Experts under cross-examination just last month in a trial which started 25/9/18. Even now, the science is not accepted as probable cause.
https://eu.jsonline.com/story/news/crime/2018/10/02/kelly-dwyer-case-cadaver-dog-handler-testifies-zocco-trial/1498006002/*snipped*
Defense attorney Craig Mastantuono, on cross-examination, raised questions about such conclusions. Corcoran admitted that Molly may have gotten some direction to some spots and that normal "human shed" such as hair, skin, nails and blood would be detectable by a cadaver dog and would likely be concentrated in trash areas of a large apartment building.
No actual remains, of Dwyer or anyone else, were found where Molly alerted.
-----------------------------------------------------------------------------------------------
During pretrial motions, Zocco's defense team moved to block Corcoran's testimony as unreliable and prejudicial and challenged her training methods with the dog. The defense called its own detection dog training expert out of order following Corcoran's testimony.
Corcoran said Milwaukee police had requested assistance from her department for Molly's expertise as a so-called cadaver-sniffing dog. She and Molly arrived during the evening of Oct. 16 and worked the scene into the next day.
She described starting in the garage before going to the 18th floor. In the hallway, she said, there were six apartment doors and a door to the trash chute. She let Molly sniff the hallway before Milwaukee officers directed her to the trash door. She opened it and the hatch to the door before Molly alerted the spot.
Corcoran also indicated that the door to Zocco's apartment was ajar, unlike the five other unit doors. After Molly alerted there, officers left while they obtained a search warrant to bring Molly inside.
When they returned, the dog alerted on the entryway floor, a pile of clothes in a laundry room, the hallway bathroom and Zocco's bed — by jumping on it and sitting. Molly did not alert in any other area of the two-bedroom apartment.
Corcoran agreed that a bedroom is another spot where human shed is often concentrated.
She also said Molly did not train extensively with residual odor — that is, detecting odors of remains no longer present. Normally, the training involved finding actual human remains. She agreed there is no way to measure a detection dog's error rate, so she couldn't know Molly's.
Defense expert questions dog's reactions
For the defense, Andre Falco Jimenez, a former law enforcement K-9 handler who now runs a California-based training and consulting company, said use of video during training has revealed how frequently a handler's subtle cues drive a dog's reactions.
He suspected that may have driven Molly's alerts in the Zocco case, since it appeared the dog did not alert on any smells on her initial cursory sniffs but only after some direction.
Jimenez told jurors a dog won't detect evidence of the first phase of human decomposition, which occurs in the first 72 hours, without training on that specific odor. He said it is difficult to obtain such recently deceased bodies for training in the United States, but easier in South America, where his company does much work.He acknowledged he has worked mostly as an expert witness for defense attorneys and that he advocates a type of training called double-blind, in which a detection dog and handler enter a place without any cues whatsoever, and where sometimes there is no target material hidden. He said he believes it results in more reliable performance than the more common training protocol that believes a dog should always end with a rewardable success.
Hill presented her own expert, Wendell Nope of the Utah Department of Public Safety, who has been involved with police K-9s for 41 years. He said he reviewed Corcoran's training records and reports and deemed her and K-9 Molly reliable and objective, noting she had more than 447 training sessions with Molly with a very high success rate.
Nope agreed it is not easy to get freshly deceased human tissue for training but said some trainers are not convinced that training shortcoming hinders effective detection. He said Molly's records showed she had successfully indicated on residual odor 13 times and that he had no concerns that Corcoran improperly used rewards to unconsciously influence Molly's behavior.
Published 5:27 PM EDT Oct 2, 2018
