Abstract: Investigated J. Grinder and R. Bandler's (1976) assertion that responding with perceptual predicates similar to those used by a speaker results in increased perceived empathy. In an analog counseling interview with 63 female undergraduates, counselors tracked the use of visual, auditory, and kinesthetic perceptual predicates and then responded with either similar or dissimilar predicates of their own. Ss completed a revised version of the Empathy scale of the Barrett-Lennard Relationship Inventory and an interviewer experience scale constructed by the author. A significant difference was found between treatments in the expected direction, with Ss in the matched predicates condition rating their counselors higher on perceived empathy. The procedure of continually tracking and matching perceptual predicates is contrasted with the method of identifying and matching a primary representational system used by other researchers, and questions are raised about the assumptions regarding representational systems in neurolinguistic programming. It is concluded that the tracking procedure was preferable for empirical and theoretical reasons. (31 ref) (PsycINFO Database Record (c) 2005 APA, all rights reserved).