Prediction

Element · All elements
Evidence · 9 claims (5 for, 4 mixed) · 10 studies (5 causal, 2 review, 1 quant-synthesis, 1 associational, 1 qualitative), q2–q3 · 3 of 10 report an effect size · 7 claims rest on one study

Description

A prediction asks learners to commit to an answer, outcome, or explanation before receiving instruction — guessing the result of an experiment, the next step in a demonstration, or the answer to a question about material not yet taught. The act of committing, even incorrectly, prepares learners to encode the corrective information that follows.

Design Implications

Predictions leverage the pretesting effect: attempting to answer before instruction improves retention of the correct answer relative to studying without a prior attempt, even when the initial guess is wrong Pretesting enhances learning [+S]. Predictions work by activating relevant prior knowledge and creating a state of curiosity or disequilibrium that makes the eventual answer more memorable Advance organizers aided learning in a 1980 meta-analysis, but activating prior topic knowledge did not improve primary pupils' text comprehension in one experiment [~M]. They are most effective when followed promptly by feedback or instruction that resolves the uncertainty — a prediction left unresolved yields little benefit Pre-instruction testing enhances memory for directly tested video content, with benefits shown for videos from 5 to 30 minutes [+M].

Context

Requirements

• A question or scenario that is answerable with the upcoming instruction — the prediction must target content the lesson will actually address
• A genuine commitment from learners (written, clicked, or stated publicly), not a rhetorical question
• Timely resolution: instruction or feedback that confirms or corrects the prediction soon after it is made
• Low stakes — learners must feel safe guessing wrong, since errors are the mechanism, not a failure (Assessment for learning, not of it)

Constraints

• Predictions about content unrelated to subsequent instruction can divert attention and impair learning of the material actually presented Effects of pre-instruction testing on untested but studied materials are highly inconsistent across experiments and materials [~M] — the "prequestioning penalty" for incidental content
• If learners lack any relevant prior knowledge, guesses are random and the activation benefit shrinks [~M]
• High-stakes grading of predictions suppresses the willingness to commit, eliminating the effect
• Overly difficult predictions that produce frustration rather than curiosity can disengage learners [~W]

Target Learners

• Novices with some, but incomplete, prior knowledge — enough to generate a plausible guess Pretesting enhances learning [+S]
• Learners prone to illusions of knowing; a failed prediction exposes gaps that passive reading conceals
• Less effective for complete novices with no relevant schema to activate, or for advanced learners who already know the answer and gain no new uncertainty

Target Learning Goals

• Conceptual understanding: surfacing and revising intuitive misconceptions before formal explanation
• Procedural learning: predicting the next step in a Demonstration keeps learners actively processing rather than passively watching
• Scientific reasoning: hypothesis generation and evidence-based revision of expectations

Affordances

• Activation — prediction enacts this principle by forcing retrieval of relevant prior knowledge before new material arrives, preparing schemas to incorporate it
• Cognitive Disequilibrium — a committed-but-wrong prediction creates a knowledge gap that motivates and focuses the resolution that follows
• Active Learning — prediction converts passive reception (watching, reading) into a generative act, even within a lecture or video format
• Assessment for Learning — aggregated predictions give instructors a real-time map of misconceptions to address in the lesson

Claims

• A prediction graphing activity revealed preschool children's classification thinking and misconceptions, which class discussion and pictures corrected [+W]
• Pre-instruction testing enhances memory for directly tested video content, with benefits shown for videos from 5 to 30 minutes [+W]
• Classroom pretesting improves high-stakes exam performance on pretested content by 8–9% over a multi-week course [+W]
• Positive information prediction errors enhance memory for low-curiosity answers increasingly with age [~W]
• Partial resolution of curiosity through immediate feedback increases subsequent curiosity [+W]

The evidence by question

• Does attempting questions on material before studying it improve later memory of the questioned content, and under what conditions on the answer and the material? Pretesting enhances learning [+S]
• What does failing first cost the learner: persistence, belief in the method, the effect of confusion? Pre-tests reduced adult learners' persistence in a MOOC, though they improved post-test scores among those who completed it [~M]

Related Elements

• Activation — prediction is a specific, commitment-forcing form of prior-knowledge activation
• Demonstration — pausing a demonstration to ask "what happens next?" interleaves prediction with modeling
• Advance Organizers — both prepare learners for incoming content; prediction adds a verifiable commitment
• Case Studies — predictions about case outcomes create engagement before the resolution is revealed
• Check-In — low-stakes response formats that make committing to a guess safe

Patterns That Use This Element

• Case-Based Learning — learners predict outcomes before the case resolution is disclosed
• Flipped Classroom — pre-class prediction questions prime learners for in-class resolution
• Cognitively Guided Instruction (CGI for Math) — students estimate and predict solutions before formal strategies are taught

Examples

Peer Instruction (Eric Mazur, Harvard) — Students commit to a conceptual answer via clicker before and after peer discussion; the initial prediction exposes misconceptions that peer debate then resolves.

PhET Interactive Simulations — Many simulations prompt learners to predict circuit or gas behavior before running the model, then compare outcome to prediction.

Prequestioning in video learning — Embedding a multiple-choice prediction before each segment of an instructional video; research shows improved learning of the queried content Carpenter & Toftness, 2017.

Predict-Observe-Explain (POE) — A published science-teaching routine (White & Gunstone, 1992) in which learners predict a demonstration's outcome, observe it, and explain any discrepancy — directly targeting misconception revision.

Key Sources

• Kornell, N., Hays, M. J., & Bjork, R. A. (2009). Unsuccessful retrieval attempts enhance subsequent learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(4), 989–998. doi:10.1037/a0015729
• Richland, L. E., Kornell, N., & Kao, L. S. (2009). The pretesting effect: Do unsuccessful retrieval attempts enhance learning? Journal of Experimental Psychology: Applied, 15(3), 243–257. doi:10.1037/a0016496
• Carpenter, S. K., & Toftness, A. R. (2017). The effect of prequestions on learning from video presentations. Journal of Applied Research in Memory and Cognition, 6(1), 104-109. doi:10.1016/j.jarmac.2016.07.014
• White, R., & Gunstone, R. (1992). Probing understanding. London: Falmer Press.

Page metadata

Field	Value
type	element
id	prediction
title	Prediction
description	A prediction asks learners to commit to an answer or outcome before instruction, activating prior knowledge and creating a gap that subsequent teaching resolves.
status	review
canonical	True
generated	by claude/unspecified on 2026-08-29
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.