“…This theory explains the finding that category exemplar generation is intact in AD when study-and test-phase operations are identical but not when they are mismatched (Maki & Knopman, 1996). Although this is a viable explanation for this result, the theory cannot explain the numerous examples of normal AD priming following unmatched study-and test-phase operations (e.g., Abbenhuis et al, 1990;Balota & Ferraro, 1996;Christensen et al, 1992;Fennema-Notestine et al, 1994;Perfect et al, 1992;Russo & Spinnler, 1994;Scott et al, 1991). For example, normal priming has been demonstrated after words were read silently in the study phase and stems of words were completed in the test phase (e.g., Deweer et al, 1994) and when words were generated in the study phase and identified in brief presentations in the test phase (Fleischman et al, 1995).…”