Original research review: Assessing affective states in domestic dogs through judgment bias tasks: a systematic review of methods, influencing factors, and welfare applications
Authors: Valentina Gazzano, Luca Ciurli, Maria Claudia Curadi, Paolo Baragli, Angelo Gazzano, Asahi Ogi.
Journal: Frontiers in Veterinary Science, volume 13, article 1926433.
Published: October 23, 2026 (publisher PDF and publication history).
DOI: 10.3389/fvets.2026.1926433.
Read the full review
Download the complete original review (PDF). This free, 29-page download includes the authors’ methods, results, tables, discussion and references. The publisher PDF is unchanged.
Original abstract
Background: Judgment bias tasks (JBTs) have emerged as a promising approach for assessing affective states in animals, including domestic dogs, by measuring optimistic or pessimistic responses to ambiguous stimuli as proxies for underlying emotion. Despite growing use, the canine evidence base had not been systematically evaluated.
Methods: We conducted a systematic review following PRISMA 2020 guidelines, searching Scopus and PubMed, with a supplementary search of Web of Science Core Collection. Thirty-two studies (15 experimental, 13 observational/quasi-experimental, 4 methodological) met the inclusion criteria. Experimental studies were appraised with the SYRCLE risk-of-bias tool. Findings were synthesized narratively, as methodological heterogeneity precluded meta-analysis.
Results: Five paradigm types were identified, but 28 of the 32 studies (88%) used the same spatial discrimination task; the four remaining paradigms were represented by one study each. JBTs showed sensitivity to some experimental manipulations, including oxytocin (d = 2.16) and environmental enrichment (nosework, p < 0.01), while observational and quasi-experimental studies also reported associations between training method and JBT performance (e.g., aversive versus reward-based training: Z = −2.38, p = 0.017). However, validity and reliability concerns were pervasive: ceiling effects at ambiguous locations, confounding by spatial-learning ability, poor test–retest reliability, and absent protocol standardization. Convergence with independent welfare indicators was only partial.
Conclusion: JBTs are responsive to some affective manipulations and are associated with relevant welfare and experiential factors, but substantial validity and reliability limitations remain. Standardized protocols and multi-method validation are needed before JBTs can serve as field measures of canine welfare.
Source and reuse
© 2026 Gazzano, Ciurli, Curadi, Baragli, Gazzano and Ogi. Originally published in Frontiers in Veterinary Science 13:1926433. Distributed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). The original abstract is reproduced verbatim with web formatting only; the complete publisher PDF has not been modified. DogTraining.World hosts this resource and is not the review’s author. Hosting does not imply endorsement by the authors or publisher.
Original publisher article, publication history and any later updates.