Can priming on perceptual implicit memory tests be achieved through semantic activation?
Roediger and McDermott (1995) presented lists of semantic associates (e.g., thread, pin, sewing) to induce false recall of related, nonpresented words (e.g., needle). This study technique was used in an attempt to determine whether semantic activation of target words would be sufficient to produce perceptual priming. Although most current theories of implicit memory would predict otherwise, reliable priming of the nonpresented items occurred both on word stem and word fragment completion. Priming did not differ reliably (but was numerically less than) priming of presented items. Implications for theories of implicit memory will be discussed.
