Phonemic, Associative, and Grammatical Context Effects With Identified And Unidentified Primes.

Number 736
Year 1990
Drawer 13
Entry Date 11/15/1999
Authors Lukatela, G., Carello, C., & Turvey, M. T.
Contact
Publication Language and Speech, 33(1), 1-18.
url http://www.haskins.yale.edu/Reprints/HL0736.pdf
Abstract Six experiments assessed priming effects on lexical decision in Serbo-Croatian when the context is identifiable (unmasked conditions) & when it is unidentifiable due to forward masking. In experiment 1 (N = 56 Serbo-Croatian-speaking high school students), word targets consisted of 72 relatively high frequency singular nouns presented visually. No context or target word was encountered more than once by any S. One set of contexts was chosen to be phonemically identical to the targets except for the middle letter; a second set was phonemically dissimilar, sharing no more than two phonemes. Reaction time & errors were measured in the lexical decision task. In experiment 2 (N = 45), the same stimuli & procedures were used except that visual forward masking was introduced. In experiment 3 (N 56), targets included pseudowords derived from associated context words. In experiment 4 (N = 24), the same context-target pairs were used, with the context masked. Experiment 5 (N = 60) examined the influence of agreement or disagreement with respect to case marking. Experiment 6 (N = 32) used a masked context. It was found that word acceptance is slowed by a phonemically similar context that is not masked, but facilitated by a phonemically similar masked context. Word acceptance was facilitated by an associatively related context that is not masked, as well as by a grammatically related context that is not masked. Results are discussed in terms of a model of the language processor that maintains the autonomy of prelexical & postlexical levels, but permits interaction among the prelexical components. Adapted from the source document.
Notes

Search Publications