In my previous post on transfer, I discussed what it is and isn’t. Here, I draw from the research and highlight the key teacher ‘take-aways’ in terms of what the research suggests for practice. Most of the quotes come from Chapter 3 in How People Learn and the Halpern & Hakel article. These and other sources are referenced at the end.

• Unfortunately, achieving significant transfer of learning has proven to be a difficult chore. Dating back to the beginning of this century, the research literature on transfer is replete with reports of failure.
• A key finding in the learning and transfer literature is that organizing information into a conceptual framework allows for greater “transfer”; that is, it allows the student to apply what was learned in new situations and to learn related information more quickly…. Transfer is affected by the degree to which people learn with understanding rather than merely memorize sets of facts or follow a fixed set of procedures; the research also shows clearly that “usable knowledge” is not the same as a mere list of disconnected facts.
• Varying the conditions under which learning takes place makes learning harder for learners but results in better learning. Like practice at retrieval, varied learning conditions pay high dividends for the effort exerted.
• In the jargon of cognitive psychology, when learning occurs under varied conditions, key ideas have “multiple retrieval cues” and thus are more “available” in memory.
• Learning is generally enhanced when learners are required to take information that is presented in one format and “re-represent” it in an alternative format.
• Cognitive research has established the fact that humans process information by means of two distinct processes…Learning and recall are thus enhanced when learners integrate information from both verbal and visuo-spatial representations.
• Ideally, an individual spontaneously transfers appropriate knowledge without a need for prompting. Sometimes, however, prompting is necessary. With prompting, transfer can improve quite dramatically. The amount of transfer depends on where attention is directed during learning or at transfer.
• Students develop flexible understanding of when, where, why, and how to use their knowledge to solve new problems if they [are instructed in] how to extract underlying themes and principles from their learning exercises.
• Transfer is enhanced by instruction that helps students represent problems at higher levels of abstraction. Helping students represent their solution strategies at a more general level can help them increase the probability of positive transfer and decrease the degree to which a previous solution strategy is used inappropriately (negative transfer).
• Knowledge that is taught in a variety of contexts is more likely to support flexible transfer than knowledge that is taught in a single context. Information can become “context-bound” when taught with context-specific examples.
• Transfer can be improved by helping students become more aware of themselves as learners who actively monitor their learning strategies and resources and assess their readiness for particular tests and performances. Metacognitive approaches to instruction have been shown to increase the degree to which students will transfer to new situations without the need for explicit prompting.
• Teaching practices congruent with a metacognitive approach to learning include those that focus on sense-making, self-assessment, and reflection on what worked and what needs improving. These practices have been shown to increase the degree to which students transfer their learning to new settings and events.
• Because metacognition often takes the form of an internal dialogue, many students may be unaware of its importance unless the processes are explicitly emphasized by teachers. Research has demonstrated that children can be taught these strategies, including the ability to predict outcomes, explain to oneself, note failures to understand, activate background knowledge, plan ahead, and apportion time and memory…
• A number of studies converge on the conclusion that transfer is enhanced by helping students see potential transfer implications of what they are learning.
• Researchers have found much greater transfer on the second day of transfer than the first: this finding suggests that transfer should be viewed as increased speed in learning a new domain—not simply initial performance.
• The fact that learners construct new understandings based on their current knowledge highlights some of the dangers in “teaching by telling.” Lectures and other forms of direct instruction can sometimes be very useful, but only under the right conditions. To counteract these problems, teachers must strive to make students’ thinking visible and find ways to help them re-conceptualize faulty conceptions.
• An especially sensitive way to assess the degree to which students’ learning has prepared them for transfer is to use methods such as “graduated prompting”. This method can be used to assess the amount of help needed for transfer by counting the number and types of prompts that are necessary before students are able to transfer.
• Some learners can transfer after receiving a general prompt such as “Can you think of something you did earlier that might be relevant?” Other learners need prompts that are much more specific. Tests of transfer that use graduated prompting provide more fine-grained analysis of learning and its effects on transfer than simple one-shot assessments.

Teaching for transfer. In light of the research, here are a set of tips for planning, teaching, and assessing to make transfer happen more by design than by luck.

• Establish and keep highlighting clear transfer goals:Explicitly and regularly alert learners to the goal of transfer. Why? Because most students do not realize that this is the goal of learning in school. They are quite convinced – from prior experience and, especially typical tests – that the aim is to recall and plug in what was previously taught. Make clear that the “transfer” game is very different from the “recall” and “plug in” game.Initially, make this clear through think-alouds and explicit reminders of what we are now doing and what its purpose is. Spend time going over the kinds of transfer performance they will need to be able to do well by the end of the unit/course. Examples: “By the end of the unit, you’ll have to do this product on your own, with no prompts or cues from me. Here are a few model student papers from past years, and a rubric describing the end-goal.” Or:  “Initially, you will just mimic some approaches I teach you. But later, you will have to invent your own approach or adapt one you have learned to a new task,” etc.

