EDUF3040 · Psychological Perspectives in Education
Practice, Retrieval and Feedback
Week 10 shows how practising, retrieving and applying knowledge consolidates long-term memory: deliberate practice, retrieval practice and the testing effect, spacing versus cramming, and the central role of feedback (APA Principle 6). It explains why high-information feedback that tells students why they erred and how to improve far outperforms bare corrective or reinforcement feedback. In University of Sydney EDUF3040 this is examined as multiple-choice items on the testing and spacing effects and short-answer questions asking you to design durable practice or rank feedback by impact.
What this chapter covers
- 01APA Principle 6: long-term knowledge/skill depends on practice, plus clear, timely feedback
- 02Deliberate practice (Ericsson): attention, rehearsal, repetition; necessary but not sufficient for expertise
- 03Retrieval practice / testing effect: retrieving strengthens memory (Karpicke); delay testing, prefer free recall over recognition
- 04Spaced (distributed) vs massed practice ('cramming'); the spacing effect (Ebbinghaus lineage)
- 05Feedback definition and purpose (Hattie & Timperley): reduce the gap between current and desired performance
- 06Feedback effectiveness (Wisniewski, Zierer & Hattie): overall d ≈ 0.55; high-information d ≈ 0.99 ≫ corrective ≈ 0.46 ≫ reinforcement/punishment ≈ 0.24
- 07'Medal and a mission' / stars-and-stairs; task-focused not person-focused; encouragement vs person-praise
- 08Study-strategy utility (Dunlosky et al.): high (practice testing, distributed practice) vs low (highlighting, rereading)
Which study plan builds durable learning - and better feedback
- +1Apply the testing effect. Retrieving knowledge (self-testing) strengthens it more than restudying, because each successful retrieval improves the ability to retrieve again (Karpicke) - Student Y's self-testing beats Student X's rereading.
- +1Apply the spacing effect. Distributing the same total study time across gaps outperforms massing it, because the gaps allow forgetting of irrelevant features while relevant ones are reactivated, aiding durable, generalisable memory - Student Y's spacing beats Student X's cramming.
- +1Make the prediction. Student Y retains more, and the advantage grows on a delayed test; Student X may feel fluent the night before (a familiarity illusion) but forgets faster.
- +1Strengthen the retrieval design. Make Y's testing effortful and useful: use free-recall/short-answer rather than recognition, test after a delay so the material must be reconstructed from long-term memory, and delay feedback slightly rather than giving it instantly.
- +1Upgrade the feedback. Replace a bare mark or 'right/wrong' with high-information feedback (d ≈ 0.99): tell the student what the error was, why it happened and how to avoid it next time - a task-focused 'medal and a mission' rather than person-praise.
Key terms
- Retrieval practice / testing effect
- Retrieving knowledge from memory changes and strengthens it - 'every time a person retrieves knowledge, that knowledge is changed, because retrieving improves the ability to retrieve it again' (Karpicke). Practically: test important content, delay testing after coverage so students reconstruct from long-term memory, prefer free-recall/short-answer over recognition (MCQ), and delay feedback.
- Spacing (distributed practice) effect
- Distributing study with gaps beats massing the same total time ('cramming'). The Ebbinghaus-lineage explanation: gaps allow forgetting of irrelevant features while relevant features are reactivated, improving durable, generalisable memory. Practically: review at the end of a lesson and the start of the next, and use spaced-repetition tools.
- Deliberate practice
- Ericsson's construct: intentionally repeating an activity to improve performance, involving attention, rehearsal and repetition over time. It is necessary but not sufficient for expertise; benefits include long-term retrievability, automaticity that frees working memory, better transfer and further motivation. Often not inherently enjoyable, so teachers design for success and provide feedback.
- Feedback (Hattie & Timperley)
- Information from an agent (teacher, peer, book, self) about aspects of one's performance or understanding, whose purpose is to reduce the discrepancy between current understanding/performance and a desired goal. Effective feedback addresses where the learner is going, how they are going, and where to next.
- High-information feedback
- Feedback that tells students not just what they got wrong but why and how to avoid it next time, adding self-regulation information. In Wisniewski, Zierer & Hattie's meta-analysis it is by far the most powerful type (d ≈ 0.99), well above corrective feedback (d ≈ 0.46) and reinforcement/punishment (d ≈ 0.24). Overall feedback averages d ≈ 0.55.
- Medal and a mission
- Chappuis's task-focused feedback format (also 'stars and stairs'): name one strength (the medal) and one specific improvement (the mission), focused on the task and next steps rather than the person. Preferred over vague person-praise ('good job'), which can reduce persistence; encouragement (process-focused, informative) is preferred over person-praise.
Practice, Retrieval and Feedback FAQ
Why is rereading such a weak strategy?
Because it creates a fluency illusion - the material feels familiar, so students think they know it - without the effortful retrieval that actually strengthens memory. Dunlosky et al. rate highlighting and rereading as low-utility, while practice testing and distributed practice are high-utility. The testing effect shows that reconstructing knowledge from memory (even when it feels harder) produces far more durable learning than passively reviewing it again.
What makes feedback effective?
Its purpose is to close the gap between where a student is and where they need to be, so the most effective feedback is high-information: it explains what the error was, why it happened and how to fix it next time, and it is task-focused rather than person-focused. Wisniewski, Zierer & Hattie found high-information feedback (d ≈ 0.99) far outperforms bare corrective feedback and especially reinforcement or punishment. Formats like 'medal and a mission' operationalise this, and encouragement is preferred over vague person-praise.
How is Week 10 assessed?
Expect multiple-choice items on the testing effect, spacing vs massing, deliberate practice and the feedback effect-size ranking, plus short-answer questions asking you to design durable practice or rank/redesign feedback. The study-strategy utility tiers (Dunlosky et al.) are also examinable. Confirm coverage on Canvas.
Can AI help me with practice and feedback?
Yes. Sia can turn a cram plan into a spaced, retrieval-based one, quiz you with free-recall prompts (a form of the testing effect), and help you rewrite feedback into the high-information form, explaining each step. It mirrors how EDUF3040 teaches this material and does not do graded work for you; University of Sydney academic-integrity rules apply.
Exam move
Use this week's own findings to revise it: self-test rather than reread, and space your sessions - which also models the testing and spacing effects you must explain. Memorise the feedback effect-size ranking (high-information ≈ 0.99 ≫ corrective ≈ 0.46 ≫ reinforcement/punishment ≈ 0.24, overall ≈ 0.55) and the 'medal and a mission' format, since these are reliable marks. Keep the Dunlosky et al. utility tiers on a card (high: practice testing, distributed practice). Be ready to design durable practice for a scenario and to convert a weak feedback comment into a high-information one. When the effects or numbers slip, ask Sia to test you with fresh scenarios and check your designs.
Working through Practice, Retrieval and Feedback in EDUF3040? Sia is AskSia’s AI Educational Psychology tutor — ask any EDUF3040 Practice, Retrieval and Feedback question and get a clear, step-by-step explanation grounded in how EDUF3040 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
