Olfactory Generalization in Detector Dogs

Detection dogs fighting transnational narcotraffic: performance and challenges under real customs scenario in Brazil

Narcotic Detection Dogs (NDDs) are essential tools in the fight against drug trafficking, acting with high precision and improving efficiency at border posts. When trained efficiently, these dogs can detect a great variety of compounds, such as cocaine, marijuana and its derivatives, and synthetic drugs, among others. Most of the knowledge on canine detection processes and efficiency has been determined in experimentally controlled conditions, but narcotic seizures detected by dogs in realistic anti-drug operations have not yet been critically determined in a Country with continental dimensions such as Brazil. This study aimed to evaluate the data set concerning the performance, operations, efficiency, and success rate of NDDs used by the Brazilian Customs Authority (Aduana) in the fight against drug trafficking. Narcotic seizure rates increased in luggage and packages detected by NDDs working at border crossings from 2010 to 2020, with an estimated value of over US$ 2 billion in losses to the cocaine drug trafficking business. NDD units also increased most narcotic groups seized in the same period. The number of NDDs and anti-drug operations, and Customs Border Post (CBP) influenced the rates of drugs seized. NDDs provided an increase of 3,157 kg/animal of drugs seized for every new dog introduced into the inspection systems.

Dogs' ability to detect an inflammatory immune response in cattle via olfaction

Introduction

Canine olfaction is a potential means for detection of respiratory disease in beef cattle. In a prior study, two dogs were trained to discriminate between nasal swabs from healthy cattle and cattle that developed Bovine Respiratory Disease. Dogs had some ability to identify samples from BRD-affected cattle, but results were ambiguous. The purpose of this study was to evaluate more dogs using better-controlled training and testing procedures.

Methods

Nasal and saliva swabs were collected from 96 cattle before and after administering a vaccine to induce an inflammatory immune response. Samples were stored at -80°C for up to 11 months before use, and samples from animals with an elevated body temperature at baseline were omitted. An automated olfactometer apparatus was constructed to improve blinding procedures and reduce opportunities for odor contamination. Four dogs were trained to distinguish between swabs from healthy and sickness-model cattle, including the two dogs from the previous study ("Runnels" and "Cheaps") and two inexperienced dogs ("Molokai" and "Amy"). During a seven-month training period, dogs were exposed to samples from 28 animals. Dogs were tested on 59 sets of unfamiliar samples.

Results

Performance varied among dogs (χ2 = 10.48, p = 0.02). Molokai's performance was above chance (0.73 ± 0.06, p = 0.0006), while Amy (0.44 ± 0.06, p = 0.43), Cheaps (0.53 ± 0.07, p = 0.79), and Runnels (0.56 ± 0.06, p = 0.43) did not respond correctly at a rate different from chance. Accuracy did not differ between nasal swabs (0.63 ± 0.08) and saliva swabs (0.53 ± 0.08, χ2 = 0.81, p = 0.37).

Discussion

The results of this study indicate that canine olfaction may be an effective means of detecting illness in beef cattle. However, individual dogs' aptitude for this detection task varies.

Investigating individual learning behaviour of dogs during a yes/no detection task

Detection dogs are frequently tested for their ability to detect a variety of targets. It is crucial to comprehend the processes for odour learning and the consequences of training on an expanding set of target scents on performance. To properly evaluate their ability to identify the target, the only true measure is the dogs' initial response to novel sources, since this excludes learning effects. In this study, we evaluated the individual learning processes of three detection dogs that were pre-trained to differentially respond to a faecal sample of a mare in oestrus (S+) and a faecal sample of the same mare in di-oestrus (S-). After reaching criterion during a test with known training samples, the dogs were tested for generalization to a novel source. Average responses to S+ and S- were calculated as a function of presentation sequence, and Signal Detection Theory was used to further analyse characteristic differences in learning. The results of this study suggest that the ability of individual scent detection dogs to learn within an olfactory discrimination test varies considerably. The information obtained in this study could be helpful for mitigation training. We show that through careful monitoring of individual learning processes, the strategy each dog followed becomes apparent: especially the observations on the dogs' responses to first encounters with novel sample sources. This provides us with more detailed information than the more traditional sensitivity and specificity measures and allows us to better predict the dog's capabilities.

