Master's Student Presentations
The processing of morphologically complex words is a matter of continued interest and debate in the field of linguistics. Recent research has provided increasing evidence that words are morphologically decomposed during visual word recognition; however, precisely when morphemes are activated during the time course of word recognition, and which word types undergo morphological decomposition remain a matter of debate. The extent to which the process of morphological decomposition interacts with and is modulated by semantic relationships, orthography, lexical status, first or second language, and experimental conditions (such as modality of stimulus presentation) is relevant to current research. The current study seeks to investigate the effects of semantic transparency, orthographic overlap, and morphological status in the mental lexicon (i.e. whether a word is argued to have a morphologically complex representation in the mental lexicon, such as “walking”, or a monomorphemic one, such as “corner”) on morphological decomposition in compound words. This is performed by comparing behavioral responses (within the context of a lexical decision task using the masked priming paradigm) for semantically transparent compound words (e.g. “teacup”), semantically opaque compound words (e.g. “honeymoon”), monomorphemic real words that contain an orthographically overlapping pseudomorpheme (e.g. “flamenco”, which begins with “flame”), and real word “false compounds”: monomorphemic pseudocompounds (e.g. “plankton”) that can be exhaustively parsed into two constituent pseudomorphemes on the basis of orthography (in this case, “plank” and “ton”). This study replicates and extends the study by Fiorentino & Fund-Reznicek (2009), which found significant and equivalent priming effects for both transparent and opaque compounds in comparison to unrelated primes, but no effects for the orthographically overlapping primes; by additionally testing responses to false compounds, this study is able to examine the extent to which semantic transparency in combination with morphological status in the lexicon interact with morphological decomposition in the absence of affixation. We find that the priming effect of such false compounds is statistically equivalent to that of both transparent and opaque compounds, but dissociates from that in the orthographically overlapping pseudomorpheme condition. The results of this study replicate the findings of Fiorentino & Fund-Reznicek (2009) for transparent and opaque compounds, and extend the findings of Rastle et al. (2004), which found significant and equivalent priming effects for both semantically transparent suffixed words (e.g. “cleaner”) and semantically opaque monomorphemic pseudosuffixed words (e.g. “corner”) in comparison to unrelated primes, but no effects for orthographically overlapping primes (e.g. “brothel”, which begins with “broth”), to false compounds. These findings indicate that morphological segmentation applies to the appearance of morphological structure on an orthographic basis, and does so even when no apparent affixation is present, supporting an account of morphological processing that involves early and automatic across-the-board morpho-orthographic decomposition.