The central aim of this paper is to investigate Štekauer's (2005 , 2006 ) notion of meaning predictability within a psycholinguistic framework. In particular, we examined whether novel compounds with low meaning predictability are more difficult to interpret than are compounds with higher meaning predictability. A second aim is to evaluate the influence of the components of meaning predictability (i.e., the goodness of a particular reading, as well as the prevalence of that reading) on comprehension. We report the results of two experiments conducted with novel compounds (e.g., wool basket and adolescent doctor). In Experiment 1, participants performed a sense/nonsense judgment task. In Experiment 2, participants performed a verification task in which they indicated whether a particular reading was appropriate. The results confirm that meaning predictability influences ease of interpretation, but also indicate that the role of the components of meaning predictability differs between the two tasks.
Automaticity in processing spatial-numerical associations: Evidence from a perceptual orientation judgment task of Arabic digits in frames
