Neurolinguistic Programming (NLP) is a currently popular therapeutic modality in which individuals organize information through three basic sensory systems, one of which is the Primary Representational System (PRS). This study was designed to investigate gender differences in PRS according to the predicate preference method. It was expected that males would be identified as primarily visual and that females would be identified as primarily kinesthetic. An ad hoc purpose was to investigate scoring procedures used in the predicate preference method. College students (N=76) were asked to describe three settings or events. Data were scored initially in a deliberate manner using literal meanings of predicates. For the ad hoc investigation of NLP scoring procedures, data were scored a second time according to the subjective evaluation of predicates. While results of multivariate analyses of variance with gender as the independent variable and three predicate scores as dependent variables were uninterpretable, PRS frequency tallies showed that when data were scored according to deliberate criteria, virtually all subjects were kinesthetic; when scored according to subjective impression, most subjects were visual. Results suggest that standard guidelines should be developed for scoring predicates and that NLP should be used cautiously until further research has been conducted. (Author/NB)