Fig. 4.

Priming effects for participants who reported being unaware of primes, combining data from Experiment 1 with data from a different group of participants reported in Taylor et al. (in press). Priming (% of trials called “old” for Primed–Unprimed) is shown as a function of Studied (Hits; upper panel) or Unstudied (False Alarms; lower panel) items, for Conceptual vs. Repetition Primes and R (red) vs. K (green) judgments in Experiment 1. The two bars for K judgments are based on raw numbers (dark green) or under independent (IRK) scoring (‘iK’, light green; see main text). Error bars are 95% confidence intervals.