Application of Analytical Chemistry to Test the Accuracy of Human Residual Odor Detection by Cadaver Dogs

Human remains detection (HRD) canines, or “cadaver dogs,” can detect volatile organic compounds (VOCs) of the odor of their target and are often deployed to aid in the search for deceased bodies. No studies have definitively linked human remains odor with canine olfactory capabilities yet courts accept dog handlers as expert witnesses. Further, claims have been made that HRD dogs can also detect odors on objects when the body is no longer present, called residual odors. The purpose of this study is to integrate a new double-blind, standardized residual odor recognition test protocol for HRD canines with the characterization of the residual odor VOCs using analytical chemistry. Samples of 12 cleaned gauze pads containing residual odor from a deceased donor (target, T) and live human scent from a living person (distractor, D) will be collected on the same day and placed in separate jars, as will twelve untreated gauze (blanks, B). This process will be repeated two more times to obtain 36 samples of each type (T,D&B). Thirty-six HRD teams will be selected. Twelve teams will be scheduled per day and will perform three replicates of the assessment. On each day the teams will be challenged with blanks, the same three live human distractors and the same residual odor targets to allow for direct comparison across all canines. In each trial, there will be 18 containers in three rows, and the T, B and D samples will be randomly placed in one container of each row. Neither the handlers nor the room attendants will know the locations. Video cameras and microphones will capture the canine behaviors and be analyzed via nested logistic and multinomial regression. Chemical analyses will be performed on the day of canine testing to ensure the integrity of the VOC mixtures. Traditional headspace sampling and headspace-solid phase microextraction (HS-SPME) will be used to generate samples for gas chromatography—high resolution mass spectrometry (GC-HRMS) analysis. A pilot test has demonstrated proof of concept for the canine assessment. We test a number of hypotheses, including whether canines can detect residual odor samples more frequently than chance and whether there is a significant relationship between VOC measurements and canine ability to distinguish between odors from living and deceased individuals. Deliverables include a training workshop to disseminate the results directly to the participating handlers and a standardized test that can be expanded to include other forensically important odors. CA/NCF