Flexible behavior or flexible methods? A cross-taxon review of experimental designs in reversal learning
Bibliographic record
Abstract
Behavioral flexibility—the ability to adapt behavior in response to changing conditions—is widely recognized as a key feature of animal cognition. It is often measured using reversal learning tasks, where individuals must inhibit a previously rewarded response and adopt a new one after contingencies shift. Despite its widespread use, the comparability of these tasks across species remains unclear. This paper establishes a foundation for resolving this issue by examining how reversal learning has been designed and implemented across taxa. We conducted a systematic review of 206 empirical studies (2014–2023) spanning eight major taxonomic groups: invertebrates, fishes, amphibians and reptiles, birds, rodents, non-human primates, humans, and other mammals. For each study, we extracted variables related to taxon coverage, sampling, learning and reversal criteria, cue types, and outcome measures. Analyses included nonparametric tests to assess group-level differences, linear discriminant analyses to explore multivariate structure, and model-based robustness checks. We identified three methodological obstacles to understanding reversal learning across diverse taxa. First, the distribution of research is highly imbalanced: birds, rodents, and humans accounted for most of the populations in the reviewed work. When considering species coverage, most animal diversity—especially invertebrates, fishes, and amphibians and reptiles—remains virtually untested, with less than 1% of described species included per taxon. Second, research is taxonomically fragmented: 99% of studies focus on a single group, limiting opportunities for direct comparison. Third, and most critically, methodological standards diverge dramatically across taxa. Humans are consistently held to the strictest learning criteria, while other taxa most often use lower thresholds. The number of reversal phases differ more than threefold among taxa. Nearly all studies of amphibians and reptiles, fishes, and invertebrates use single-reversal designs, whereas multi-reversal protocols are much more common in humans and non-human primates. Sample sizes, evaluation window lengths, cue types, and outcome metrics also display taxon-specific patterns. These systematic differences in experimental design introduce structural asymmetries that complicate cross-taxon comparisons, blurring the line between true cognitive variation and methodological artifacts. Although research to date has advanced our understanding, further progress will depend on greater methodological coordination and broader taxonomic coverage. Emerging large-scale collaborations are beginning to address these gaps, offering a promising path toward a more robust and equitable science of behavioral flexibility.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.144 | 0.283 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.006 | 0.006 |
| Bibliometrics | 0.007 | 0.008 |
| Science and technology studies | 0.001 | 0.005 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".