Content uploaded by Robert A. Bjork
Author content
All content in this area was uploaded by Robert A. Bjork on Jan 03, 2021
Content may be subject to copyright.
Bjork, R. A., & Bjork, E. L. (2020). Desirable difficulties in theory and practice. Journal of
Applied research in Memory and Cognition, 9 (4), 475-479.
2
Desirable Difficulties in Theory and Practice
Robert A. Bjork and Elizabeth L. Bjork
University of California, Los Angeles
The articles in this special issue have triggered memories of the events and research
findings that led us to the idea that difficulties can be desirable, but they have also emphasized
the complexities and challenges of trying to incorporate such difficulties into teaching and self-
regulated learning. Before we go on to comment on the individual papers in this special issue,
we provide a bit of history with respect to the considerations that led to the idea that difficulties
can be desirable.
Some Reminiscing Re the Desirable Difficulties Idea
In the framework we titled A New Theory of Disuse (Bjork & Bjork, 1992), which was
written for a Festschrift honoring William K. Estes, we tried to capture what we labelled some
“important peculiarities” (p. 36) of human learning and memory. We came up with the
somewhat awkward title for our framework by reference to Thorndike’s (1914) original Law of
Disuse, which stated that learned habits, without continued practice, fade or decay from memory
with the passage of time. We wanted to give Thorndike credit for emphasizing that use is critical
for keeping memories accessible, but to also point out that the decay idea, which remains
appealing to most people, had been completely discredited by McGeoch (1932) and others.
Instead, we wanted to convey that memory representations remain in memory, but can become
inaccessible—other than, perhaps, in the presence of rare and unique cues.
In our Festschrift chapter we sought to provide linkages to some of the dynamics that
emerged from Estes’ stimulus fluctuation theory, especially the “peculiar” idea that forgetting
can enable learning. In Estes’ theory, which focused on learning by non-human animals, but was
extended by Bower (1972) and others to human learning, an animal in a conditioning experiment
was assumed to sample “stimulus elements” in the environment and responding was determined
by the proportion of sampled elements associated with a given response. Estes assumed that—
owing to attentional and other factors—some elements were “available” and others not available
at any given time and that forgetting took the form of “conditioned” stimulus elements
fluctuating out of the set “available” to an animal—to be replaced by yet-to-be conditioned
elements. Such fluctuation would then lead to forgetting (non-responding), but also create the
potential for additional conditioning/learning (that is, associating additional stimulus elements to
the response in question).
In our framework we assumed that an item in memory can be characterized by two
strengths—storage strength (how well learned an item is, as defined by how interconnected it is
with related items in memory) and retrieval strength (the current ease of access to that item
given the current cues). Such a distinction was certainly not new with us; it corresponds to
Estes’ (1955) distinction between habit strength and response strength and to Hull’s (1943)
distinction between habit strength and momentary reaction potential. And more broadly, it
corresponds to the time-honored distinction between performance, which we can measure at any
given point, and learning, which can only be measured at a delay—and indirectly by the rate of
3
loss of retrieval strength or the rate of relearning as measured by the regaining of retrieval
strength. Subsequent interactions with motor-skills colleagues, especially Richard Schmidt and
Robert Christina, made us aware of linkages to related research on learning versus performance
in the motor-skills domain (see Christina & Bjork, 1991; Schmidt & Bjork, 1992; for a review
see Soderstrom & Bjork, 2015).
What was “new” about our New Theory of Disuse, versus those precursors, is our
specification of how storage strength and retrieval strength interact. In our framework the higher
the current level of storage strength the larger the gain in retrieval strength that results from
restudying or retrieving, whereas—and much less intuitively—the higher the current level of
retrieval strength the smaller the gain in storage strength that results from restudying or
retrieving. Thus, forgetting (loss of retrieval strength) can enhance learning (the gain in storage
strength), which is why, in the theory, manipulations such as spacing and variation, which reduce
retrieval strength, can enhance learning, as measured by performance at a delay.
Metamemory Considerations
The fact that conditions of learning that make performance improve rapidly often fail to
support long-term retention and transfer, whereas conditions that create challenges (i.e.,
difficulties) and slow the rate of apparent learning often optimize long-term retention and
transfer, means that learners—and teachers—are vulnerable to mis-assessing whether learning
has or has not occurred. Thus, to the extent that we interpret current performance as a valid
measure of learning we become susceptible not only to mis-judging whether learning has or has
not occurred, but also to preferring poorer conditions of learning over better conditions of
learning.
