Abstract: Tested R. Bandler and J. Grinder's (1975, 1979) neurolinguistic programming theory that eye movement direction and spoken predicates are indicative of sensory modality of imagery. 39 undergraduates reported on modality, sequence, and vividness of images to questions that evoked either no images or visual, auditory, or kinesthetic images. Eye movement direction and spoken predicates were matched with sensory modality of the questions. Ss reported images in the 3 modes, but no relation between imagery and eye movements or predicates was found. The visual modality was dominant. Visual images were most vivid and often reported. Most Ss rated themselves as visual, and most spoken predicates were visual. Data are discussed in the context of an ever- growing literature that does not support Bandler and Grinder's model and in the context of the difficulties in interpreting the model itself.