Dr. Wallace's Favorite Mental Models

The Spacing Effect (Distributed Practice)

Why ten minutes of review spread across two weeks beats two hours crammed into one night β€” and why the spread-out version is the one still there months later.

130-Second Overview

The idea in one sentence: the same total amount of study or training time produces dramatically better long-term memory when it's broken into separate sessions with real gaps between them, instead of done all at once β€” even though the crammed version feels more confident and effective while you're doing it.

Where it comes from

German psychologist Hermann Ebbinghaus spent the 1880s memorizing lists of nonsense syllables on himself and timing how fast he forgot them, publishing the results in 1885 as Über das GedΓ€chtnis ("On Memory"). He mapped the "forgetting curve" β€” a sharp early drop in recall that gradually levels off β€” and showed that reintroducing material before it's fully lost resets that curve more efficiently than one long initial study session. Twentieth-century research cemented the finding into the "spacing effect," and a landmark 2006 meta-analysis of 254 studies by Cepeda, Pashler, Vul, Wixted, and Rohrer confirmed it holds overwhelmingly across ages, subjects, and task types.

Why it matters

Training budgets, study hours, and onboarding time are all limited; this model is a free lunch inside that constraint β€” it doesn't ask for more hours, it asks for the same hours arranged differently. Anyone designing a curriculum, a certification program, or their own exam prep is already making a scheduling choice whether they realize it or not, and the cramming default is usually the losing one.

2Mental Model Compass

Type of thinking
Memory science / scheduling
Best for
Anything you need to retain for weeks, months, or indefinitely
Difficulty
Simple idea, hard to follow under deadline pressure
Time to apply
Minutes to plan a schedule, then it runs itself

Reach for it any time you're designing a study plan, a training curriculum, an onboarding sequence, or a certification program β€” anywhere the real goal is "still correct in three months," not just "passed the quiz on Friday."

3Walk Through It

4Build a Spacing Schedule

Enter how many days you have until you need this material to be solid, and how many separate review sessions you can realistically fit. This generates an expanding-interval schedule β€” reviews start close together and the gaps widen as the material gets more secure, which is the pattern spacing research consistently favors over evenly-spaced reviews. Treat it as a practical planning heuristic inspired by that research, not a clinically validated algorithm like the ones purpose-built spaced-repetition apps use for individual flashcards.

5Same Total Hours, Two Schedules

Pick a situation below to see how the exact same amount of total study or training time produces very different outcomes depending on whether it's delivered all at once or spread out.

Massed practice (cramming)

Distributed practice (spaced)

6Diverse Examples

Medical education
Board-exam prep programs deliberately re-cover the same high-yield topics across multiple later blocks instead of teaching each organ system once and moving on, because licensing knowledge needs to survive years of practice, not just one exam day.
Corporate compliance
Many companies have moved from one long annual compliance session to short quarterly refreshers on the same material, specifically to counter the steep early drop Ebbinghaus mapped in the forgetting curve.
Consumer apps
Spaced-repetition software β€” SuperMemo, built by Piotr WoΕΊniak starting in the late 1980s, and its popular open-source descendant Anki β€” automates the expanding-interval idea, scheduling each flashcard's next appearance based on how well you just recalled it.
Music & sports coaching
Coaches and music teachers generally favor short daily practice sessions over occasional long ones for the same total hours, because durable skill retention tracks the same spacing pattern researchers find for verbal memory.
New-hire certification
Certification programs that test once at the end of week one often see strong pass rates but weak competence months later; programs that re-test the same material at 30, 60, and 90 days see far better durable performance on the job.
Marketing & brand recall
Advertisers commonly spread the same campaign across several weeks rather than one saturation burst, because a brand name needs to survive the gap until the next purchase decision, not just register once.

7Worked Application

Situation: an operations manager needs a 40-person support team to correctly apply a new refund policy that replaces one the team has used for three years β€” and needs it to hold up in month six, not just pass Friday's quiz.

