Abstract: Investigated the impact of varied imaging tasks on the use of sensory predicates by 45 right-handed white females (aged 18-40 yrs). Ss completed a background questionaire and 2 imagery questionaires before completing pleasant and unpleasant imagery tasks in visual, auditory, and kinesthetic sensory modalities. Four additional tasks included having Ss report a pleasant and an unpleasant image using 5 sensory modalities, and earliest memory, and an accomplishment experience. Randomly selected images were coded by therapists. Previous studies of the neurolinguistic programming have considered sensory predicates as a trait measure, indicative of a person's preferred or primary representation system. Results of the present study demonstrates that Ss were able to vary their type of sensory predicates according to the task demands or situational context. Thus, most Ss were auditory types during auditory imaging tasks and kinesthetic types during kinesthetic imaging tasks. Findings are incongruent with R. Bandler and J. Grinder's (1979) conceptualization of representational systems, but they support A. Hammer's (see PA, Vol 70:6385) recommendations for therapists to continuously track and match clients' sensory predicates.