Further evidence for sublexical components in implicit memory for novel words.

This study provides further evidence for the notion, proposed by Dorfman (1994), that activation of sublexical components plays an important role in implicit memory for novel words. Priming in an identification task was observed consistently when nonwords were formed out of morphemes; weaker, more variable effects were observed when nonwords were composed of syllables or pseudosyllables. Priming was insensitive to changes in the surface features (type case) of the stimuli and was disrupted by rearrangement of the stimulus components. Results are interpreted in terms of the activation and integration of preexisting sublexical representations and as evidence against the role of newly established episodic or perceptual representations. It is concluded that findings of priming for novel information should not be taken as evidence against an activation view.