Analogical Alignment for Teaching Feedback Loops

Bibliographic Data

ID22433105
AuthorsAlexandra E Davatzes (Temple University), Thomas F Shipley (0000-0001-8397-2486, Temple University), Thorne Shipley (Temple University), Kim A Kastens (0000-0001-8292-5098, Lamont-Doherty Earth Observatory)
Year2026
Volume16
Issue7
Pages1141
Publication date2026-07-17
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueEducation Sciences (JOURNAL)
Journal identifiersISSN: 2227-7102 • E-ISSN: 2227-7102
PublisherMDPI AG (PUBLISHER • IT)
DOI10.3390/educsci16071141
OpenAlexW3108109286
LanguageEN
References cited48

Feedback loops are fundamental to understanding complex systems across the sciences, yet undergraduate students consistently struggle to correctly identify and distinguish positive and negative feedback loops. Traditional instructional approaches, which typically provide a definition and a single example, often fail to produce robust understanding. We tested an alternative pedagogical approach grounded in cognitive science research on analogical reasoning: mutual alignment of analogies. In this approach, students compare two feedback loop examples drawn from different domains (e.g., Earth science and economics) and identify their shared structural features, thereby extracting the abstract schema of a feedback loop. We compared student performance across two consecutive offerings of an introductory geoscience course at a large urban university. In the first year (Year 1, n = 71), feedback loops were taught using a traditional definition-plus-example approach. In the second year (Year 2, n = 100), a mutual alignment activity was introduced. Students in the mutual alignment condition outperformed those in the traditional instruction condition at all three time points following instruction, with a large effect size on the immediate assessment (Cohen’s d = 0.86) that persisted at three months (Cohen’s d = 0.51). We also identified two factors that may impede feedback loop understanding: the perceived desirability of a loop’s outcome and the readability of the loop description.

Computer Science · Experimental Learning in Engineering · Numerical Methods and Algorithms

Tools

Open DOIOpen Access