Abstract: Two specific hypotheses were examined in this study. The first was that children will exhibit a consistent, observable relationship between stimulus question type and their subsequent eye movements in the direction predicted by Bandler and Grinder (1979). The second hypothesis was that older children, who have been in the educational system for many years, would exhibit a significantly higher degree of uniformity in their eye movements during responses to spelling questions than would very young school children. The subject sample was comprised of children from three age groups -- first grade, fifth grade, and high school. Participants were administered questions which had been categorized as either visual, auditory, or kinesthetic in nature, and asked to spell words of average difficulty from the Spelling subtest of the Wide Range Achievement Test, Revised Edition. The results provided only limited support for the first hypothesis, and no support for the second. In the Question Phase the first- and fifth-graders responded with "auditory" eye movements significantly more often than the other types, regardless of stimulus question type. High school subjects did not evidence any significant difference in the type of responses they generated regardless of question type. A between grades comparison for the Question Phase indicated that the first- and fifth-graders generally made significantly more "auditory" eye movements than did the high school subjects, while high school subjects made significantly more "visual" eye movements. This same trend was also seen in the Spelling Phase data: younger subjects made significantly more "auditory" eye movement responses, while those made by high school students were generally more "visual". Rather than being consistent with the Neurolinguistic Programming (NLP) hypothesis of eye movements, the results are suggestive of a developmentally based bias in the direction of eye movement responses.