Compare retrieval and restudy on two matched fictional vocabularies

This pack is a small within-person comparison, not a promise to make learning permanent. Two matched sets each contain six invented word–symbol pairs. A coin flip assigns one set to retrieval-plus-feedback and the other to equal-time restudy before any results are known. Both sets receive the same initial study time and equal three-minute review windows at T+10 minutes and T+24 hours. At T+72 hours, test all twelve prompts in mixed order without warming up. The 72-hour interval is this exercise’s declared configuration, not a universally best spacing rule. Record cues, assistance, missed reviews and burden. The difference may be informative for these arbitrary items, but small sets can differ by chance. The result cannot establish lifelong memory, intelligence, transfer or a cognitive diagnosis. Stop if the schedule becomes distressing, compulsive or an indefinite streak; formal assessment or concerning memory change requires an appropriate professional.

Start the tutorial Open the canonical Power page All full tutorials

Tutorial curriculum · current public edition · Revision 7T · Full Tutorial Edition

Revision 7T · Full Tutorial Edition · complete governed practice tutorial

Compare retrieval and restudy on two matched fictional vocabularies

Authority and safety

Use only the bounded configuration. Use neutral material and low-stakes self-comparison. The experiment is not an intelligence, memory-health or educational placement test.
Stop if the frame becomes a spaced schedule, repeated testing loop, streak or lifelong-memory promise.
Stop if tracking forgetting causes distress or compulsive checking.

Exact materials

Set X cards: NERAL–amber hexagon; POVIN–teal star; RAKU–silver circle; SELA–violet square; TOVIN–green triangle; WEX–blue crescent.
Set Y cards: BIRA–green crescent; CAVO–violet triangle; DENI–amber circle; FALU–blue square; GOMA–silver star; HESI–teal hexagon.
One coin and assignment card: ‘Heads: X retrieval, Y restudy. Tails: Y retrieval, X restudy. Record before studying and never swap after seeing performance.’
Schedule card: T0 study each set for three minutes; T+10 minutes retrieval set attempts for two minutes plus feedback for one minute, restudy set for three minutes; repeat at T+24 hours; mixed no-warm-up test at T+72 hours.
Mixed test sheet in the fixture with answers covered, plus log fields for coin result, actual times, missed exposure, cues, assistance and burden.
Accessible text, audio, symbol descriptions and external response methods are allowed when declared; no speed scoring.

Set up in this order

Choose an accessible response format and a realistic T0. Write the exact T+10-minute, T+24-hour and T+72-hour times before seeing any cards.
Flip the coin once and record the assignment. Do not relabel or exchange sets after baseline, even if one set feels easier.
Test prior knowledge once by attempting all twelve pairs before study; because the terms are invented, record any accidental familiarity or exposure.
Study Set X for three minutes and Set Y for three minutes in the same format. At T+10 minutes, apply the assigned retrieval or restudy procedure for equal three-minute windows.
At T+24 hours, repeat the same assigned conditions once. Record any missed window or added exposure rather than hiding it.
At T+72 hours, do not warm up. Answer the twelve mixed prompts, score by assigned condition, calculate the difference, record burden and stop.

Supplied fixture

T+72-hour prompt order. Exact vocabulary answers remain in the staged key until every response is locked.
Test order	Prompt	Source set	Condition after coin flip
1	SELA	X	Retrieval if Heads; restudy if Tails
2	GOMA	Y	Restudy if Heads; retrieval if Tails
3	NERAL	X	Retrieval if Heads; restudy if Tails
4	FALU	Y	Restudy if Heads; retrieval if Tails
5	TOVIN	X	Retrieval if Heads; restudy if Tails
6	BIRA	Y	Restudy if Heads; retrieval if Tails
7	WEX	X	Retrieval if Heads; restudy if Tails
8	DENI	Y	Restudy if Heads; retrieval if Tails
9	POVIN	X	Retrieval if Heads; restudy if Tails
10	HESI	Y	Restudy if Heads; retrieval if Tails
11	RAKU	X	Retrieval if Heads; restudy if Tails
12	CAVO	Y	Restudy if Heads; retrieval if Tails

Check your work only after you finish

Lock every response first. The checking page is separate so the answers are not exposed inside this exercise.