Drug-detecting bioelectronic nose based on odor cue memory combined with a brain computer interface

The international drug situation is increasingly, various new drugs are hidden in public places through changing forms and packaging, which brings new challenges to drug enforcement. This study proposes a drug-detecting bioelectronic nose based on odor cue memory combined with brain-computer interface and optogenetic regulation technologies. First, the rats were trained to generate positive memories of drug odors through food reward training, and multichannel microelectrodes were implanted into the DG region of the hippocampus for responsible memory retrieval, the spike signals of individual neurons and the local field potential signals of population neurons in the brain region were collected for pattern recognition and analysis. Preliminary experimental results have shown that when low-dose drugs are buried in a hidden area, rats can find the location of the drugs in a very short time, and when close to the relevant area, there is a significant change in the energy value and time-frequency spectrum signal coupling of the returned data, which can be extracted to indicate that the rats have found the drugs. Second, we labled the neuronal activity marker c-fos and revealed more robust activation in the DG region following odor detection. We modulated these neurons through neuroregulatory technology, so that the rats could recognize drugs by retrieving memories more quickly. We conceive that the drug-detecting rat robot can detect trace amounts of various drugs in complex terrain and multiple scenes, which is of great significance for anti-drug work in the future.

Scent detection dogs as a novel method for oestrus detection in an endangered species, the Tasmanian devil (Sarcophilus harrisii)

Captive breeding is a critical tool for conservation of endangered species. Identifying the correct time to pair males and females can be a major challenge for captive breeding programmes, with current methods often being invasive or slow. Detection dogs may provide a non-invasive way to determine female receptivity, but this has not been explored in captive wildlife. This exploratory study investigated the use of detection dogs as a novel method of oestrus detection in the endangered Tasmanian devil (Sarcophilus harrisii). Faecal samples were collected from 11 captive female devils during the breeding seasons of 2020 and 2021. Three dogs with prior detection experience were trained and subsequently assessed (n = 188 searches per dog), on their ability to discriminate between oestrus and non-oestrus devil faecal samples, in a one sample set-up. When assessed on training samples, dogs were able to correctly discriminate oestrus from non-oestrus with a mean sensitivity of 69.1% and mean specificity of 65.7%. When assessed on novel samples, their sensitivity to oestrus dropped (mean sensitivity of 48.6%). However, they were still able to correctly identify non-oestrus samples (mean specificity of 68.1%). This study is the first to explore detection dogs' ability to identify oestrus in a captive breeding programme for endangered wildlife, providing a promising tool for non-invasive monitoring of reproductive status in wildlife.

Canine olfactory detection of SARS-CoV-2-infected humans-a systematic review

PURPOSE

To complement conventional testing methods for severe acute respiratory syndrome coronavirus type 2 infections, dogs' olfactory capability for true real-time detection has been investigated worldwide. Diseases produce specific scents in affected individuals via volatile organic compounds. This systematic review evaluates the current evidence for canine olfaction as a reliable coronavirus disease 2019 screening tool.

METHODS

Two independent study quality assessment tools were used: the QUADAS-2 tool for the evaluation of laboratory tests' diagnostic accuracy, designed for systematic reviews, and a general evaluation tool for canine detection studies, adapted to medical detection. Various study design, sample, dog, and olfactory training features were considered as potential confounding factors.

RESULTS

Twenty-seven studies from 15 countries were evaluated. Respectively, four and six studies had a low risk of bias and high quality: the four QUADAS-2 nonbiased studies resulted in ranges of 81%-97% sensitivity and 91%-100% specificity. The six high-quality studies, according to the general evaluation system, revealed ranges of 82%-97% sensitivity and 83%-100% specificity. The other studies contained high bias risks and applicability and/or quality concerns.

CONCLUSIONS

Standardization and certification procedures as used for canine explosives detection are needed for medical detection dogs for the optimal and structured usage of their undoubtful potential.

A role for GnRH in olfaction and cognition: Implications for veterinary medicine

Pulsatile secretion of gonadotropin-releasing hormone (GnRH) is essential for the activation and maintenance of the function of the hypothalamic-pituitary-gonadal (HPG) axis, which controls the onset of puberty and fertility. Two provocative recent studies suggest that, in addition to control reproduction, the neurons in the brain that produce GnRH are also involved in the control postnatal brain maturation, odour discrimination and adult cognition. Long-acting GnRH antagonists and agonists are commonly used to control fertility and behaviour in veterinary medicine, primarily in males. This review puts into perspective the potential risks of these androgen deprivation therapies and immunization on olfactory and cognitive performances and well-aging in domestic animals, including pets. We will also discuss the results reporting beneficial effects of pharmacological interventions restoring physiological GnRH levels on olfactory and cognitive alterations in preclinical models of Alzheimer's disease, which shares many pathophysiological and behavioural hallmarks with canine cognitive dysfunction. These novel findings raise the intriguing possibility that pulsatile GnRH therapy holds therapeutic potential for the management of this behavioural syndrome affecting older dogs.

