Temporal consistency of behavior trait measurements in guide dogs
Bibliographic record
Abstract
Guide dog organizations have strict criteria to breed, raise, and select dogs to assist people with visual impairments. In collaboration with Dr. James Serpell, several guide dog training organizations developed a scoring tool called the Behavior Checklist (BCL) to evaluate candidate guide dogs. The tool's use has expanded to the entire assistance dog industry and is rapidly emerging as the standard behavior assessment. Since 2003, Guiding Eyes for the Blind (GEB) has used the BCL to measure individual dogs' behaviors up to 8 times between puppyhood and final placement. Here, we evaluate the consistency of the BCL over multiple evaluations in a population of 3,969 Labrador Retrievers raised by Guiding Eyes. We grouped BCL evaluations by two methods, factor analysis, and trainer-defined groups, and summarized groupings of behavior in two ways, using mean and lowest scores. We then determined the agreement between pairs of evaluations using kappa statistics and the predictive capacity of early BCL scores to predict later scores using positive and negative predictive values. Evaluations that are similar in nature and those that are scored within 3 to 6 months of one another agree the most. When a dog scores well early in life, they are likely to consistently score well and the dog's behavior is unlikely to regress over time. We also found that dogs who score poorly early in life either improve their scores on later evaluations with training intervention or are removed from training. One limitation of this data is that dogs who score poorly at early time points are often removed from training and the data from later BCL evaluations is biased toward higher-scoring dogs. Regardless, these data show that the BCL is an effective way to evaluate assistance dog behavior and has some predictive capacity.
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 | 0.003 | 0.011 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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.
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".