On the Measurement of Priming: What is the Correct Baseline?

We argue, contrary to Ostergaard (this issue), that the baseline for measuring priming effects in "remgnit~odidentification" experiments is correctly estimated by the subjects' performance with non-primed items. There is no reason to assume, as Ostergaard (this issue) does, that priming that occurs in a single presentation of the target item is qualitatively different from priming that occurs, in the same experimental situation, in two or more presentations, or that empirical findings. such as stmhastic independence between recognition and primed identification, are affected hy the number of presentations of the target items at study.