  • Step 1 β€” Name what needs to survive, and for how long: the policy will be used indefinitely, so the real goal is "correctly applied on a live call in month six," not "scored well on a one-time quiz right after training."
  • Step 2 β€” Stop treating this as a single training event: a 90-minute session, however thorough, fades on the same curve Ebbinghaus mapped in 1885 β€” most of it is gone within a few weeks without any reinforcement at all.
  • Step 3 β€” Build an expanding-interval schedule around the launch: day 1 (initial training), day 3 (a five-question recap quiz), day 10 (recap plus three realistic refund scenarios), day 30 (a spot-check during routine call review), day 90 (a short refresher only for anyone whose live-call accuracy has slipped).
  • Step 4 β€” Measure retention, not attendance: track live-call accuracy on refund decisions at each checkpoint rather than just training completion, since the forgetting curve is invisible on a sign-up sheet β€” it only shows up when someone actually needs the information under real conditions.

8Common Mistakes

Confusing more hours in one sitting with more learning. The same total hours split with real gaps between them produces far better delayed recall than the identical hours spent all at once β€” this is a scheduling change, not an extra-effort one.
Spacing sessions evenly instead of letting the gaps expand. Reviewing every single day isn't the goal; if the gaps never widen, you're paying for more sessions than durable retention actually requires.
Testing immediately and declaring victory. A quiz given minutes after training measures short-term memory, which is almost always intact right then β€” it says nothing about what survives the forgetting curve's first steep drop over the following days.
Treating "spaced" as a synonym for "less." The research gain comes from rearranging the same amount of exposure over time, not from cutting the total amount of exposure β€” cutting hours to "make room for spacing" gives up the benefit it's trying to capture.
Spacing without any real retrieval. Passively re-reading material on a delay still helps somewhat, but pairing each spaced session with an actual retrieval attempt β€” a quiz, a self-test, a live scenario β€” produces a stronger effect than re-exposure alone.
Ignoring how long the memory actually needs to last. A one-time pitch delivered tomorrow doesn't need a three-month refresher plan; a safety protocol used for years does. Scale the spacing plan to the real retention horizon, not a one-size-fits-all calendar.

9When Not to Use It

  • When something is used once, immediately, and never again. A one-off improvised task with no future recall requirement doesn't justify building a multi-week review schedule around it.
  • When there's genuinely no calendar time before the deadline. Spacing requires days or weeks to work with; if the test is tomorrow, massed practice tonight isn't useless, it just won't hold up as long afterward as a spaced plan would have.
  • For skills still being physically acquired in a single block. Spacing's benefits are most consistent and best-documented for verbal and declarative memory (facts, procedures, vocabulary); some fine motor sequences still in the earliest stage of acquisition benefit more from short-term repetition within one session before spacing is introduced.
  • As an excuse to barely practice at all. "Spacing it out" is not the same as doing very little β€” the schedule still needs enough total reinforcement, just distributed with gaps rather than removed.
Common myth to retire: "practice makes perfect" is often repeated as if more repetition always helps, full stop. The research says how repetition is distributed matters as much as how much of it there is β€” cramming is still repetition, and it reliably underperforms the identical amount of spaced repetition on anything you need to remember past the next few days.

10Related Models

The Serial Position Effect

Concerns where something sits within a single list or session β€” items at the start and end are recalled best. The spacing effect is about the gap between separate sessions, not position within one; they describe different mechanisms inside the same study of memory.

The Zeigarnik Effect

Explains why unfinished tasks nag at attention and get remembered involuntarily. A different mechanism from spacing β€” incompleteness versus gap-then-retrieval β€” but both are about how memory behaves over time rather than at a single moment.

Retrieval Practice (The Testing Effect)

A close companion, not a duplicate: spacing is about when you review; retrieval practice is about how β€” actively recalling information rather than passively re-reading it. The two combine, and most real training programs that work well are using both at once.

11Try It Yourself

12Save as Skill

Turn this mental model into a reusable instruction you can hand to any AI assistant β€” or to yourself β€” before designing any study plan, training curriculum, or onboarding sequence that needs to survive longer than the session it's delivered in.

Copied!