Even though it has been established that the incorporation of computers into the teaching and learning process enhances student performance, the underlying mechanisms through which this is accomplished have been largely unexplored. The present study aims to shed light on this issue. Two groups of 10 secondary school students were tutored by their geography teacher in how to solve correlational problems. Students in the one group used paper and pencil while students in the other group used a computer spreadsheet. All tutorials were videotaped, transcribed verbatim, and subsequently all transcripts were segmented and coded. The mean frequencies for teacher and student behaviours between the two conditions were then compared. Results indicated that teacher behaviour in the two conditions differed in terms of error feedback, factual and conceptual questions asked, regulation of students, and task management. Regarding student behaviours, the findings showed that the two conditions differed in terms of task engagement, goal setting, and explanations given. On the basis of these findings the issue of mechanisms is discussed and three main implications for the teaching and learning practice are drawn.