Desirable versus Undesirable Difficulties
The term desirable difficulty, coined in 1994 (Bjork 1994a, 1994b), has a nice
alliteration, but it has led to our having to emphasize that the word desirable is important. Many
difficulties are undesirable during instruction and forever after. Desirable difficulties, versus the
array of undesirable difficulties, are desirable because they trigger encoding and retrieval
processes that support learning, comprehension, and remembering. If, however, the learner does
not have the background knowledge or skills to respond to them successfully, they become
undesirable difficulties. We entitled a short chapter Making things hard on yourself, but in a
good way: Creating desirable difficulties to enhance learning
1
to emphasize that the level of
difficulty matters (Bjork & Bjork, 2011; 2014).
The level of difficulty that is optimal, therefore, will vary with the degree of a learner’s
prior learning. In general, for example, it is desirable to have learners generate a skill or some
knowledge from memory, rather than simply showing them that skill or presenting that
knowledge, but a given learner needs to be equipped by virtue of prior learning to succeed at that
generation—or at least succeed in activating relevant aspects of the skill or knowledge, which
may then potentiate subsequent practice or study (e.g., Little & Bjork, 2016; Richland, Kornell,
1
We thank Steve Smith, Texas A&M University, for suggesting the “making things hard on yourself” title.
4
& Kao, 2009). Guadagnoli and Lee’s (2004) “Challenge Point” framework picks up on that idea
in the domain of motor skills.
Comments on the Articles in this Forum
The articles the Editor recruited for this special issue illustrate in a compelling way the
range of potentially important applications of desirable-difficulties research. The articles also,
however, do an excellent job of documenting the challenges that are inherent in trying to
introduce desirable difficulties into real-world settings where the necessary changes may well be
undesired by learners—and perhaps by teachers as well. The Editor has provided a compelling
overview of the articles in this special issue—to which we add a few specific comments on each
article.
Enhancing Law School Instruction
In his contribution, Schulze (2020) reports on an effort to upgrade Florida International
University’s law-school instruction by drawing on the cognitive science of learning and, in
particular, by incorporating desirable difficulties. What Shulze and his FIU colleagues have
achieved is both amazing and inspiring. That they were able, by revamping FIU’s law-school
instruction, to increase the rate of FIU students passing the bar exam from about fifth among
Florida law schools to first in the majority of recent exams is an amazing achievement. That
achievement is truly inspiring because it far exceeds any prediction based on the entering
credentials of FIU students, as measured by their LSAT scores and other metrics, versus other
Florida law schools. It may be an unwarranted and over-optimistic generalization on our part,
but such findings suggest that across education more broadly optimizing instructional practices
may act as a kind of leveler.
We found Shulze’s comments on testing to be especially interesting. For the two of us
and other cognitive scientists, testing is viewed as having multiple pedagogical advantages from
both a memory standpoint and meta-memory standpoint: The retrieval processes triggered by
testing can enhance later retrieval, reduce the likelihood of recalling competing incorrect
information, and provide feedback to learners as to what has and has not been understood and
learned. In the law-school climate, though, according to Shulze, “our obsession with summative
assessment leads students to believe that testing, or retrieval practice, is meant in all cases only to
assess the student’s ability, knowledge, and aptitude,” and, as a consequence, “students cannot
fathom the idea of self-testing unless they are fully prepared for the real exam; and no law
student has every felt fully prepared for an exam” (Schulze, 2020).
Spacing Effects in Mathematics Education: Unanswered Questions
There often seems to be skepticism as to whether laboratory findings that seem to have
important educational implications will actually transfer to the real world of education—even
when the evidence is strong, as in the case of the time-honored finding that long term retention of
skills and knowledge benefit from spacing instruction or practice. As a consequence,
instructional procedures are often guided by intuition and/or by whatever are standard practices,
rather than by experimental research. In a truly important—and in some ways, heroic—project,
5
Rohrer (2012) took on the challenge of examining the extent to which mathematics education in
the real world of schools would profit from spacing (and, hence, interleaving), such as when and
how to use the Pythagorean Theorem. He found strong support for the benefits of spacing, a
result that is especially important given that most workbooks, as well as classroom exercises,
involved blocked, not interleaved, practice.
