Novel popout without novelty

Localization of a novel word in an array with several familiar words is typically enhanced relative to localization in an all-novel array (between-array novel popout) and sometimes enhanced relative to familiar words in one-novel arrays (within-array novel popout). Christie and Klein (1996) have questioned the reality of the latter effect, suggesting that it may be an artifact of guessing bias. The present Experiment 1 replicated within-array novel popout with the novel word probed at chance (i.e., on only one quarter of trials). Experiments 2 and 3 demonstrated a similar popout effect for a categorically unrelated word among three categorically related words, despite superior performance on all-related arrays relative to all-unrelated arrays. Repetition of constant sets of words within the experimental context is therefore unnecessary for a popout effect, contrary to assertions by Johnston and Hawley (1994). Interitem associations appear to be sufficient to produce a popout effect; as such, “novel popout” appears to be a misnomer for a phenomenon that does not depend on novelty.