Is diagnostic performance of SARS-CoV-2 detection dogs reduced -due to virus variation- over the time?

Medical detection dogs have a high potential for use as alternative diagnostic tools not only for organic diseases, but also for infectious diseases. However, new variants emerging over time may affect the accuracy and sensitivity of diagnostic methods including medical detection dogs in case of viral pandemics. To the best of our knowledge, this is a pioneer study aimed to investigate diagnostic performances and generalization ability of SARS-CoV-2 detection dogs against the new variant after being trained with the original virus. Two SARS-CoV-2 detection dogs were used in this study. In total, 1002 samples including the Omicron variant were introduced to the dogs using a double-blinded design. Two different refresher training sessions were conducted to train the dogs to identify the scent of the Omicron variant. In the first refreshment training, mixed samples (original virus and Omicron variant) were used. The diagnostic performances of the dogs were significantly increased only after the second refreshment training where only the Omicron variant was introduced. This study illustrates that diagnostic performances of SARS-CoV-2 detection dogs were not consistent over time with the emerging new variants. Thus, refreshment training with new variant(s) should be conducted with every new variant which may affect the diagnostic performances of those dogs in such infectious outbreaks.

Use of a habituation-dishabituation test to determine canine olfactory sensitivity

The habituation-dishabituation (H-D) paradigm is an established measure of sensory perception in animals. However, it has rarely been applied to canine olfaction. It proposes that animals will lose interest in, or habituate to, a stimulus after successive exposures but will regain interest in, or dishabituate to, a novel stimulus if they can perceive it. This study assessed an H-D test's practicability to determine dogs' olfactory detection thresholds (ODTs) for a neutral odorant. A random selection of mixed-breed pet dogs (n = 26) participated in two H-D tests in a repeated-measures crossover design. They were first habituated to a carrier odor and then presented with either ascending concentrations of n-amyl acetate in the known ODT range (experimental condition) or repeated carrier odor presentations (control condition). No single odor concentration elicited dishabituation in the majority of the dogs. However, individual dogs dishabituated at differing experimental concentrations significantly more often than in the control condition (p = .012). These findings provide some tentative support for using this method in studying canine olfaction. However, further assessment and refinement are needed before it can be a viable alternative to traditional ODT measurement.

Vapor Signatures of Double-Base Smokeless Powders and Gunshot Residues for Supporting Canine Odor Imprinting

Non-intrusive means to detect concealed firearms based on magnetometry are widely accepted and employed worldwide. Explosive detection canines can also detect concealed firearms provided that they are imprinted on materials that may be related to firearms such as nitroglycerin in double-base smokeless powders. However, there are hundreds of possible smokeless powder formulations across various manufacturers, presenting a challenge for trained canines to generalize across all possible powder compositions. In response, this paper reports a set of potential imprinting vapor(s) that may help detection canines generalize across a variety of double-base smokeless powders and gunshot residues. Statistical analysis was conducted on the smokeless powder database maintained by the National Center for Forensic Science, and headspace measurements targeting nitroglycerin and diphenylamine were collected from several powders. In addition, measurements were taken to track nitroglycerin and diphenylamine vapor concentration changes over time on the spent casings and gun barrels of four types of ammunition. The observed vapor concentration mixing ratios for nitroglycerin and diphenylamine from residues were in the part-per-billion to part-per-trillion range, which would be challenging to detect for many field-deployed explosive vapor detectors and indicate continued importance of canines for forensic investigation and crime prevention. Analyses suggest four potential vapor compositions for imprinting. For unburnt powders, 90% nitroglycerin and 10% diphenylamine appear adequate for most powders, and 90% dinitrotoluene and 10% diphenylamine is a possible candidate to increase generalization to powders that contain dinitrotoluene instead of nitroglycerin. 100% nitroglycerin appears adequate for many gunshot residues (GSRs). Diphenylamine may be present in some GSRs, and equal compositions of nitroglycerin and diphenylamine may be adequate for imprinting against these residues as they age (this study tracked signatures up to 7 weeks after discharge).