Open the separate checking page →

Scoring rule

At T+72 hours, award 1 point only for each exact colour-and-shape pair and 0 otherwise. Each assigned condition contains six items. Calculate Retrieval correct/6, Restudy correct/6, and the raw difference Retrieval minus Restudy; percentage-point difference is (Retrieval/6 − Restudy/6) × 100. Report missed reviews, cues, assistance and burden beside the scores. Do not average in the practice attempts. A difference on twelve invented items is a bounded within-person observation, not proof that one method is universally superior or that anything was learned permanently.

Worked example

The fictional learner flips Heads before study, so Set X is retrieval and Set Y is restudy. Both receive three minutes at T0 and three-minute condition windows at T+10 minutes and T+24 hours.
At T+72 hours, the learner correctly answers SELA, NERAL, TOVIN, WEX and POVIN from Set X, but answers RAKU as ‘silver square.’ Retrieval is 5/6.
For Set Y, GOMA, FALU and DENI are correct; BIRA, HESI and CAVO are incorrect. Restudy is 3/6.
The raw difference is 5 − 3 = +2 correct pairs. Retrieval is 5/6 = 83.3% and restudy is 3/6 = 50.0%, a +33.3 percentage-point descriptive difference after rounding to one decimal.
The learner records no missed review, no cue and burden 2/5, then states that item mismatch and the tiny sample prevent a universal conclusion.

Correct result: In this Heads-assigned example, retrieval is 5/6 and restudy is 3/6: +2 pairs or +33.3 percentage points for the retrieval condition. The outcome applies only to these items and this 72-hour configuration; it does not demonstrate permanent learning or far transfer.

Right and wrong

Right and wrong comparison
Moment	Right / safer	Wrong / riskier	Why
Assigning conditions	Flip once and record the mapping before study.	Move the easier-looking set into the preferred method after baseline.	Preassignment limits a result-driven set-selection bias.
Equalising review	Give each condition a three-minute window at both review points.	Give retrieval extra feedback time without recording it.	Unequal total exposure weakens the comparison.
Running the delayed test	Test all twelve prompts in mixed order before any warm-up.	Review the difficult condition immediately before scoring.	A warm-up would change the declared retention interval and assistance.
Interpreting a difference	Report +2 pairs for one small fictional comparison and name uncertainty.	Conclude that retrieval makes all learning permanent.	The best schedule varies with delay and content, and this is a tiny within-person sample.
Missing a review	Record the missed window and continue only if the result remains useful as a changed configuration.	Backdate the log or add several catch-up repetitions.	Transparent attrition is more informative than hidden extra practice.

Common mistakes and fixes

Common mistakes and corrections
Mistake	Fix
The coin result is not recorded.	Write Heads or Tails and the X/Y mapping before turning over any answer.
One correct colour without its shape earns a point.	Apply the complete-pair criterion consistently to all twelve items.
Practice-round answers are included in the retention total.	Use only the no-warm-up T+72-hour mixed test for the primary score.
A +2 raw difference is called a 2% difference.	Report +2 pairs; for percentages calculate 83.3% minus 50.0% = 33.3 percentage points.
The three review points become an indefinite card streak.	Stop after the declared T+72 test and decide whether the fictional material has any reason to be retained.

Evidence boundary

What the tutorial may—and may not—claim.

Supportable claim

Some learning methods improve delayed retention for specified material or skills, and the delay and testing conditions must be stated.

Measurement boundary

Declare interval, maintenance practice, cues, assistance, attrition, accuracy, errors and transfer task; repeated exposure is not unaided retention.

Myth

Make learning permanent

Metric

Performance after a predeclared delay with maintenance and assistance declared

Boundary

A delayed test cannot by itself establish lifelong memory or transfer.

Negative and limiting findings

The best retrieval schedule changed with the test delay.
Sleep protected some motor performance but did not automatically enhance learning.

Prohibited wording

never forget
permanent learning
one session lasts forever
retention score proves transfer

Direct source register