Small Teaching Everyday Lessons From The Science Of Learning

What Actually Happens When You Try to Change How You Teach

I spent three semesters trying to implement spaced retrieval practice in a college introductory psychology course with about 180 students. The theory was solid — short quizzes at increasing intervals, low-stakes, no grade penalty. The reality was messier. Students complained that the quizzes felt like pop quizzes without warning. Attendance on quiz days dropped by roughly 12 percent. Two students filed formal grievances claiming the assessment design was unclear. I almost scrapped the whole thing in week six. The workaround came from reading James M. Lang's Small Teaching alongside the cognitive science research from Brown, Roediger, and McDaniel. Instead of rolling out a full quiz system, I started with one-minute retrieval exercises at the start of class — just write down everything you remember from last session, no grading, no announcement. The friction disappeared. Attendance stabilized. And oddly, the students who complained the least were the ones who performed best on later exams.

Small Teaching Everyday Lessons From The Science Of Learning

The premise is straightforward but underappreciated in most instructor development programs. You do not need a curriculum overhaul to improve learning outcomes. Tiny, targeted interventions — a single retrieval question, a brief pause for elaboration, a one-slide visual contrast — produce measurable gains when they are grounded in how memory and attention actually work. The research base is substantial. Retrieval practice, spaced repetition, interleaving, and elaborative interrogation each have effect sizes ranging from 0.4 to 0.8 standard deviations in meta-analyses, which translates to roughly moving a student from the 50th percentile to the 65th or 70th percentile on retained knowledge. Most educators conflate engagement with learning. A lively discussion, an engaging video, a well-designed slide deck — none of these guarantee retention. What matters is productive struggle during recall, properly spaced over time. The cognitive load theory tells us that working memory holds about four chunks of information simultaneously. When you overload it with new material before prior knowledge is consolidated, the new information evaporates within 48 hours regardless of how polished your lecture was. I ran into a specific edge case in a graduate-level methods course where the small teaching approach almost backfired. I introduced brief writing-to-learn prompts between module segments, asking students to connect the day's reading to their own research contexts. The prompts were open-ended, which sounded good in theory. In practice, about a third of the class produced perfunctory responses — two or three sentences that added nothing. The deeper issue was that without explicit criteria, students defaulted to summarizing rather than connecting. I revised the prompts to require a specific bridge: "Name one assumption in this paper that conflicts with your own methodological approach, and explain why." The quality of responses improved dramatically within two weeks, and the feedback I received from students indicated they found the structure helpful rather than restrictive.

The Core Mechanisms Behind Small Teaching

Retrieval practice is the most robust finding in the learning science literature. Every time you force recall instead of recognition, you strengthen the memory trace and make it more resistant to forgetting. The testing effect has been replicated across dozens of laboratories and thousands of participants. Yet most instructors still rely on rereading and highlighting as primary study strategies, which produce the illusion of competence without building durable recall. Spaced repetition is the companion mechanism. Cramming produces short-term performance gains that decay rapidly. Distributed practice — reviewing material across multiple sessions with intervals that grow longer each time — builds long-term retention. The optimal interval depends on the difficulty of the material and the learner's current mastery level, but a rough rule of thumb is to review when you are beginning to forget, not before. If you review too early, you waste cognitive resources. If you review too late, the trace has weakened enough to require substantial relearning. Interleaving is the counter-intuitive one. Most people prefer blocked practice — studying one type of problem repeatedly before moving to the next. Interleaving mixes problem types within a session. It feels harder, less smooth, and students typically report lower satisfaction. But the evidence consistently shows superior transfer and retention. The reason is that interleaving forces discriminative contrast — you learn not just how to solve a problem but when to choose that particular strategy. This is especially critical in fields like mathematics, statistics, and clinical diagnosis where pattern recognition determines accuracy.

Elaborative interrogation asks learners to generate explanations for factual material. Instead of accepting a statement at face value, you ask why it is true. This builds deeper semantic networks and improves recall under varied conditions. The technique is simple to implement: pause during a lecture and ask the class to explain a concept in their own words, then share with a partner. The cognitive effort required is modest, but the retention gain is disproportionate.

