Discovering which chemicals might be involved in the manufacture of mobile phones can be a surprisingly difficult task. According to Compound Interest (2014), “there’s actually remarkably little easily accessible information out there that details the specific compounds used” and even less information regarding any associated odours.
That said, there seems little doubt that mobile phones “smell” – our dogs can find them – but what odour, or odours, are they actually latching on to? What produces the odour that is significant to our dogs? Possibilities include chemicals associated with the phone’s memory chip, touchscreen and rechargeable lithium ion battery and probably a lot else besides.
Triphenylphosphine oxide – a flame retardant used to protect the memory chips in electronic devices including memory sticks, hard drives, SD cards and mobile phones.
Lithium Niobate – used in surface acoustic wave (SAW) devices. These convert electric signals to acoustic waves.
Lithium Ion Battery
Dimethyl Carbonate – produces an alcohol-like odour.
Diethyl Carbonate – has a nail-polish-like odour.
Lithium Niobate – used in electrodes of the lithium ion battery.
But how do we know whether our dogs are using any of these chemicals to find mobile phones?
A few months ago, one of my client’s reported that her dog, without any specific training, was able to locate hidden memory sticks. The client had discovered this ‘natural talent’ a few weeks previously when she’d asked her dog to find another, already trained, odour source. Instead, her dog found a memory stick that, earlier in the day, had been hidden for another dog!
So, was this evidence that her dog was searching for memory stick odour – perhaps one, or more, of the chemicals listed previously – or was there something else going on?
Given that her dog had received no previous Scent Detecting training with memory sticks – there had been no familiarisation process – and that memory stick odour is unlikely to be an Unconditioned Stimulus (UCS) for any dog (as far as I’m aware, no dog is naturally attracted to memory stick odour), it seems highly unlikely that her dog was locating memory sticks because of their very specific, chemical, profile and associated odour.
If you’d like to know a little bit more about the familiarisation process and unconditioned stimuli, take a trip over to one of my previous Blog posts, Smoothing out the Lumpy Bits; Why it might be wise to teach the indication before the search and all that back chaining stuff.
So, what was going on?
The simplest explanation is usually the correct one
During the 12th Century, a Franciscan friar – William of Ockham – argued that “the simplest answer – that is, the answer that requires the fewest assumptions – is generally the correct one” (Salter, ?date). “Ockham’s Razor”, as it has become known, or the “Law of Parsimony”, is the view that “of any given set of explanations for an event occurring, the simplest one is most likely the correct one … [it] is a vital tool in rigorous thought. By reducing the number of unsupported assumptions in an explanation, you reduce the likelihood of being wrong” (Salter, ?date).
So, rather than my client’s dog demonstrating some sort of ‘natural’, or ‘super-natural’, Scent Detecting ability, might there be a simpler, more straightforward, explanation? An explanation that fits better with the current knowledge of odour, olfaction and learning?
Rather than being an example of ‘natural talent’ at work, the simplest explanation for this dog’s apparent ability to locate memory sticks (with no previous training) is provided by … contamination. In all likelihood, the memory stick had been contaminated with an odour that the dog had previously been trained to detect alongside human scent and, if they were used, the odour of gloves. Yes, those gloves you’ve been told to wear to avoid contaminating your scent source are a potential contaminant themselves (Learning Center, 2016) … an odour that your dog learns to detect!
As Gadbois (2016) has pointed out, our dogs olfactory abilities are so good that no amount of cleaning or careful handling of the scent source or use of gloves will reduce contamination sufficiently for our dogs not to detect it. So, throw away your gloves and tweezers and work with contamination rather than against it!
Mobile phone odour vs A N Other odour
In addition to some of the possible chemical sources of mobile phone odour (as listed previously), the overall “smell” of a phone is likely to comprise of a number of additional odours – plastics from the casing and human scent from handling the device. With the possible exception of a mobile phone straight from the factory, human scent is very much part of the mobile phone odour picture. It may be a contaminant but it’s also part of the mobile phone cocktail of odours.
So, here are your first two, of many, potential difficulties. Is your mobile phone Scent Detecting dog using human scent or plastic odour to detect the phone rather than any of the previously listed chemicals? If so, just as my client didn’t have a memory stick Scent Detecting dog, you don’t have a mobile phone Scent Detecting dog.
Be Systematic – Rule things out!
After familiarising my dogs to the odour of mobile phones – I have a huge collection of old, well-used, phones that I use for this purpose – I then begin to make sure that they’re searching for the cocktail that is the mobile phone odour as opposed to human scent, or plastics, more specifically.
As well as hiding mobile phones, I hide other, similar sized, items that I, and other people, have handled thoroughly. What I want to see is my dog either checking these items out, or ignoring them, before moving on to locate and indicate on the mobile phone – the target odour.
