Four experiments contrasted the cue-familiarity hypothesis of feeling-of-knowing judgments (FKJs) and tip-of-the-tongue feelings (TOTs) to the target-retrievability hypothesis. Familiarity of the cues was contrasted to memorability of the targets in a paired-associate design (e.g., A-B A-B, A-B A-B', A-B A-D, A-B C-D), in which the number of repetitions of the cue A terms was dissociated from the memorability of the target B terms. Little support was found for the target-retrievability hypothesis, because in none of the 4 experiments were FKJs related to target memorability. In one experiment, an omnibus retrieval hypothesis (which implicates total retrieval rather than just correct retrieval) and the cue-familiarity hypothesis produced isomorphic predictions that were borne out by the FKJ and TOT results. All 4 experiments supported the cue-familiarity hypothesis, because FKJs and TOTs were directly related to the number of presentations (and thereby the familiarity) of the cues.