In the current commentary, Rohrer and Hartwig (2020) list some “unanswered questions”
about spaced/interleaved mathematics practice. They point out that it is by no means a given that
research-based changes will get into the classroom (“Too often, the classroom is where
promising interventions go to die”), given beliefs that students and teachers may hold. One
question has to do with the extent to which learners are deterred from adopting
interleaving/spacing by the combination of lower performance and greater experienced difficulty
when spacing is introduced. Another has to do with whether learners actually believe spaced
practice is effective and, relatedly, whether massing and blocking provide an “illusion of
mastery” that is difficult to overcome. They also, importantly, summarize evidence that the
benefits of interleaving can go beyond the benefits of the spacing that interleaving creates.
Reflections on Teaching Desirably Difficult Learning Strategies
In talking to various audiences over the years, we have often been asked whether we have
written some kind of manual on how to incorporate desirable difficulties into one’s teaching or
self-regulated learning. That question has made us to realize that we have been prone to
assuming, unrealistically, that simply telling learners and teachers about relevant research
findings is enough. In their contribution, Biwer, De Bruin, Schreurs, and oude Egbrink (2020)
discuss their impressive effort to implement a “study smart” program. What they have learned
about the opportunities and obstacles in creating a program to make students more effective
learners is interesting and important.
The challenges Biwer et al. have confronted are as informative as the successes they have
achieved. Among those challenges is that learners bring with them “naïve theories” about
learning strategies that need to be “debunked.” Another challenge is that what students have been
doing has worked—in the sense of getting them to where they are—and what they are doing may
also be consistent with what they have been taught earlier in their academic careers.
When and What Difficulties Are Desirable for Children?
The essay by Knabe and Vlach (2020) draws our attention to the fact that while the
desirable difficulty of spacing, rather than massing, practice is one of the most reliable and
extensively studied phenomena in the field of learning, it remains the case that little research has
been focused on the potential of spaced practice to improve children’s learning, particularly their
learning in a classroom setting. While pointing out the need to investigate the promise of this
strategy for children’s learning more fully, they also provide an important warning—namely, that
for such studies to be meaningful, they must be conducted in ways that take into account the
potential effects of the many individual differences among children in the early years of their
living and learning.
6
They predict, for example, that individual differences in visual attention, memory
capacity, prior knowledge, and metamemory abilities—all of which are rapidly developing, but
at different rates, during the early years of life—will play a major role in determining whether
the learning of an individual child or age group can profit from a spaced versus a massed
schedule of study. Such developmental differences will clearly present K-12 instructors with a
very difficult task of deciding when the learning of a given child would or would not profit from
spaced versus massed study schedules, which makes the need for careful research on the
limitations and boundary conditions of the spacing effect in early education a critical need.
The Role of Cognitive Effort in Motor Skill Learning
Hodges and Lohse provide a thought-provoking analysis of three areas within motor
learning where they argue that our thinking about desirable difficulties needs to be refined. Their
analysis focuses on three intriguing questions: (a) If, in general, learners feel they are learning
less well when faced with desirable difficulties, how can that finding be reconciled with those in
the field of motor learning showing that learning is enhanced when individuals feel they are
succeeding?; (b) can the concepts or assumptions about desirable difficulties, the role played by
errors in learning, and cognitive effort be reconciled with findings obtained in the area of implicit
motor learning?; and (c) can partners serve as a desirable difficulty under conditions of shared
practice? Their comments regarding possible answers to these questions are insightful and
suggest avenues for future research that may prove quite productive.
Also provided is an insightful discussion of the need to identify desirable difficulties
prospectively, highlighting that such difficulties be task relevant, novel (i.e. not something the
learner is already doing), and potentially solvable by the learner. This latter criterion is one we
frequently find we must emphasize when speaking with educators and creators of instructional
materials: Namely, that for difficulties to be desirable—that is, promote learning—they must
present challenges to the learner but not be of such difficulty that the learner cannot eventually
meet or overcome them. Use of “adaptive” learning schedules where levels of difficulty are
tailored to an individual’s past successes represent good instantiations of this necessary feature.