Using Canine Olfaction to Detect Bovine Respiratory Disease: A Pilot Study

Bovine respiratory disease (BRD) is the leading cause of morbidity and mortality in feedlot cattle and is a major welfare and economic concern. Identification of BRD-affected cattle using clinical illness scores is problematic, and speed and cost constraints limit the feasibility of many diagnostic approaches. Dogs can rapidly identify humans and animals affected by a variety of diseases based on scent. Canines' olfactory systems can distinguish between patterns of volatile organic compounds produced by diseased and healthy tissue. In this pilot study, two dogs (“Runnels” and “Cheaps”) were trained for 7 months to discriminate between nasal swabs from cattle that developed signs of BRD within 20 days of feedlot arrival and swabs from cattle that did not develop BRD signs within 3 months at the feedlot. Nasal swabs were collected during cattle processing upon arrival to the feedlot and were stored at −80°C. Dogs were presented with sets of one positive and two negative samples and were trained using positive reinforcement to hold their noses over the positive sample. The dogs performed moderately well in the final stage of training, with accuracy for Runnels of 0.817 and Cheaps of 0.647, both greater than the 0.333 expected by chance. During a double-blind detection test, dogs evaluated 123 unique and unfamiliar samples that were presented as 41 sets (3 samples per set), with both the dog handler and data recorder blinded to the positive sample location. Each dog was tested twice on each set of samples. Detection test accuracy was slightly better than chance for Cheaps at 0.451 (95% CI: 0.344–0.559) and was no better than chance for Runnels at 0.390 (95% CI: 0.285–0.496. Overall accuracy was 0.421 (95% CI: 0.345–0.496). When dogs' consensus response on each sample set was considered, accuracy was 0.537 (95% CI: 0.384–0.689). Detection accuracy also varied by sample lot. While dogs showed some ability to discriminate between BRD-affected and healthy cattle using nasal swabs, the complexity of this task suggests that more testing is needed before determining whether dogs could be effective as a screening method for BRD.

Evaluation of canine detection of covid-19 infected individuals under controlled settings

Reverse transcription polymerase chain reaction (RT‐PCR) is currently the standard diagnostic method to detect symptomatic and asymptomatic individuals infected with Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). However, RT‐PCR results are not immediate and may falsely be negative before an infected individual sheds viral particles in the upper airways where swabs are collected. Infected individuals emit volatile organic compounds in their breath and sweat that are detectable by trained dogs. Here, we evaluate the diagnostic accuracy of dog detection against SARS‐CoV‐2 infection. Fifteen dogs previously trained at two centres in Australia were presented to axillary sweat specimens collected from known SARS‐CoV‐2 human cases (n = 100) and non‐cases (n = 414). The true infection status of the cases and non‐cases were confirmed based on RT‐PCR results as well as clinical presentation. Across dogs, the overall diagnostic sensitivity (DSe) was 95.3% (95%CI: 93.1–97.6%) and diagnostic specificity (DSp) was 97.1% (95%CI: 90.7–100.0%). The DSp decreased significantly when non‐case specimens were collected over 1 min rather than 20 min (p value = .004). The location of evaluation did not impact the detection performances. The accuracy of detection varied across dogs and experienced dogs revealed a marginally better DSp (p value = .016). The potential and limitations of this alternative detection tool are discussed.

Discrimination of SARS-CoV-2 Infections From Other Viral Respiratory Infections by Scent Detection Dogs