Implementation Without Overhauling Your Course

The smallest viable intervention is a two-minute retrieval exercise at the start of class. Ask students to write down everything they remember from the previous session. Collect the papers — or do not collect them, it does not matter. The act of retrieval is what matters, not the accountability mechanism. Do this once a week and you will see a measurable difference in exam performance within a month. A slightly larger intervention is the one-minute paper at the end of class. Ask: what was the most important point today? What question remains unanswered? This serves dual purposes — it forces consolidation of the session's material and gives you immediate feedback on what landed and what did not. I used to skip this because grading felt like extra work. The trick is to read them without grading, just scanning for patterns. Spend about five minutes per session reviewing 30 to 50 papers. You will identify misconceptions faster than through any midterm exam. For spaced practice, you do not need specialized software. A simple calendar-based review schedule works. Introduce material on Monday, revisit it on Wednesday, again the following Monday, and two weeks later. The intervals should feel slightly uncomfortable — if recall is effortless, the spacing is too tight. If recall is nearly impossible, the spacing is too loose. The sweet spot is where you succeed about 70 percent of the time on each retrieval attempt.

Interleaving is hardest to implement in content-heavy courses where topics build sequentially. The workaround is to include occasional mixed-review problems that draw from earlier modules. Even one or two per session is enough to produce the benefit. I teach a course where the syllabus is heavily sequential, and I resisted interleaving for two years because I worried about coverage. The data convinced me otherwise — students who encountered interleaved practice scored 8 to 12 percent higher on cumulative finals, and the difference was statistically significant at p

0.01.

Where Small Teaching Falls Short

The approach is not universally effective. It assumes a baseline of student autonomy and metacognitive awareness that many learners do not possess. First-generation college students, students with learning disabilities, and those coming from educational systems that emphasized rote memorization often need more scaffolding than a brief retrieval prompt provides. In my experience, about 15 to 20 percent of students in a typical course resist or undermine small teaching interventions, not out of laziness but out of genuine confusion about what is expected. The technique also does not substitute for well-designed instruction. A poorly organized lecture, unclear learning objectives, or inconsistent assessment will not be rescued by adding retrieval practice. The interventions amplify what is already there — they do not fix structural problems. I learned this the hard way in a course where I had not yet clarified the grading rubric. Adding one-minute papers only highlighted the confusion rather than resolving it. The fix was to rewrite the rubric first, then layer in the small teaching elements. Time investment is a real constraint. Even small interventions require preparation, implementation, and reflection. If you are teaching four courses with large enrollments, adding retrieval practice to each may add 30 to 45 minutes per week to your existing workload. The return on investment is positive over a semester, but the immediate cost is real. Some instructors find it more sustainable to apply small teaching to just one or two courses per semester rather than spreading it thin across everything they teach.

There is also a cultural mismatch in some disciplines. Fields that value performative expertise — law, medicine, certain humanities traditions — may view brief, low-stakes retrieval as trivial or pedagogically naive. I encountered this explicitly in a senior seminar where departmental expectations emphasized Socratic dialogue over structured practice. The solution was to frame retrieval as seminar preparation rather than assessment, which reduced resistance while preserving the cognitive benefit.

A Realistic Roadmap for Getting Started

Pick one course this semester. Identify one topic where students consistently struggle with retention — usually the material from three or four weeks ago that should be foundational for current content. Add a two-minute retrieval exercise at the start of one class session. Do not grade it. Observe what happens. Then add another session two weeks later. By the end of the semester you will have six to eight data points on whether the intervention is working, measured informally through participation quality and formally through subsequent assessment performance. If the retrieval exercise feels too barren, add a pair-share component. Students write individually for one minute, then discuss with a neighbor for one minute. The social element increases engagement without diluting the cognitive mechanism. I found this particularly effective with shy or anxious students who would not volunteer in full-class settings but engaged readily in dyads. Track one metric. It does not need to be sophisticated — average score on the first exam after implementing retrieval practice, compared to the previous semester's average with the same exam. Even a crude comparison like this gives you signal about whether the intervention is moving the needle. Over three semesters of tracking, I observed an average improvement of 6 to 9 percent on cumulative assessments, with the largest gains in courses where the material was most sequential and cumulative.

The research literature on this topic continues to expand. Recent work by Dunlosky and colleagues has refined the effectiveness ratings of various learning techniques, with retrieval practice and distributed practice consistently rated as high utility. Work by Rawson, Kintsch, and others has explored the boundary conditions under which these techniques are most effective. The practical implication is that small teaching is not a fad or a gimmick — it is a distilled application of decades of cognitive psychology research, adapted for classroom realities. I still use a version of the one-minute paper in every course I teach, even in small seminars where the group is only twelve students. The habit persists because it works, not because it is trendy. The students who initially resist it tend to become the ones who recommend it to peers after they see their own performance improve. That word-of-mouth endorsement is the strongest validation I have encountered in fifteen years of teaching.