Also, of value is the discussion of the role of making errors, often considered something to be
avoided, in optimizing learning. Indeed, from the standpoint of the desirable difficulty
framework learners should interpret errors as opportunities for enhanced learning, but that is
much easier said than done.
Interactions of Motivation and Cognition in Self-Regulated Learning
Finn (2020) provides a convincing case that bringing desirable difficulties into the real
world of education requires addressing the motivational factors she sketches in her essay. The
two of us have actually been quite guilty of ignoring such factors, the importance of which
comes through, if less explicitly, in some of the other articles in this forum as well. If one’s goal
is to have students replace less-effective learning activities—activities that may have become
habitual and may even have been encouraged by teachers—with more effective activities, the
issues of motivation mentioned in Finn’s essay must be addressed.
7
Among other important observations, Finn points out that students’ memories of their
past academic experiences—and achievements, or lack thereof—provide a basis for their
expectations and goals. Such expectations and goals, in turn, can heavily influence both
students’ effort to learn and their selection of learning procedures. From that standpoint, as she
argues, research of achievement motivation and on judgment and decision making becomes
highly relevant.
Strategies to Motivate Students to Embrace Desirable Difficulties
Zepeda, Martin, and Butler (2020) focus on a challenge that appears explicitly or
implicitly in a number of the other essays in this special issue: How to get learners to embrace
and employ desirable difficulties in managing their own learning. Years ago, we thought—
somewhat naively—that simply showing the benefits of incorporating desirable difficulties
would be enough for students and others to introduce such difficulties into their own learning.
Desirable difficulties are, however, difficulties, and any benefits are long-term benefits, whereas
the short-term consequences are typically poorer performance, so it seems obvious at this point
that convincing learners to introduce desirable difficulties is a major challenge.
In their scholarly analysis of how learners might be motivated to incorporate desirable
difficulties Zepeda et al. review relevant research in the broader domain of psychological
research on motivation. More specifically, they provide brief reviews of five approaches that
may provide insights into how learners can be motivated to introduce desirable difficulties into
the management of their own real-world learning. That there is important research to be carried
out on how motivational factors influence and interact with learning strategies comes through
very clearly from their analysis.
Concluding Comments
The commentaries in this special issue have made us look both backward and forward
with respect to the real-world applications and implications of desirable difficulties findings.
Looking back, it is surprising from the current vantage point that it took us a while to realize that
the dynamics we viewed as theoretically interesting actually had real-world importance for
teaching and self-regulated learning as well. Looking forward, the commentaries in this issue
have, if anything, increased our estimation of the potential to enhance teaching and self-regulated
learning by introducing desirable difficulties, but have also opened our eyes with respect to the
challenges that remain if that potential is to be realized.
8
References
Biwer, F., de Bruin, A. B. H., Schreurs, S., & oude Egbrink, M. G. A. (2020). Future steps in
teaching desirably difficult learning strategies: Reflections from the Study Smart
program. Journal of Applied Research in Memory and Cognition, 9, 439-446.
Bjork, E. L., & Bjork, R. A. (2011). Making things hard on yourself, but in a good way:
Creating desirable difficulties to enhance learning. In M. A. Gernsbacher, R. W. Pew, L. M.
Hough, & J. R. Pomerantz (Eds.), Psychology and the real world: Essays illustrating
fundamental contributions to society (pp. 56-64). New York: Worth Publishers.
Bjork, E. L., & Bjork, R. A. (2014). Making things hard on yourself, but in a good way:
Creating desirable difficulties to enhance learning. In M. A. Gernsbacher and J. Pomerantz
(Eds.), Psychology and the real world: Essays illustrating fundamental contributions to
society (2nd edition). (pp. 59-68). New York: Worth.
Bjork, R.A. (1994a). Memory and metamemory considerations in the training of human
beings. In J. Metcalfe and A. Shimamura (Eds.), Metacognition: Knowing about
knowing (pp.185-205). Cambridge, MA: MIT Press.
Bjork, R.A. (1994b). Institutional impediments to effective training. In D. Druckman and
R.A.Bjork (Eds.), Learning, remembering, believing: Enhancing human performance
(pp.295-306). Washington, DC: National Academy Press.
Bjork, R. A. (2015). Forgetting as a friend of learning. In D. S. Lindsay, C. M. Kelley, A. P.
