistical perspective. Then, I will present data from two published experimental studies evaluating the role of word and phrase frequencies in the processing of literal combinations of words like "alcoholic beverages" and "psychic nephew" in language comprehension-based recognition memory tasks (Jacobs, Dell, Benjamin, & Bannard, 2016, Journal of Memory and Language) and language production-based free recall tasks (Jacobs, Dell, & Bannard, 2017, Journal of Memory and Language). Altogether, the evidence points to a need for linguistic and cognitive theories to account for the critical role of component words in the processing and production of morphologically and syntactically complex linguistic structures, while also permitting the learning and representation of higher-order linguistic entities in memory.
January 16, 2020 - 10:59am