A study investigated cognitive behaviors and processes that contribute to writer's block. Subjects were 10 college undergraduates who had scored at the extreme ends of a writer's block measurement instrument. The 10, 6 "high-blockers" and 4 "low-blockers," varied in their English experience, class standing, and majors. Each was given a writing topic, allowed time to become familiar with it, and then left alone to write for 1 hour. Two videotape machines recorded the students' behaviors and the writing they were doing. At the end of the hour, each student was shown the tapes and questioned about various behaviors that had been recorded. In addition, the amount of time spent in prewriting, planning, and pausing, and the number of words written and deleted were recorded. Finally, the final products were evaluated by two independent raters. Results showed that on the average, high-blockers paused longer, produced shorter drafts, and received lower evaluations than low-blockers. Furthermore, in several cases, high-blockers spent one-half to two-thirds of their time prewriting and planning rather than producing text. Both groups expressed roughly the same number of positive and negative evaluations of their work. The expression of negative evaluations was highly related to low English experience. (Excerpts from three student case studies are included.) (FL)