Analogical Encoding¶

Strategy · All strategies
Evidence · 3 claims (2 for, 1 mixed) · 11 studies (5 review, 4 causal, 1 quant-synthesis, 1 theoretical), q1–q4 · 1 of 11 report an effect size

Description¶

Analogical encoding asks learners to compare two or more surface-different cases that share the same underlying structure (e.g., two negotiation dilemmas both hinging on a trade-off between value creation and value claiming). Through comparison, learners extract the shared relational schema and encode it as the case's central meaning, making it more likely they will recognize that structure in a new, surface-dissimilar problem. It is carried out by presenting paired cases, prompting explicit comparison ("How are these two situations alike?"), and naming the abstract principle the comparison reveals.

Design Implications¶

Comparison is what drives the abstraction: a single case tends to be encoded in terms of its surface story, while two aligned cases force attention to shared relations Analogical Reasoning Improves Transfer [+S]. The strategy enacts Analogical Reasoning as a learning mechanism rather than only a presentation device. Because comparing two full cases adds working-memory load, cases should be presented side by side, aligned element-by-element, and kept short Chunking reduces working-memory load. [+S].

Context¶

Requirements¶

• Two (or more) cases with identical relational structure but different surface features
• Side-by-side presentation or interleaved presentation so both cases are active simultaneously
• Explicit comparison prompts that direct attention to relationships, not surface details
• A named, stated schema — learners rarely abstract the principle without it being articulated (Advance Organizers can supply the label)

Constraints¶

• If cases share surface features as well as structure, learners may rely on surface similarity and fail to abstract the schema [-S]
• Comparison without guidance often fails: unguided learners notice surface differences, not structural parallels, and derive no usable schema [-S]
• Adding a second case increases load for novices; with very high-element-interactivity material, sequential single-case study may outperform simultaneous comparison [~M]
• Less effective for learners with strong domain expertise, who already encode relationally and gain little from forced comparison Expertise reverses the benefit of added instructional support. [~M]

Implementation Variability¶

• Case-pair then transfer problem: compare two cases, then solve a third structurally similar one (the classic Gentner & Gentner / Gick & Holyoak sequence)
• Within-case comparison: contrast a good and flawed application of the same principle (Non-Examples style)
• Learner-generated analogies: learners supply their own parallel case, deepening processing but requiring more schema support
• Delayed comparison: cases studied separately, then compared — weaker than simultaneous comparison but easier to schedule

Target Learners¶

• Novices who default to surface-based encoding and need help seeing relational structure Analogical Reasoning Improves Transfer [+S]
• Learners in domains where problems look different on the surface but share structure (negotiation, physics, statistics, management)
• Less useful for experts, who already abstract structure and may find comparison redundant Expertise reverses the benefit of added instructional support. [~M]

Target Learning Goals¶

• Transfer: recognizing a learned principle in novel, surface-dissimilar situations
• Schema formation: abstracting general principles from concrete instances
• Conceptual discrimination: distinguishing structurally similar from merely superficially similar cases

Instructions¶

1. Select two cases with the same deep structure and maximally different surface stories.
2. Present them side by side, aligned so parallel elements are visually or textually matched (Clear Structure Presentation).
3. Prompt explicit comparison: "What do these two situations have in common? What is the same decision in each?"
4. State and label the shared schema explicitly; do not leave abstraction to chance (Advance Organizers).
5. Follow with a transfer problem that shares the structure but not the surface (Application of Knowledge), so learners must retrieve the schema from memory.
6. As expertise grows, fade the pairing and the prompts (Fading).

Related Strategies¶

• Case-Based Learning — analogical encoding specifies how cases should be processed: comparison, not sequential reading
• Activating Prior Knowledge — a familiar case can serve as the first analog in the pair
• Erroneous Examples — comparing a correct and a flawed case is analogical encoding applied to error analysis

Examples¶

• Gentner, Loewenstein & Thompson's negotiation training (Kellogg School) — MBA students compared two negotiation cases (e.g., a sister/brother silverware dispute and a union/management dispute) both structured around a trade-off; compared groups later spontaneously used the winning strategy on a new negotiation far more often than single-case groups.
• Gick & Holyoak's radiation problem studies — learners read the fortress story and a second analog, compared them, and were far more likely to solve the Duncker radiation problem; the comparison-plus-schema-statement condition produced the strongest transfer.
• Harvard Business School case method — when sections deliberately compare two cases sharing a decision structure, the discussion functions as analogical encoding; when cases are treated one at a time, it does not.

Key Sources¶

• Gentner, D., Loewenstein, J., & Thompson, L. (2003). Learning and transfer: A general role for analogical encoding. Journal of Educational Psychology, 95(2), 393–408. doi:10.1037/0022-0663.95.2.393
• Gick, M. L., & Holyoak, K. J. (1983). Schema induction and analogical transfer. Cognitive Psychology, 15(1), 1–38. doi:10.1016/0010-0285(83)90002-6

• Alfieri, L., Nokes-Malach, T. J., & Schunn, C. D. (2013). Learning through case comparisons: A meta-analytic review. Educational Psychologist, 48(2), 87–113. doi:10.1080/00461520.2013.775712