Yonelinas, & H. L. Roediger, III (Eds.) Remembering: Attributions, processes, and control in
human memory: Papers in honour of Larry L. Jacoby (pp. 15-28). New York: Psychology
Press.
Bjork, R. A., & Bjork, E. L. (1992). A new theory of disuse and an old theory of stimulus
fluctuation. In A. Healy, S. Kosslyn, & R. Shiffrin (Eds.), From learning processes to
cognitive processes: Essays in honor of William K. Estes (Vol. 2, pp. 35-67). Hillsdale, NJ:
Erlbaum.
Bjork, R. A., & Bjork, E. L. (2019). Forgetting as a friend of learning: Implications for teaching
and self-regulated learning. Advances in Physiology Education, 43, 164-167.
Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-regulated learning: Beliefs, techniques,
and illusions. Annual Review of Psychology, 64, 417-444.
Bower, G. H. (1972). Stimulus-sampling theory of encoding variability. In A. W. Melton & E.
Martin (Eds.), Coding processes in human memory (pp. 85-123), Washington, DC: V. H.
Winston.
9
Christina, R. W., & Bjork, R. A. (1991). Optimizing long-term retention and transfer. In D. Druckman
& R. A. Bjork (Eds.), In the mind’s eye: Enhancing human performance (pp. 23-56). Washington,
DC: National Academy Press.
Druckman, D. & Bjork, R.A. (Eds.) (1994). Learning, remembering,believing: Enhancing
human performance. Washington, DC: National Academy Press.
Estes, W. K. (1955). Statistical theory of distributional phenomena in leaning.
Psychological Review, 62, 369–377.
Finn, B. (2020). Exploring interactions between motivation and cognition to better shape self-
regulated learning. Journal of Applied Research in Memory and Cognition, 9, 461-467.
Guadagnoli, M. A., & Lee, T. D. (2004) Challenge point: A framework for conceptualizing the
effects of various practice conditions in motor learning. Journal of Motor Behavior, 36, 212-
224.
Hertel, P. (2020). Everyday challenges to the practice of desirable difficulties, Journal of
Applied Research in Memory and Cognition, 9, 425-427.
Hodges, N., & Lohse, K. R. (2020). Difficulty is a real challenge: A perspective on the role of
cognitive effort in motor skill learning. Journal of Applied Research in Memory and
Cognition, 9, 455-460.
Knabe, M. L., & Vlach, H. A. (2020). When are difficulties desirable for children? Firtst steps
toward a development and individual differences account of the spacing effect. Journal of
Applied Research in Memory and Cognition, 9, 447-454.
Little, J. L. & Bjork, E. L. (2016). Multiple-choice pretesting potentiates learning of related
information. Memory & Cognition, 44, 1085-1101.
McGeoch, J. A. (1932). Forgetting and the law of disuse. Psychological Review, 39, 352-370.
Richland, L. E., Kornell, N., & Kao, L. S. (2009). The pretesting effect: Do unsuccessful
retrieval attempts enhance learning? Journal of Experimental Psychology: Applied, 15, 243-
257.
Rohrer, D. (2012). Interleaving helps students distinguish among similar concepts. Educational
Psychology Review, 24, 355-367.
Rohrer, D., & Hartwig, M. K. (2020). Unanswered questions about spaced and interleaved
mathematics practice. Journal of Applied Research in Memory and Cognition, 9, 433-438.
Schmidt, R. A., & Bjork, R. A. (1992). New conceptualizations of practice: Common principles
in three paradigms suggest new concepts for training. Psychological Science, 3, 207-217.
10
Schulze, L. N. (2020). Legal education’s difficulty with “desirable difficulties” and its impact on
student success and bar passage rates. Journal of Applied Research in Memory and
Cognition, 9, 428-432.
Soderstrom, N. C., & Bjork, R. A. (2015). Learning versus performance: An integrative review.
Perspectives on Psychological Science, 10, 176-199.
Thorndikc, E. L. (1914). The Psychology of Learning. New York: Teachers College.
Zepeda, C. D., Martin, R. S., & Butler, A. C. (2020). Motivational strategies to engage learners
in desirable difficulties. Journal of Applied Research in Memory and Cognition, 9, 468-
474.
