Proceedings of the Annual Meeting of the Cognitive Science Society

The validity of verbal working memory depends on language experience-independent capacities. We tested how grammat-ical knowledge impacts memory in the absence of overt language production and while controlling for semantic meaningof word pair stimuli. Native English speakers (n=129) completed: (1) ratings of unattested noun-noun compounds (e.g.ice-wallet) on meaningfulness; the (2) Author Recognition Test, measuring language experience; and (3) an old/newrecognition task, where previously presented noun-noun compounds appeared in either old (ice-wallet) or new (wallet-ice)orders. Order of nouns in compounds either resembled order consistently found in English (i.e. typical noun modifier +typical head noun) or was reversed. If grammatical knowledge affects verbal working memory, consistency with natu-ral language should predict old ratings, controlling for meaningfulness ratings and old status. As predicted, participantswere more likely to rate consistent compounds as old compared to reversed. All analyses pre-registered on OSF prior toexperimenter access to data.