Why Do The Math? The Impact of Calculator Use on Participants’ Actual and Perceived Retention of Arithmetic Facts

What is the impact of calculator use on the acquisition of arithmetic facts? Some, but not all, prior research reports that mental practice promotes better subsequent performance than calculator practice (i.e., the generation effect). Is answer production faster and more accurate on a test after practice with versus without a calculator? If so, to what extent does mental practice promote retention of the fact, enabling retrieval (semantic memory) versus streamlined computation algorithms (procedural memory)? To investigate this issue, 32 participants practiced sets of 6 problems (3 large, 3 small) 36 times each, either with or without a calculator. Then, in the test phase, participants produced answers to practiced as well as novel problems, without a calculator. Practice without a calculator led to faster, more accurate responses on the test than practice with a calculator. The data further suggest this speed advantage after no-calculator practice was due to retrieval of the facts (e.g., no problem-size effect, many retrieval reports) rather than optimized computation. Interestingly, participants subjectively reported a comparable increase in the proportion of facts memorized over the course of practice with and without a calculator, but fewer retrievals were reported on the actual test after calculator practice, and a substantial problem-size effect remained on response times. Some theoretical and pedagogical implications are discussed.