Word Frequency and Morpheme Frequency Effect in Naming Mandarin Chinese Compounds: More than a Replication

An unresolved debate regarding whether compound words are stored in our mental lexicon in a decomposed or full-listing way prompted Janssen et al. to investigate the representation of compounds using word and morpheme frequency manipulations. They employed three conditions, i.e., H(h) where both word and constituent frequencies are high, L(h) where word frequency is low but constituent frequency is high, and L(l) where both word and constituent frequencies are low. While their results initially aligned with a compositional representation, issues arose with the stimuli they used, particularly the frequencies of the high and low conditions, showing no significant differences.
In our study, we tackled this issue using a new set of stimuli and incorporating EEG data for a more detailed investigation. Despite ERP analyses revealing no word frequency or morpheme frequency effects across the three conditions, behavioral outcomes indicated similarity between H(h) and L(h) conditions in naming latencies, suggesting compounds may not be sensitive to word frequency. Furthermore, the L(h) condition exhibited significantly faster naming than the L(l) condition, highlighting a morpheme frequency effect in naming Mandarin compounds. These findings challenge the compositional model while potentially bolstering the decompositional model.