Transfer-Appropriate Processing: What It Means for Studying

Transfer-appropriate processing is the idea that the value of a study method depends on the test: you remember best when the mental work you did while studying matches the work the test asks for. So practise the kind of thinking your exam demands, and treat its surface format as a weaker guide than it looks.

Both halves of that answer have evidence behind them. A 1977 experiment found that the better of two study tasks swapped places when the test changed. Two studies from 2007 found that short-answer practice with feedback did more good than multiple choice practice, in one case on exams made entirely of multiple choice questions.

What is transfer-appropriate processing?

It is a claim about the fit between studying and testing. Whenever you study, you process the material in a particular way: you work out what a term means, notice how it sounds, or run a method on numbers. Transfer-appropriate processing says a memory is easiest to use when the test asks for the same kind of processing you did at study.

The idea is defined against an older one. Levels of processing is the view that working with the meaning of material leaves a stronger memory than working with surface features such as sound. It ranks study methods by depth, and transfer-appropriate processing adds a second question: depth for which test?

A 1977 paper by C. Donald Morris, John Bransford and Jeffery Franks put the two views side by side in its title, Levels of processing versus transfer appropriate processing. Its abstract describes three experiments that varied the study task and the type of recognition test, and reports the result of the first: “Experiment I showed semantic acquisition to be superior to rhyme acquisition given a standard recognition test, whereas rhyme acquisition was superior given a rhyming recognition test.”

In plain terms, studying for meaning won on a standard recognition test, and studying for rhyme won on a rhyming recognition test. Which task counted as the better one depended on the test it was measured against. The abstract gives no results for the second and third experiments, so nothing here describes them.

For a student, the practical reading is to find out what your exam will ask you to do with the material, and to practise that: producing definitions, running methods on unseen numbers, applying ideas to new cases. Rereading the chapter rehearses none of those actions.

The room you study in is a separate variable. Our post on context-dependent memory and the exam room covers the physical setting and why a strong exam question tends to swamp its cues. Transfer-appropriate processing concerns the mental work, which comes with you into any room.

Should you practise in the same format as the exam?

The obvious conclusion is to copy the format: essays for an essay paper, multiple choice for a multiple choice paper. For papers that ask you to produce answers, copying the format and copying the thinking amount to the same thing. For a multiple choice paper, two studies from 2007 point somewhere more demanding.

Mark McDaniel, Janis Anderson, Mary Derbish and Nova Morrisette took the question into a college course, in Testing the testing effect in the classroom. Students took weekly quizzes of multiple choice or short-answer questions, with feedback after each quiz. The tests that counted were multiple choice unit tests and a multiple choice cumulative final. In some weeks, students got target material for additional reading as a comparison. The abstract reports: “Quizzing, but not additional reading, improved performance on the criterial tests relative to material not targeted by quizzes. Further, short answer quizzes produced more robust benefits than multiple choice quizzes.”

So on exams made entirely of multiple choice questions, practice that made students produce the answer did more good than practice in the exam’s own format. Multiple choice quizzing helped as well, less robustly, and additional reading did not improve performance.

A second 2007 study adds a condition. In Test format and corrective feedback modify the effect of testing on long-term retention, Sean Kang, Kathleen McDermott and Henry Roediger had participants study four short journal papers. After each one, participants took a multiple choice test, took a short-answer test, read a list of statements, or did a filler task, and the tests and statements covered identical facts. A final test 3 days later mixed multiple choice and short-answer questions.

Experiment 1 gave no feedback on those first tests. The abstract reports that “having had an intervening MC test led to better performance than an intervening SA test, but the intervening MC condition did not differ significantly from the read statements condition.” So with no feedback, multiple choice practice came out ahead of short-answer practice, and the abstract reports no significant difference between it and reading the facts.

For Experiment 2 the researchers added corrective feedback, in their words “to better equate exposure to test-relevant information.” With feedback, “having had an intervening SA test led to the best performance on the final test,” which they read as “suggesting that the more demanding the retrieval processes engendered by the intervening test, the greater the benefit to final retention.”

A multiple choice question puts the correct answer on the page, while a short-answer question you cannot answer leaves you with whatever you managed to write. Corrective feedback, added to equate that exposure, narrows the gap, and once it was in place, the harder format pulled ahead.