• Always work on a Gradual Release of Responsibility sequence – in units, and in the course as a whole. The Gradual Release of Responsibility (GRR) model was first articulated for reading instruction by Pearson (1983): I do, you watch; I do, you help; you do, I help; you do, I watch. More formally: Model, Guided Practice, Independent Practice, Independent use of all Strategies. But GRR can also be thought of as the common flow used in athletic training:  simple drill, game situation, game-like conditions, the game. Note that in both cases – reading and athletics – the movement to the final stage of self-regulated complex performance is done in each unit of study. Independent and self-regulated behavior is practiced all the time, not postponed until many discrete “sideline” activities are done over many lessons. You have to practice transfer to master it!

• Highlight Essential Questions to suggest the kinds of connections students will have to make all year: Knowing that essential questions will be used to explore connections between units will make students more likely to make connections on their own – particularly if the assessments regularly involve the questions. Examples: “How should this data be modeled?” when problem solving in math and science. “Who is an American? Says who?” when considering each major topic in US history. “What should we eat?” in health. “What is the author saying without stating?” in English/language arts, etc.

• Have learners practice judgment, not just skill. Transfer is about judging which skill and knowledge to use when. Transfer is thus not about plugging in a “skill” but “judgment” – smart strategy – in the use of a repertoire of skills. (Psychologists refer to this as “conditional” knowledge as opposed to “declarative” and “procedural” knowledge). Make sure learners have opportunity to hear think-alouds of your problem-solving or text-interpreting. Give students practice and get feedback on their attempts to judge which skill or knowledge might be best here. Have learners do think-alouds and provide reports of why they did what they did when they did it. Learning to self-monitor in this way improves both self-assessment and self-adjustment. Thus –

• Assess (without grading) student self-cueing, knowledge retrieval, self-assessment, and self-adjustment on their own, i. e. minus teacher cues. As in sports and independent reading, there have to be countless opportunities for the student to self-prompt, self-assess, and self-adjust – with teacher feedback on the attempts. What does the student do when teachers don’t supply the graphic organizers or a big hint that they should use the writing process we studied yesterday? The research is clear, alas: many students do notself-prompt, in the absence of explicit direction.  “You didn’t say to use it!” is a common comment.The irony here is that this is precisely where students often fall down on standardized tests! Now, there are no teacher or textbook cues as to where the item comes from in the course content, and no overt cues about what content applies are typically given. So, constantly “test” (without necessarily grading them and/or entering the grade in the grade-book) student ability to self-cue. Examples: Give them unfamiliar looking items, writing prompts, problems, etc. – with no mention of which knowledge is being tapped and which strategies and tools they should use. See what they do on their own, then go over the assessment carefully in class soon after – debrief like a coach: What kind of task did they think it was? Why didn’t they think to use Graphic Organizer X or Strategy Y since it should have seemed so clearly related to the task? etc.

• Change the set-up so that students realize that a possible use of prior learning comes in many guises: The research on transfer stresses that students need to be given tasks in which the setting/format/context/mode/language is sufficiently varied over time that students learn they have to think more flexibly in tapping their knowledge. The student too often thinks that – and wishes that! – a recipe or plug-in formula will solve all future needs. Make clear that the initial recipe/structure/scaffold is just that – a scaffold or crutch to be eventually replaced by fluid decision-making. Examples: After learning about gravitational force using balls dropped from towers, give a problem about a ball of a different size and material dropped down a big tunnel in the earth. (See Perkins and Salomon 1989). Teach 2-3 ways to solve every major kind of problem. Give students increasingly odd ‘looks’ at a task or problem that requires the same knowledge (e.g. increasingly non-routine and unobvious problems involving the Pythagorean theorem).

• Have students constantly generalize from (increasingly challenging) specific instances and cases: Transfer is about using helpful ‘big ideas’ to find familiarities and connections where others see only newness and difference. Ask students to generalize from their experience and immediate past lessons to more widely applicable principles, rules, and ideas. Example: After studying westward expansion, ask, “What big generalizations about human migration does this movement west suggest? Can you support your generalizations by other evidence you know of?” Then, ask the same question after studying early 20^th-century immigration, and help them understand that this kind of transfer will be more and more requested of them – i.e. using ideas to see connections and transfer.

• Practice the whole/part/whole development of transfer early and often. Think of what coaches do to break the complex game down into easier but game-like games, e.g. 3 v 3 and 6 v 6 in soccer, to practice the full 11 v 11 game in a more manageable way. In the arts the play or musical piece is broken down into its elements and practiced in chunks, then put back into the whole. Reading instruction proceeds similarly, as does the best problem-solving instruction in mathematics.

• Make sure that any tool or technique is seen as one of many: Too often students work too rigidly or mechanically in applying their learning, rather than seeing application as use of an idea. Example: Teach the 5-paragraph essay, the 3-paragraph essay and the no-paragraph argument (i.e. a powerful advertisement).  Make clear that the transfer goal is “rational persuasion,” not “plug in the 1 tool called 5 paragraphs.”

