Wednesday, September 22, 2010

Introducing Difficulty During Learning = Improved Use in New Contexts

I am curious to know what you think and what your experiences have been with the ideas presented by Bjork, D.R. (1994) Memory and metamemoryL Considerations in the training of human beings. In J. Metcalfe & A.P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 185-205). Cambridge, MA: MIT Press. Below is my summary.  Beth

If we want the knowledge and skills that are developed in a class or a learning session to be durable and usable in a real world setting with different conditions then the information needs to be multiply encoded and not bound to a single set of semantics of situational cues. Bjork proposes that to meet that end we need to introduce difficulties for the learner as described below. It is important to note that these strategies will impair performance during the initial learning or training session but enhance long term performance.
  1. Varying the conditions of practice.
    “It has been demonstrated in a variety of ways, and in a variety of motor, verbal and problem solving tasks, that introducing variability and or unpredictability in the training environment causes difficulty for the learner but enhances long term performance and the ability to transfer the training to novel task environment” (p. 181).
    1. Schedule practice trials in a random fashion rather than block by type of task.
    2. Vary parameters of the task like speed or distance to the target.
    3. Increase variability, types, and range of experience or problem.
      I think Trey’s example of having his students develop a story using concepts that they worked with on the test is an example of this. The story activity was a new experience.
    4. Vary incidental environment context.
  2. Provide contextual interference
    An example of this is to make the task environment more variable or unpredictable.  This could involve designing or interweaving material to be learned in a way that creates, at least temporary interference.
    1. Research was conducted on 2 groups of learners who had to learn content in a technical article. One group studied an outline organized like the article and the other group studied an outline inconsistent with that of the article. The inconsistent group’s verbatim recall and recognition was impaired but their performance on tests that required inferred answers and problem solving was enhanced.
  3. Distributed practice on a given task
    This has been repeatedly demonstrated over time.
  4. Reducing feedback to the learner.
    This will make life more difficult during the training but again will enhance post-training performance. In research done involving changing the frequency with which feedback was provided (after every 5minutes or 15 trials or fading the frequency over the length of the training), it was discovered that when the feedback was “faded” over time, simple motor skills were impeded initially but long-term retention of those same skills was enhanced (p. 191).
  5. Using tests as learning events
    There is evidence that the act of retrieval by a recall test can be considerably more potent than a study opportunity in facilitating future recall.
    1. The rational for this is as follows. In responding to the difficulties and challenges induced by the test, the learner is forced into elaborate encoding processes and more substantial and varied retrieval processes (p.192).
If our goal is to produce transfer of the learning or training related to knowledge and skills to a post learning environment then these might be strategies to consider.


  1. Here's a link to a free .pdf of the article - I hope to read it later tonight.


  2. Trey,

    Thank you so much for this link. I had accessed the information in the book itself but didn't have a copy of the article.