Background: Testing of possibly infected individuals remains cornerstone of containing the spread of SARS-CoV-2. Detection dogs could contribute to mass screening. Previous research demonstrated canines' ability to detect SARS-CoV-2-infections but has not investigated if dogs can differentiate between COVID-19 and other virus infections. Methods: Twelve dogs were trained to detect SARS-CoV-2 positive samples. Three test scenarios were performed to evaluate their ability to discriminate SARS-CoV-2-infections from viral infections of a different aetiology. Naso- and oropharyngeal swab samples from individuals and samples from cell culture both infected with one of 15 viruses that may cause COVID-19-like symptoms were presented as distractors in a randomised, double-blind study. Dogs were either trained with SARS-CoV-2 positive saliva samples (test scenario I and II) or with supernatant from cell cultures (test scenario III). Results: When using swab samples from individuals infected with viruses other than SARS-CoV-2 as distractors (test scenario I), dogs detected swab samples from SARS-CoV-2-infected individuals with a mean diagnostic sensitivity of 73.8% (95% CI: 66.0-81.7%) and a specificity of 95.1% (95% CI: 92.6-97.7%). In test scenario II and III cell culture supernatant from cells infected with SARS-CoV-2, cells infected with other coronaviruses and non-infected cells were presented. Dogs achieved mean diagnostic sensitivities of 61.2% (95% CI: 50.7-71.6%, test scenario II) and 75.8% (95% CI: 53.0-98.5%, test scenario III), respectively. The diagnostic specificities were 90.9% (95% CI: 87.3-94.6%, test scenario II) and 90.2% (95% CI: 81.1-99.4%, test scenario III), respectively. Conclusion: In all three test scenarios the mean specificities were above 90% which indicates that dogs can distinguish SARS-CoV-2-infections from other viral infections. However, compared to earlier studies our scent dogs achieved lower diagnostic sensitivities. To deploy COVID-19 detection dogs as a reliable screening method it is therefore mandatory to include a variety of samples from different viral respiratory tract infections in dog training to ensure a successful discrimination process.

Highly sensitive scent-detection of COVID-19 patients in vivo by trained dogs

Timely and accurate diagnostics are essential to fight the COVID-19 pandemic, but no test satisfies both conditions. Dogs can scent-identify the unique odors of volatile organic compounds generated during infection by interrogating specimens or, ideally, the body of a patient. After training 6 dogs to detect SARS-CoV-2 by scent in human respiratory secretions (in vitro diagnosis), we retrained 5 of them to search and find the infection by scenting the patient directly (in vivo screening). Then, efficacy trials were designed to compare the diagnostic performance of the dogs against that of the rRT-PCR in 848 human subjects: 269 hospitalized patients (COVID-19 prevalence 30.1%), 259 hospital staff (prevalence 2.7%), and 320 government employees (prevalence 1.25%). The limit of detection in vitro was lower than 10-12 copies ssRNA/mL. During in vivo efficacy experiments, our 5 dogs detected 92 COVID-19 positive patients among the 848 study subjects. The alert (lying down) was immediate, with 95.2% accuracy and high sensitivity (95.9%; 95% C.I. 93.6-97.4), specificity (95.1%; 94.4-95.8), positive predictive value (69.7%; 65.9-73.2), and negative predictive value (99.5%; 99.2-99.7) in relation to rRT-PCR. Seventy-five days after finishing in vivo efficacy experiments, a real-life study (in vivo effectiveness) was executed among the riders of the Metro System of Medellin, deploying the human-canine teams without previous training or announcement. Three dogs were used to examine the scent of 550 volunteers who agreed to participate, both in test with canines and in rRT-PCR testing. Negative predictive value remained at 99.0% (95% C.I. 98.3-99.4), but positive predictive value dropped to 28.2% (95% C.I. 21.1-36.7). Canine scent-detection in vivo is a highly accurate screening test for COVID-19, and it detects more than 99% of infected individuals independent of key variables, such as disease prevalence, time post-exposure, or presence of symptoms. Additional training is required to teach the dogs to ignore odoriferous contamination under real-life conditions.

The "olfactory fingerprint": can diagnostics be improved by combining canine and digital noses?

A sniffer (detecting) dog is conventionally defined as an animal trained to use its olfactory perceptions for detecting a vast array of substances, mostly volatile organic compounds (VOCs), including those exceptionally or exclusively generated in humans bearing specific pathologies. Such an extraordinary sniffing performance translates into the capability of detecting compounds close to the femtomolar level, with performance comparable to that of current mass spectrometry-based laboratory applications. Not only can dogs accurately detect "abnormal volatilomes" reflecting something wrong happening to their owners, but they can also perceive visual, vocal and behavioral signals, which altogether would contribute to raise their alertness. Although it seems reasonable to conclude that sniffer dogs could never be considered absolutely "diagnostic" for a given disorder, several lines of evidence attest that they may serve as efficient screening aids for many pathological conditions affecting their human companions. Favorable results have been obtained in trials on cancers, diabetes, seizures, narcolepsy and migraine, whilst interesting evidence is also emerging on the capability of early and accurately identifying patients with infectious diseases. This would lead the way to proposing an "olfactory fingerprint" loop, where evidence that dogs can identify the presence of human pathologies provides implicit proof of the existence of disease-specific volatilomes, which can be studied for developing laboratory techniques. Contextually, the evidence that specific pathologies are associated with abnormal VOC generation may serve as reliable basis for training dogs to detect these compounds, even (or especially) in patients at an asymptomatic phase.