• Provide many examples of ‘think-alouds’ in transfer situations: Talk out, demonstrate, model the kind of pro-active thinking that needs to take place in one’s head if transfer goals are to be achieved. Example: A math teacher demonstrates how a problem might be solved by “thinking aloud” to reveal strategic thinking and efficient reframing of the problem based on prior knowledge.

• Shift perspective: After any lesson in teaching a particular skill or approach, shift gears. Challenge the prior assumptions, look at the problem or situation from a new point of view, read a different opinion – anything to help students see that the goal is active learning and understanding, not merely taking in the “official” single (glib) answer.

• Require self-assessment and self-adjustment as part of all major assessments. Learning to transfer is greatly facilitated by learning to self-monitor, self-assess, and self-adjust. Initially, make the accuracy of the self-assessment and the self-adjustment more central to scoring than the answer. Example: 5 points for doing the task on your own; 3 points for getting it based on 1 teacher hint; 2 points for getting it after 2 hints, etc.

• Require students to constantly re-word/re-phrase/re-present what they learn: Whether in just taking notes or creatively placing a complete text in a new genre, time, and place: making learners re-cast what they have learned in their own terms is a significant aid to long-term memory and flexible use of knowledge, according to the research on learning and transfer.

Want to learn more? First, we highly encourage you to read Chapter Three on transfer in How People Learn, and the new companion volume How Students Learn: History, Mathematics, and Science in the Classroom, both of which can be read online at the National Academy Press web site. David Perkins has written often and clearly about transfer in numerous articles and in various books, most notably Smart Schools, and his widely-read article with Gavriel Salomon in 1988 in Educational Leadership. Diane Halpern and Milton Hakel have written a very clear summary of the research on transfer (“Applying the Science of Learning to the University and Beyond”) that was published in Change and is available online as a PDF file. Transfer on Trial, a recent collection of essays provides a scholarly but informative look at the research on transfer over the past twenty years. Formal references appear below.

Resources on Transfer

Bransford, J., Brown, A., & Cocking, R. (Eds.). (2001). How people learn: Brain, mind, experience, and school. Washington, DC: National Research Council.
Bruner, J. (1960). The process of education. Cambridge, MA: Harvard University Press.
Dewey, J. (1933). How we think: A restatement of the relation of reflective thinking to the educative process. Boston: Henry Holt.
Gardner, H. (1991). The unschooled mind: How children think and how schools should teach. New York: Basic Books.
Gick M. L., & Holyoak K. J. ( 1987). “The cognitive basis of knowledge transfer”. In S. M. Cormier & J. D. Hagman (Eds.), Transfer of learning: Contemporary research and applications (pp.9-47). New York: Academic Press.
Halpern, Diane F. “Teaching Critical Thinking for Transfer Across Domains: Dispositions, Skills, Structure Training, and Metacognitive Monitoring.” American Psychologist  53.4 (1998): 449-455.
Halpern, Diane & Hakel, Milton “Applying the science of learning to the university and beyond.”  Change; Jul/Aug 2003; 35, 4.
Hestenes, D., M. Wells, M. & Swackhamer, G. (1992)  “Force Concept Inventory,” The Physics Teacher, Vol. 30, March, 141-158. The revised Force Concept Inventory can be found at: http://modeling.asu.edu/R&E/Research.html (I. Halloun, R.R. Hake, E.P. Mosca, and D. Hestenes)
Larkin J. H. ( 1989). “What kind of knowledge transfers?” In L. B. Resnick (Ed.), Knowing, learning and instruction: Essays in honor of Robert Glaser (pp. 284-306). Hillsdale, NJ: Lawrence Erlbaum Associates.
McKeough, A., Lupart J., & Marini, Q. (Eds.).  (1995).  Teaching for transfer: fostering generalizations in learning.  Mahwah, NJ: Lawrence Erlbaum Associates.
Ma, Liping A. (1999) Knowing and Teaching Elementary Mathematics: Teachers’ Understanding of Fundamental Mathematics in China and the United States. Mahwah, NJ: Lawrence Erlbaum Associates.
Perkins, D. (1992). Smart schools: From training memories to educating minds. New York: Free Press.
Perkins, David N., and Tina A. Grotzer. “Teaching Intelligence.” (1997) American Psychologist  52.10: 1125-1133.
Salomon G., & Perkins D. N. ( 1989). “Rocky roads to transfer: Rethinking mechanisms of a neglected phenomenon”. Educational Psychologist, 18, 42-50.
Spiro, R., et al. (1988). Cognitive flexibility theory: Advanced knowledge acquisition in ill-structured domains. Hillsdale, NJ: Lawrence Erlbaum.
Sternberg, R. J., & French, P. A. (1993). Mechanisms of Transfer. In D. K. Detterman & R. J. Sternberg (Eds.), Transfer on Trial: Intelligence, cognition, and instruction . Norwood, NJ: Ablex.
Vanderstoep, Scott W., and Colleen M. Seifert. “Learning “How” Versus Learning “When”: Improving Transfer of Problem-Solving Principles.” Journal of the Learning Sciences  3.1 (1993): 93-111.