Abstract: The present study addresses several issues raised by Neurolinguistic Programming's (NLP) conceptualizations of the function of individual differences in preferred sensory predicate usage. One hundred twenty-four subjects completed the necessary procedures for inclusion in the present study. A language sample was obtained for each subject. This language sample was transcribed and coded with regard to the number and type (auditory, visual, kinesthetic, olfactory, gustatory, and neither) of predicates employed. The first NLP proposition studied is that a subject's language behavior can be characterized by a predominant preference in sensory modality of predicates utilized. Based on a five minute language sample it was possible to identify a predominant sensory mode of predicates for all subjects. This was accomplished using a conservative decision rule which required a spread of at least 20% of total sensory predicates between the proportion of predicates in the dominant sensory modality and that in the next most frequently utilized modality. Individual differences in the type of predicate usage were compared with self-reports of habitual imaginal modalities and preferences. The subjects were asked to complete two such self-report instruments: the Adjective Questionnaire (constructed by the author for the purpose of this study); and the Verbalizer- Visualizer Questionnaire (Richardson, 1977). There was an extraordinarily powerful correspondence found between the subject's dominant mode of sensory predicate use and his self-reports of preferences for a visual versus auditory (verbal) imaginal style in activities of daily living. In addition, individual differences in type of predicate usage were compared with performance on standard memory tasks where mode of item presentation varied. Performance on these tasks was facilitated when mode of item presentation was congruent with subjects' preferred mode of language predicate usage. These findings support the second NLP hypothesis studied. Finally, subjects' reactions to a simulated therapy transcript, which was constructed so as to cross dominant predicate mode of therapist (auditory or visual) with dominant predicate mode of patient (auditory or visual), were obtained and assessed as a function of preferred predicate mode of the subject and the type of predicate usage employed by the therapist and the patient. The findings support the third NLP proposition studied, that individual differences in preferred representational system influences a person's responsiveness and evaluation of an interpersonal transaction in such a way that a person will evaluate more positively another individual whose predicate usage is similar than one whose predicate language is dissimilar to his own and in such a way that interactions between persons whose language preferences are congruent will be evaluated more positively than interactions between persons whose language preferences are incongruent.