Rather than relying on a change of behaviour (COB) to tell me when my dog has located the scent source, I train a rock-solid passive indication, with duration. This gives me the confidence to let my dog check things out, dismiss things and move on to the correct location before letting me know that they know they’ve found the correct scent source – the mobile phone. My dog has shown me that he’s differentiated between human odour (more generally) – mine and other peoples – and human odour (more specifically) – mine and other peoples – associated with the mobile phone.
If you’d like to know more about the power of the passive indication, follow this link to one of my previous Blog posts – The Indication. It’s not an optional extra!
If you take a look at your mobile phone you’ll notice just how much plastic is used in its construction. Again, in order to confirm that my dog is using the whole cocktail of mobile phone odour to find its location rather than focusing on one odour in particular – plastic – I hide a variety of other, similar sized, plastic items – Spectacles. Plugs. Spectacle cases. Pencil cases. Backs off mobile phones – alongside the mobile phone. I want my dog to demonstrate to me that he can differentiate between plastic items (more generally) and plastic items (more specifically) that are part of the mobile phone odour cocktail.
Interestingly, last week, BB demonstrated very clearly that he was searching for mobile phone odour as opposed to any other associated odours. The mobile phone had been thrown into a wooded area and, as it landed, part of its plastic casing fell off and came to rest a few feet away from the remaining phone and its contents. BB located the area, checked out the separated piece of plastic casing before dismissing it and moving on to indicate, correctly, on the main part of the phone. Embrace these moments of happenstance!
Elemental vs Configural Odour Cues
As Thomas-Danguin et al (2014) point out, “smelling monomolecular odors hardly ever occurs in everyday life, and the daily functioning of the sense of smell relies primarily on the processing of complex mixtures of volatiles that are present in the environment”. The mobile phone is just one example of a scent source that is comprised of a complex mix of odours.
Some Scent Detecting trainers attempt to address this issue by taking an elemental approach. That is, they identify each odour molecule within a mixture of odours and train their Scent Detecting dog on each of these separate odours. According to an article by Chammah (2016), this is the approach taken by the California Department of Corrections and Rehabilitation when training their mobile phone detecting dogs. They work on each element of the phone – its battery, its SIM card – before training on the whole phone.
But here’s a question for you. Take a look at this picture …
Do you smell roast chicken (A) or broccoli (B) or beans (C) or cauliflower cheese (D) or the whole roast dinner (U)?
Interestingly, but not surprisingly perhaps, “an odor mixture can carry, beside the elements, another stimulus that is unique to the combination of those elements … usually noted U (unique cue)” (Thomas-Danguin et al, 2014). This mixed odour – U – “can stimulate cortical neurons that are not stimulated by their individual component odorants. This complex combinatorial coding system is consistent with the fact that often a mixture of odorants gives rise to novel perceptual qualities that are not present in each component” (Thomas-Danguin et al, 2014).
And this is why I train my dogs on the whole mobile phone rather than dissecting it into individual olfactory parts. I want my dogs to find mobile phones rather than mobile phone batteries or SIM cards or circuit boards or touch screens. If I want my dogs to find these individual parts, I’ll train them to find these individual parts – a completely separate exercise. As Hall et al (2018) state, “mixtures tend to be perceived configurally, such that the mixture produces a unique percept distinct from the constituent elements, and this may vary depending on the chemical similarities of the components in the mixture”.
Interestingly (and importantly), in a study by Lazarowski et al (2014), “most dogs trained with pure PC [potassium chlorate] did not correctly signal the presence of one or more PC-based explosive mixtures, despite having demonstrated the ability to reliably detect a trained amount of pure PC”. According to Lazarowski et al (2014), possible explanations for a dog’s inability to detect a previously trained substance when presented to him in a mixture include the way in which two or more chemicals might interact with one another making them unrecognisable to the dog. This is referred to as mixture suppression.
In contrast, the ability of a dog to successfully identify individual components in a complex mix is referred to as figure-background segregation (Hall et al, 2018). If, for example, I only wanted my dogs to detect a specific component of a mobile phone or other electronic device – the memory chip – then I would start by familiarising my dogs to that odour alone. During later training I would secrete the memory chips with other substrates – the whole phone, perhaps, or other materials – and continue with figure-background segregation training. This will ensure that my dogs can detect memory chips in a variety of environments and under a number of different conditions. I’m training them to detect memory chips within a roast chicken dinner! This is generalisation in action.
Another generalisation issue in scent detecting involves the concept of categorisation. According to Wright et al (2017), “categorisation is the ability to treat comparable but non-identical stimuli as equivalent by responding to them according to the category to which they belong. Thus, the ability to identify a novel stimulus as a member of a known category allows the organism to respond to it in an appropriate way”.
