Invention

Element · All elements
Evidence · 8 claims (2 for, 6 mixed) · 11 studies (5 causal, 3 theoretical, 1 quant-synthesis, 1 review, 1 design), q2–q4 · 4 of 11 report an effect size · 7 claims rest on one study

Description

An invention task presents learners with a novel problem and asks them to create their own solution procedure, formula, or representation — typically before any direct instruction on the canonical method. The goal is not to produce the correct answer but to surface gaps in prior knowledge and build readiness for the instruction that follows. This is the core move of "inventing to prepare for future learning" Invention tasks before instruction improve later learning from that instruction. [+M].

Design Implications

Invention tasks work by activating relevant prior knowledge and making learners aware of the deep features of a problem — the reason a canonical method exists — rather than just its surface procedure Invention tasks before instruction improve later learning from that instruction. [+M]. Learners should be given contrasting cases that make the critical features visible, and the task must be followed by explicit instruction that connects their invented methods to the canonical one. The value lies in the preparation, not the invention itself: learners who invent poorly but then receive instruction often outperform those who received instruction alone Struggling before instruction improves subsequent learning. [+S].

Context

Requirements

• A problem that is solvable in principle with learners' existing (partial) knowledge, but for which the canonical method is genuinely better
• Contrasting cases or multiple data sets that highlight the features the canonical method addresses
• A follow-up instructional phase that explicitly compares invented solutions to the expert method (Demonstration, Direct Instruction)
• Sufficient time and low-stakes framing, since learners are expected to struggle

Constraints

• Invention without subsequent instruction produces little durable learning — the task alone does not teach the canonical method Struggling before instruction improves subsequent learning. [~S]
• Ineffective when learners lack the prerequisite knowledge to engage meaningfully, producing frustration rather than preparation Struggling before instruction improves subsequent learning. [-M]
• Time-intensive; less efficient than direct instruction for simple procedural goals where no deep structure needs to be discovered
• Can entrench misconceptions if the follow-up instruction does not explicitly confront and resolve learners' invented methods

Target Learners

• Learners with moderate prior knowledge who can partially engage with the problem Invention tasks before instruction improve later learning from that instruction. [+M]
• Novices benefit from the contrasting-case structure more than from open invention; pure unguided invention suits them poorly Struggling before instruction improves subsequent learning. [~M]
• Less effective for complete novices with no relevant knowledge to activate

Target Learning Goals

• Conceptual understanding: grasping why a method or representation works, not just how
• Transfer and adaptive expertise: preparing learners to apply knowledge to novel problems Invention tasks before instruction improve later learning from that instruction. [+M]
• Knowledge integration: connecting informal intuitions to formal representations

Affordances

• Activation — invention forces retrieval and use of prior knowledge, revealing what learners do and don't have before instruction builds on it
• Cognitive Disequilibrium — the experience of one's own method failing creates the awareness of a gap that instruction then fills
• Comparing Cases — contrasting cases are the standard vehicle for invention tasks, making deep features discriminable
• Productive Failure — invention is the exploration phase of the productive failure sequence, followed by consolidation instruction

Claims

• Productive failure, problem-solving before instruction, shows a measured benefit of about d = 0.36 rising to about 0.58 at high design fidelity (as reported) [+W]
• Contrary to expectation, the worked-example group applied suboptimal criteria more often, and fewer suboptimal applications mediated inventing's transfer [~W]
• Solution quality correlates with conceptual knowledge and gap awareness only for individual problem solvers [~W]
• PS-I sequence outperforms I-PS on conceptual knowledge only when instruction builds on student solutions [~W]
• In PS-I, prompting students to generate multiple strategies during the initial phase generally outperformed inventing one strategy or I-PS [~W]
• Number of canonical components in students' PF solutions correlates with conceptual knowledge, but the range of distinct components does not [~W]
• The increase in students inventing their own ways of solving tasks was minor, but there was a marked increase in students agreeing the CCS tasks required inventing their own ideas [~W]

Related Elements

• Demonstration — the canonical instruction that must follow invention; the pairing is what produces the effect
• Comparing Cases — the contrasting-case structure that makes invention tasks work
• Activation — the cognitive function invention serves before instruction
• Problem-Solving — invention is a special case of problem-solving positioned before instruction rather than after

Patterns That Use This Element

• Productive Failure — invention is the core exploration phase
• 5E Learning Cycle — the "explore" phase before "explain"
• Cognitively Guided Instruction (CGI for Math) — students invent strategies for word problems before formal algorithms are taught

Examples

Invention Tasks (Schwartz & Martin) — Learners devise a "reliability index" for bowling averages or density measures from contrasting data sets before being taught the canonical formula; the invention phase measurably improves learning from the subsequent lecture.

Productive Failure lessons (Kapur) — Singapore mathematics classrooms where students attempt to solve complex problems in groups before the teacher consolidates with canonical methods.

Physics by Inquiry (McDermott) — University physics curriculum in which students invent operational definitions and measurement procedures before formal physics concepts are introduced.

Key Sources

• Schwartz, D. L., & Martin, T. (2004). Inventing to prepare for future learning: The hidden efficiency of encouraging original student production in statistics instruction. Cognition and Instruction, 22(2), 129–184. doi:10.1207/s1532690xci2202_1
• Kapur, M. (2014). Productive failure in learning math. Cognitive Science, 38(5), 1008–1022. doi:10.1111/cogs.12107
• Schwartz, D. L., Chase, C. C., Oppezzo, M. A., & Chin, D. B. (2011). Practicing versus inventing with contrasting cases: The effects of telling first on learning and transfer. Journal of Educational Psychology, 103(4), 759–775. doi:10.1037/a0025140
• Kapur, M. (2008). Productive failure. Cognition and Instruction, 26(3), 379–424. doi:10.1080/07370000802212669

Page metadata

Field	Value
type	element
id	invention
title	Invention
description	An invention task asks learners to devise their own solution method or representation for a problem before receiving canonical instruction, preparing them to learn more deeply from subsequent teaching.
status	review
canonical	True
generated	by claude/unspecified on 2026-08-30
trust	unverified — no one has confirmed this page's content

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