Search

We apply the concept of meaning predictability developed by Štekauer (2005, 2006, Štekauer, Diaz-Negrillo & Valera 2011) to the cognitive processing of modifier-noun phrases (e.g., wool basket) and determine whether meaning predictability is a viable psycholinguistic construct. We review and describe three sets of experiments that examined the role of meaning predictability. In the first set of experiments, we investigated how each of the two components of meaning predictability (i.e., prevalence and goodness) influenced the reaction times in a series of tasks that involved the interpretation of modifier-noun phrases. The results of these experiments showed that highly predictable meanings elicit faster responses in a sense/nonsense judgment task and in a meaning verification task. Furthermore, the results indicated that goodness and prevalence played different roles in the two tasks, which suggests that these two components of meaning predictability are best evaluated separately rather than as a single predictability rate measure. In the second set of experiments, we investigated the influence of sentential context on response times in a comprehension task and found that the influence of context depended on meaning dominance. Finally, in the third set of experiments we investigated whether the influence of meaning dominance extends to written production. We found that meaning dominance influences inter-key typing latencies but that this influence differs across the various positions in the phrase. [ABSTRACT FROM AUTHOR]