Presented at the First International Quantitative Morphology Meeting (IQMM1); University of Belgrade, July 11–15 2015

Previous work on morphological processing of complex Greek words (nouns and verbs) examined whether lexical processing times of Greek noun and verb targets are facilitated by prior presentation of morphologically related vs unrelated words, in a lexical decision task with masked priming. Results showed that morphological relations affect processing of morphologically complex words (verbs and nouns) or, in other words, that Greek readers are sensitive to the morphological structure of these morphologically complex Greek words. This is consistent with studies using the same paradigm in other languages (e.g., Rastle, Davis, Tyler, & Marslen-Wilson, 2000) and provides evidence for morpheme-based representations in the Greek lexicon. Importantly, facilitation between morphologically related word pairs was present after only 50 ms of prime presentation, which was not the case with orthographically related word pairs. This indicates that morpheme-based representations are accessed early in visual word recognition, and that semantic information is not completely ignored at early stages (cf. Feldman, O'Connor, Moscoso del Prado Martín, 2009).

In the present study, we further investigate morpheme-based representations in Greek and their early processing, using transposed-letter effects. Specifically, we examined whether priming is observed with letter transpositions across vs. within morphemes. The same target words as in the previous experiment were used, again in a masked-priming lexical decision task. Primes were based on the morphologically related primes of the previous experiment, manipulated by transposing (vs. replacing) two adjacent letters that either straddle the boundary between stem and suffix or are contained within one morpheme (either at the edge or internally). If Greek morphologically complex words are represented in a decomposed manner, and if decomposition occurs at early stages of visual word recognition, we expected an interaction with transposed-letter effects, such that priming facilitation would occur only when transpositions occurred within morpheme boundaries. Results showed that transposed-letter primes generated significant priming, relative to replaced control primes, when transpositions occurred within the stem (either at the edge or internally) but not when transpositions occurred across morphemes, in line with our predictions. However, there was no significant difference between response latencies to targets with a different locus of transposition (across vs. within morphemes). This might suggest a role for supralexical morphological representations in modulating masked morphological priming.

References:
Feldman, L. B., O'Connor, P. A., & del Prado Martín, F. M. (2009). Early morphological processing is morphosemantic and not simply morpho-orthographic: A violation of form-then-meaning accounts of word recognition. Psychonomic Bulletin & Review, 16(4), 684–691.
Rastle, K., Davis, M. H., Tyler, L., & Marslen-Wilson, W. (2000). Morphological and semantic effects in visual word recognition: A time-course study. Language and Cognitive Processes, 15(4/5), 507–537.