Case Study: An Evaluation of Detection Dog Generalization to a Large Quantity of an Unknown Explosive in the Field

Simple Summary

This case study is a series of experiments to investigate a real-life event where two highly trained and certified detection dogs did not find an explosive in a suspicious bag. We tested seven dog teams from the agency in three experiments and confirmed that dogs were able to detect the agency’s training sample in a small quantity (30 g) but not the large amount of the confiscated explosive (13 kg) found in a similar scenario. To further evaluate a possible generalization deficit, we tested dogs with a 30 g subsample of the confiscated explosive, and most of the dogs were able to detect it (but with some decrement) even though they largely failed to detect 13 kg of the same material. Finally, we trained dogs to detect the 30 g subsample until reaching proficiency and found that after training with the small sample, dogs still showed poor generalization to the large-quantity sample until explicit training with the large sample was conducted. Altogether, this series of studies shows the importance of explicitly training for larger quantity finds and has led to changes in agency practices to mitigate future risks.

Abstract

Two explosive detection dogs were deployed to search a suspicious bag, and failed to detect 13 kg of explosive within. The aim of this research was to further evaluate this incident. First, dog teams (N = 7) searched four bags in a similar scenario. One bag contained the same 13 kg of explosive, two bags were blanks, and the other contained the training sample that the agency routinely used for training. All dogs detected the training sample, but most (5/7) did not alert to the 13 kg sample. Subsequently, dogs received two trials in a line up with a 30 g subsample of the explosive to evaluate whether they could generalize to a smaller quantity. Most dogs (6/7) alerted to the subsample at least once. Finally, dogs were trained with the 30 g subsample and later tested with the 13 kg sample. Only three dogs spontaneously generalized to the large sample after training with the small subsample. Dogs’ alert rate to the 13 kg sample was improved with training in subsequent trials with the 13 kg sample. This result indicates that explosive detection dogs may not generalize to a target odor at a significantly higher quantity relative to the one used in training, highlighting the importance of conducting such training.

Diving in Nose First: The Influence of Unfamiliar Search Scale and Environmental Context on the Search Performance of Volunteer Conservation Detection Dog-Handler Teams

Simple Summary

Conservation detection dogs (CDDs) are trained to locate biological material from plants and animals of interest to conservation efforts and are often more effective and economical than other detection methods. However, the financial costs of developing and appropriately caring for CDDs can make them inaccessible for smaller conservation organizations. Training skilled volunteers to work with suitable pet dogs may help increase accessibility. We sought to further develop the skills of 13 volunteer dog–handler teams that were trained in a previous study to detect myrrh essential oil in controlled laboratory conditions. We recorded the proportion of targets found, false alerts made and search duration of the dog–handler team group through progressive training stages outdoors that increased in size and environmental complexity. First, teams searched various-sized areas before and after 12 weeks of search training on a sports-field. Next, teams searched various-sized areas before and after seven weeks of training in bushland. Overall, teams found approximately 20% fewer targets in each unfamiliar context, compared to performance in familiar contexts. However, teams typically found 10–20% more targets after a period of training compared to baseline searches. Search performance varied between teams, yet six teams found at least 78% of targets after training in bushland. Our results help to validate our stepped approach to training and highlight the need to train volunteer CDD teams to work in various-sized areas and environments.

