Anderson, J. R., & Schunn, C. D. (2000). Implications of the ACT-R Learning Theory: No Magic Bullets. In R. Glaser (Ed.), Advances in instructional psychology (Vol. 5). Erlbaum

John Anderson and Christian Schuss examine some of the implications of ACT-R for learning. Some of the key claims include:

• practice is really what matters
  • “the ACT-R theory makes it clear that there is no magic bullet that allows some way out of these enormous differences in time on task. For competences to be displayed over a lifetime, time on task is by far and away the most significant factor”
  • a high order bit is that students actually practice what needs to be practiced (they claim that this is really all that’s happening in Deliberate practice)
  • they observe a “power law of learning”: the more practice problems, the lower the time to solution
  • alongside the power law of forgetting, this suggests a multiplicative power law of performance
  • degree of initial learning matters much more than anything else about conditions of initial learning (they allow a possible exception for the Spacing effect)
  • they don’t think mnemonics matter except for speed of initial acquisition, which washes out in the long run
• they’re skeptical of contrasts drawn between “real understanding” and “rote learning”
  • they think “real understanding” demonstrated in “insight” experiments really come from the student simply learning different things in the “insight” conditions—more focused on connections, etc. But they’re still acquiring declarative chunks and production rules, still through practice.
  • “understanding is best thought of as having a rich network of highly available declarative chunks and production rules that can be used to solve problems involving those concepts flexibly in many contexts”