The effects of field dependence-independence and graphical/non-graphical user interfaces upon word processing errors

Recent studies suggest that word processing errors can arise from individual differences along the continuum of field dependence-independence, or from features of the word processing software per se. Experiment I compared the performance of field-dependent and field-independent subjects on three types of word processors: Non-graphical, Graphical, and one which combined features of these two major types. Naive subjects were asked to find errors which had been inserted into texts, and to create their own brief essays. Experiment I found that under high-resolution conditions software type influenced the total number of errors found during the proofreading task. An interaction effect was found between software type and field-dependence for the number of transposition errors remaining during the creation task. For Experiment II, naive subjects were asked to perform the same tasks using a cluttered or uncluttered word processing environment. Experiment II found that under less than high-resolution conditions field dependence interacted with interface type for the number of proofreading errors found and total number of errors remaining during the creation task. Implications for future research are discussed.