Abstract

Conservation detection dogs (CDDs) are trained to locate biological material from plants and animals of interest to conservation efforts and are often more effective and economical than other detection methods. However, the financial costs of developing and appropriately caring for CDDs can nonetheless prohibit their use, particularly by smaller conservation organizations. Training skilled volunteers to work with suitable pet dogs may help address this constraint. We sought to further develop the skills of 13 volunteer dog–handler teams that were trained in a previous study to detect myrrh essential oil in controlled laboratory conditions. We assessed search sensitivity, search effort, search precision and false-alert instances through progressive training stages increasing in size and environmental complexity. First, teams searched various-sized areas before and after 12 weeks of search training on a sports-field. Next, teams searched various-sized areas before and after seven weeks of training in bushland. Overall, search sensitivity decreased by approximately 20% in each unfamiliar context, compared to performance in familiar contexts. However, sensitivity typically improved from baseline performance by 10–20% after a period of training. Six teams found at least 78% of targets after training in bushland, yet sensitivity ranged from 29% to 86% between teams. We maintain that the foundational skills developed previously were necessary to prepare volunteer teams for field surveys involving conservation related targets. However, our results highlight the need to also train volunteer CDD teams in search scale and environmental contexts similar to their intended working conditions.

Biosecurity Dogs Detect Live Insects after Training with Odor-Proxy Training Aids: Scent Extract and Dead Specimens

Detector dogs could be trained to find invasive insect pests at borders before they establish in new areas. However, without access to the live insects themselves, odor training aids are needed to condition dogs to their scent. This proof-of-concept study assessed 2 potential training aids for insect detection: a scent extract and dead specimens of the target species. Using Musgraveia sulciventris (Hemiptera: Tessaratomidae) as an experimental model, gas chromatography-mass spectrometry (GC-MS) analyses were carried out to compare the chemical headspaces that make up the odors of live specimens and these 2 training aids. This was then followed by canine scent-detection testing to investigate biosecurity detector dogs' (n = 4) responses to training in an ecologically valid context. Both the scent extract and the dead specimens shared the majority of their volatile organic compounds (VOCs) with live insects. Of the dogs trained with scent extract (n = 2), both were able to detect the live insects accurately, and of those trained with dead specimens (n = 2), one detected the live insects accurately. These findings lend support for these training aids as odor-proxies for live insects-particularly scent extract, which is a relatively novel product with the potential for broad application to facilitate and improve insect-detection training.

Training with varying odor concentrations: implications for odor detection thresholds in canines

Detection dogs are required to detect trace quantities of substances, many times in the parts per billion or parts per trillion concentration range. Frequently, detection of trace quantities is not explicitly trained but rather assumed when dogs show proficiency at higher concentrations to which they are trained. The aim of this study was to evaluate the effect of the odor concentration of the training sample on the minimum concentration dogs will subsequently detect. We expected that dogs may not spontaneously generalize to trace odor concentration when trained with higher concentrations, but when trained to a range of lower concentrations, dogs will show superior detection to lower untrained concentrations. A total of 11 dogs were randomly assigned to 2 groups and were trained to alert to isoamyl acetate at 0.01% odor dilution (v/v with mineral oil) using a 3-alternative forced choice test. Once reaching proficiency, odor detection threshold was assessed using a 2-down 1-up descending staircase procedure. Next, experimental dogs received training with systematically lower concentrations of isoamyl acetate and threshold re-assessed. Control dogs were yoked to experimental dogs in terms of training time, but only received training to the 0.01% dilution between threshold assessments. Experimental dogs showed significantly improved detection thresholds, outperforming control dogs by detecting an average dilution about 100-fold lower. Results suggest that explicitly training for lower concentrations is critical for generalization for trace odor detection.

Training with Multiple Structurally Related Odorants Fails to Improve Generalization of Ammonium Nitrate Detection in Domesticated Dogs (Canis familiaris)

Simple Summary

Domestic dogs are used by military and police forces to detect improvised explosive devices (IEDs) and other explosives. A challenge with training explosive detection dogs is that the ingredients used by someone to make an IED can vary. It is therefore critical that dogs be able to detect an IED with unfamiliar ingredients. This ability can be improved if the dog’s training allows them to categorize similar odors together. Many IEDs are created using ammonium nitrate, which was the focus of our study. Based on preliminary odor training performance, we equally assigned dogs to two experimental groups. Dogs in the first group were trained with two odors related to ammonium nitrate, while dogs in the second group were trained to six related odors. We anticipated that dogs trained to six odors would be more likely to form a category. However, this was not the case since dogs in both experimental groups were unable to form a category that allowed them to identify a novel ammonium nitrate mixture. Based on our results, the use of authentic explosive materials likely remains the most cost-effective and efficient way to train explosive scent detection dogs.

