Strengthening operational performance in canine detection teams with double-blind certification testing.
Double-blind scoring exposes hidden cueing and can flip pass rates by twenty points.
01Research in Context
What this study did
The researchers swapped the usual single-blind dog-certification test for a double-blind one.
Same dogs, same handlers, same search area—only the evaluator now did not know where the target scent was hidden.
They then counted how many teams still passed.
What they found
Pass rates fell off a cliff. Under single-blind rules, 94–100% of teams passed.
Once nobody knew the answer, only 72–88% passed. The drop shows handlers had been reading evaluator cues.
How this fits with other research
Bellon-Harn et al. (2020) saw the same mess when they compared three quality-rating tools: scores for the same FCT studies did not line up. Both papers warn that the way you check something can change the result.
Willis-Moore et al. (2024) found that just changing task order skewed people’s delay-discounting choices. Quigley-McBride et al. now show that evaluator knowledge can skew dog choices—same breed of hidden influence.
Williams et al. (2023) looked at live detection dogs in ports but focused on public reactions, not test accuracy. Together the trio says: if you care about true performance, control the test setting first.
Why it matters
Your behavior-chain is only as clean as your measurement. Swap any staff-evaluation probe to double-blind this week—hide the correct answer from both therapist and observer. You will spot false mastery fast and stop reinforcing cue-dependent responding.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Put the answer key in a sealed envelope and have a blind observer score the next skill probe.
02At a glance
03Original abstract
The present study evaluated the performance of professional canine–handler teams on narcotics detection certification trials conducted under single-blind and double-blind conditions. Across six years of annual testing (2012–2017), we analyzed 667 first-attempt trials and 132 second-attempt trials from 133 distinct canine-handler teams. Teams demonstrated high accuracy under single-blind conditions (94% pass rate for vehicle searches and 100% for luggage searches), but performance dropped substantially under double-blind conditions (72% pass rate for vehicle searches and 88% for luggage searches), where neither handlers nor evaluators knew the number or location of the target odors. Many teams that failed an initial double-blind trial passed on a second attempt, suggesting that at least some observed deficits in performance may be easily remedied with additional practice participating in double-blind trials. A follow-up survey of 20 handlers indicated generally positive perceptions of double-blind testing—although double-blind trials are more difficult, handlers believe that these types of trials increase their confidence, improve training strategies, and more closely reflect real-world scenarios. Incorporating routine double-blind exercises into certification and maintenance training may provide agencies with a reliable means of preparing teams for unpredictable real-world scenarios. Thus, double-blind testing represents a straightforward, cost-efficient strategy for enhancing the accuracy, credibility, and overall integrity of canine detection.
Forensic Science International: Synergy, 2026 · doi:10.1016/j.fsisyn.2026.100658