For example, in Cablk et als (2008) study, dogs were able to detect tortoises of all sizes, age and sex, not only those that they’d encountered during training. And this is where my very large collection of old mobile phones comes into its own – different makes, models, styles, sizes, previous owners and colours. My dogs are able to generalise and categorise. Their initial training is conducted using a variety of mobile phones and then, during subsequent searches, they’re able to successfully detect mobile phones they’ve never encountered before.
In contrast, our dogs are also able to discriminate between “individuals” within a target class. If you really wanted to, it is feasible to train your dog to only detect a certain type of phone. According to Pinc et al (2011), specially trained German Shepherds were able to distinguish the individual scents of identical twins who lived in the same environment and ate the same food!
Decide what it is you want your dog to detect – an individual within a target class or a target class more generally – a certain type of mobile phone or all mobile phones. Train your dog according to this decision. For instance, if you want your dog to detect a specific type of mobile phone then reward successful detection of that specific type of phone amongst other mobile phones.
You might just be surprised by what your dog can achieve!
Different Search Environments
Training your dog in different environments would seem to be a given. In terms of generalisation, it’s something that we’re always advised to do – from a straightforward Sit or Down to, arguably, some more complex Scent Detecting challenges. But with Scent Detecting it’s far more than simply a question of whether or not your dog can perform in different types of environment.
Placing a mobile phone in an outdoor area is likely to act as an attractant to your dog – a big red flag in the environment drawing your dog towards it. I’ve seen many a novice mobile phone detecting dog being trained in outdoor environments. They appear highly successful. They seem (like my client’s dog) to possess some sort of ‘natural talent’. As Ockham would point out, the more likely explanation for their apparent success is that they’re attracted to the unusual in an environment. A mobile phone in woodland is unusual – but so is a rubber duck or pair of spectacles. Your dog may not be detecting what you think he’s detecting!
Environmental factors need to be considered when training your mobile phone detecting dog and that means … start your training indoors.
and in conclusion …
Short and sweet … train systematically! It’s highly unlikely that your dog has any natural mobile phone detecting talents but he does have a super-natural ability to learn. Don’t waste it.
As with all of my blogs, I include a reference list. This allows you to investigate the topic a little further, check out the sources of my information and decide for yourself whether my interpretations of the literature represent an accurate reflection of the author’s work. Happy reading.
© Lesley McAllister – Scent : Detect : Find Ltd 2020
References / Further Reading
- Cablk ME, Sagebrel JC, Heaton JS and Valentin C (2008) Olfaction-based Detection Distance: A Quantitative Analysis of How Far Away Dogs Recognize Tortoise Odor and Follow It to Source. Sensors. 8. 2208-2222
- Chammah M (2016) The Amazing iPhone-Sniffing Prison Dogs. Who knew cell phones had a smell? https://www.themarshallproject.org/2016/06/07/the-amazing-iphone-sniffing-prison-dogs Accessed: 25.08.2020
- Compound Interest (2014) The Chemical Elements of a Smartphone. https://www.compoundchem.com/2014/02/19/the-chemical-elements-of-a-smartphone/. Accessed: 20.8.2020
- Gadbois S (2016) Modern Dog Training and Science: Are We Doing it Right? Common Misconceptions and Pitfalls. The Science of Sniffer Dogs. Oxford: Two day Conference
- Hall NJ and Wynne CDL (2018) Odor mixture training enhances dogs’ olfactory detection of Home-Made Explosive precursors. Heliyon 4. 4. 12. December
- Lazarowski L and Dorman DC (2014) Explosives detection by military working dogs: olfactory generalization from components to mixtures. Applied Animal Behaviour Science. 151. 84-93
- Learning Centre Valutek Blog (2016) Odor and Color Determine the Quality of Your Cleanroom Gloves. https://blog.valutek.com/home/odor-color-determine-the-quality-of-your-cleanroom-gloves# Accessed: 27.8.2020
- Pinc L, Bartos L, Reslova A, Kotrba R (2011) Dogs Discriminate Identical Twins. PLoS ONE. 6(6)
- Salter M (? date) Examples of Occam’s Razor. https://examples.yourdictionary.com/examples-of-occam-s-razor.html Accessed: 24.8.2020
- Thomas-Danguin T, Sinding C, Romagny S, Mountassir FE, Atanasova B, Le Berre E, Le Bon AM, Coureaud G (2014) The perception of odor objects in everyday life: a review on the processing of odor mixtures. Frontiers in Psychology. June. 5. 504
- Wright HF, Wilkinson A, Croxton RS, Graham DK, Harding RC, Hodkinson HL, Keep B, Cracknell NR, Zulch HE (2017) Animals can assign novel odours to a known category. Scientific Reports. 21 August. 1-6