Abstract

A critical aspect of canine scent detection involves the animal’s ability to respond to odors based on prior odor training. In the current study, dogs (n = 12) were initially trained on an olfactory simple discrimination task using vanillin as the target odorant. Based on their performance on this task, dogs were assigned to experimental groups. Dogs in group 1 and 2 (n = 5 dogs/group; 1 dog/group were removed due to low motivation or high error rates) were trained with either two or six forms of ammonium nitrate (AN), respectively. Dogs were then assessed with a mock explosive with AN and powdered aluminum. Dogs in both groups failed to respond to the novel AN-aluminum odor. Mean success rates were 56 ± 5 and 54 ± 4% for groups 1 and 2, respectively. Overall, and individual dog performance was not statistically higher than chance indicating that dogs did not generalize from AN to a similar AN-based odorant at reliable levels desired for explosive detection dogs. These results suggest the use of authentic explosive materials, without the added complication of including category-learning methods, likely remains a cost-effective and efficient way to train explosive scent detection dogs.

A Review of the Types of Training Aids Used for Canine Detection Training

The canine detection community is a diverse one, ranging from scientific fields such as behavior, genetics, veterinary medicine, chemistry, and biology to applications in law enforcement, military, medicine, and agricultural/environmental detection. This diversity has allowed for a flourishing and innovative community, yet it has also led to little acceptance and agreement on terminology. This is especially true when discussing the variety of training aids used in olfactory-based exercises. In general, authentic materials and pseudo-scents are the most commonly discussed, with the former accepted widely for training and certification, and the latter more often disregarded. However, as advances are made in the creation of training materials, alternative training aids are being introduced that do not fit into either of these categories. The misconceptions surrounding how these alternative training aids are manufactured has led to confusion on their classification, and therefore their reliance as an effective tool. This manuscript will review the existing language surrounding canine training aids, address relevant research revealing effectiveness, and clarify the different types based on their manufacture, chemical nature, and fundamental function.

Generalization and Discrimination of Molecularly Similar Odorants in Detection Canines and the Influence of Training

Operationally-deployed canine detectors are often trained on one or a limited number of materials representing a single target odor, and training frequently occurs using materials of a high purity grade in controlled scenarios with minimal other background odors. Conversely, in the field, canine detectors are expected to generalize and identify variations of the target odor, while discriminating from similar extraneous or background odors. This exemplifies the balance between generalization and discrimination required for effective canine detectors. This research explored the tendency for detection canines to generalize or discriminate between similar odorants. Two groups of related odorants were used in two separate studies; (1) odorants of similar functional groups with differing carbon chains, and (2) odorants of similar carbon chain length but differing functional groups. Within each odorant set, the effect of training was addressed by incrementally increasing the number of odorants each canine was trained to detect. Initially, discrimination increased with increasing molecular dissimilarity in both odorant groups. After further training on additional related odorants, generalization increased across the set of odorants of the same carbon chain length, but there were no significant changes in either generalization or discrimination across the set of odorants of the same functional group. The results suggest that the canines in this study were more likely to generalize across compounds of the same chain length with differing functional group than across compounds of the same functional group, but differing chain lengths. Furthermore, some variation in performance between individual canines indicated that the tendency to generalize differed with experience, breed, and other factors affecting olfaction.

A jump to the left and a step to the right: A test of two accounts of peak shift

“Peak Shift” usually occurs following intradimensional-discrimination training and involves a shift of the peak of the generalization gradient away from the original discriminative stimulus (S+) in a direction away from an S-. Two theoretical accounts of peak shift, the gradient interaction theory (GIT) and adaptation level theory (ALT), were compared. The effects of asymmetric test stimuli and the impact of instructions to participants for them to treat stimuli as members of categories on generalization gradients were investigated. In Experiment 1, the relation between peak shifts obtained when an extended asymmetric set of test stimuli was employed and the occurrence of categorization of the stimuli involved was investigated in four separate conditions. Two involved temporal discrimination, one involved line-angle discrimination, and one involved a compound line-angle and temporal cue discrimination. If participants treated the stimuli as belonging to discrete categories, such as hands-on a clock, rather than as being on continuous dimensions then responding to the compound cue was expected to result in attenuation of blocking of a peak shift. However, the peak shift obtained to the three cue types were the same. In Experiment 2, an independent group of participants was given explicit instructions to treat the line angles as if they were the hands of a clock face and this eliminated peak shift. The results from the present experiments support an ALT interpretation, although the peak shifts were significantly smaller in magnitude than predicted by this account.