How does the 2007 finding fit transfer-appropriate processing?

What follows is our reading of the three abstracts, and it goes beyond any one of them. In the 1977 experiment, the task that won on each test was the one whose processing suited that test: rhyme for the rhyming test, meaning for the standard one. In the 2007 studies, the practice and the final tests aimed at the same thing, the facts and ideas in the material. Short-answer practice asked for those facts with more retrieval effort than picking from a list, and with feedback it did more good.

So the rule has two parts. First, match the kind of thinking: aim your practice at what the test will ask you to know and do with the material. Second, inside that match, make retrieval effortful and check every answer. A multiple choice exam on cell biology asks you to know cell biology well enough to spot it among convincing wrong options, and producing the answer from memory is a demanding way to rehearse exactly that.

This also squares with our guide to desirable difficulties, which advises varying the conditions you practise under. Those conditions are the surface: the room, the order of topics, the wording of a question, and, on our reading of the 2007 results, even the question format, provided you check your answers. Keep varying them. The part to hold steady is the kind of thinking the exam will ask of you.

How do you use transfer-appropriate processing when you study?

1. Read the exam before you plan. Find past papers, the syllabus, or the assessment instructions, and list the verbs: define, calculate, explain, compare, apply. Those verbs describe the processing the test will demand.
2. Name the thinking each topic needs. A formula you must use on new numbers needs unseen problems. A process you must describe in order needs recall of the sequence. A concept you must apply needs fresh cases.
3. Practise by producing from memory. Write the definition, solve the problem, or outline the argument before you look at anything, which is active recall aimed at the exam. For a multiple choice exam, answer each question in your head before reading the options, the habit our guide to multiple choice test taking strategies recommends on the day itself.
4. Check every answer against a key. In the Kang experiments, short-answer practice pulled ahead only once corrective feedback followed it. Mark each attempt, and write the correction beside anything you missed. Our guide to reviewing a practice test properly covers turning those misses into a log you redo later.
5. Vary everything else. Change rooms, shuffle topics together, and reword questions you have seen before, so the thinking stays constant while the surface moves.

What does this research leave open?

All three sources were read as abstracts, and nothing above reports a number or method detail that its abstract does not state. The 1977 paper gives a result for its first experiment only. The classroom study ran in one college course, with feedback after every quiz and multiple choice exams throughout. The second 2007 study measured retention 3 days after participants read short journal papers. Each is a single study, and none of them tells you how large a gain to expect in your own course.

Where GeniusPal fits

The evidence above favours short-answer practice with feedback, and a free GeniusPal quiz is multiple choice throughout, so here is what each plan does.

By default, when you pick an option the quiz marks it straight away and shows the correct one. A switch lets you hold every verdict until the end of the round, where the answer review shows the correct option for each question. Free quizzing is therefore multiple choice practice with corrective feedback, and in the McDaniel classroom study, multiple choice quizzing with feedback still improved exam performance, which additional reading did not. To add a production step on a free account, cover the options and answer in your head first. No study above tested that habit, so treat it as a sensible routine.

Student and Genius add written answers inside the quiz. A setting called Question types offers A few written, Half and half, or All written, and typed answers are marked when you finish the round by comparing the meaning of what you wrote with the expected answer, which the recap shows beside yours. If you think an answer was marked wrong by mistake, you can count it as correct. The paid plans also add flashcards and written recall. In written recall you write an answer in your own words, compare it with a model answer and the key points you covered or missed, then grade your own attempt.

The free account covers 2 generations for the life of the account, at 10 questions per set, with 2 full runs of each set. Student is 14.99 dollars a month for 100 generations, with sets of up to 30 questions. Genius is 59.99 dollars a year, shown as 5.00 dollars a month, under a fair-use ceiling. The daily review is free on every plan: up to 20 questions from across your sets, with the ones you missed first, then the ones you have never answered, then the ones still short of mastery.

GeniusPal reads a PDF, a Word (.docx) file, a PowerPoint (.pptx) deck, plain text, Markdown or CSV up to 10 MB, or a pasted link. It has no optical character recognition, so a scan with no text layer is refused. It writes no summaries, draws no mind maps, and offers no chat window or tutoring dialogue. And it cannot tell you what your exam will ask, so listing the verbs your exam uses stays with you.