MétaCan
Menu
Back to cohort

Prediction of successful training outcomes for drug detection dogs using subjective ratings and behavioral test measures: A case study in Japan customs

2025· article· en· W4406062098 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

aboutThe title or abstract carries a Canadian signal from the geographic lexicon.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueApplied Animal Behaviour Science · 2025
Typearticle
Languageen
FieldBiochemistry, Genetics and Molecular Biology
TopicHuman-Animal Interaction Studies
Canadian institutionsnot available
FundersJapan Society for the Promotion of Science
KeywordsTest (biology)PsychologyAnimal-assisted therapyHUBzeroClinical psychologyTraining (meteorology)Applied psychologyPet therapyAnimal welfareEcologyBiologyGeography

Abstract

fetched live from OpenAlex

Drug detection dogs, primarily employed by customs and police forces, play a crucial role in preventing the spread of illegal drugs worldwide. To minimize training costs, accurately predicting which dogs will succeed in scent detection training is essential. Local training organizations seek validated behavioral assessment methods for this purpose, but the wide range of methods used and the lack of scientific verification pose challenges. Previous research on detection dogs in Japan focused on genetics, but behavioral assessment methods for training have been understudied. To bridge the gap, the current study aimed to outline and evaluate the predictive validity of behavioral assessment systems used for drug detection dogs at Japan Customs. We compared the relative effectiveness of two different behavioral assessment methods: subjective ratings by chief trainers and behavioral measures in a novel test situation. For subjective ratings, we used subscales of Training Focus (i.e., interest in play, independence, concentration, activity, and boldness) and Tolerance (i.e., friendliness to humans and tolerance to dogs) to characterize a dog’s personality. For behavioral tests, a simple behavioral test measured a dog’s approach behavior and reactivity to an unfamiliar person. Data from 196 dogs (159 Labrador Retrievers and 37 German Shepherds) showed high inter-rater agreement for both methods. A GLMM model revealed that Training Focus subscale scores significantly predicted training success of candidate dogs. On the other hand, Tolerance scores and behavioral test measures were poor predictors for scent detection work. Dog breed and sex did not significantly influence final training outcomes. Receiver Operating Characteristic (ROC) curves indicated that Training Focus scales' classification performance for training success is comparable to or better than previous reports for assistance and detection dogs. These findings demonstrate the predictive validity of subjective Training Focus ratings, aiding in the selection of drug detection dogs at Japan Customs. While generalizability to other detection dog populations and identification of alternative behavioral predictors remains uncertain, this study provides valuable insights into the predictive accuracy of trainer ratings in a dog behavior assessment system. • Trainer’s ratings of Training Focus strongly predicted the success of detection dogs. • Trainer’s ratings of Tolerance did not predict the success of detection dogs. • Behavioral test measures did not predict the success of drug detection dogs. • Dog breed and sex did not significantly influence final training outcomes.

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.264
Threshold uncertainty score0.643

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.055
GPT teacher head0.376
Teacher spread0.321 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it