Smoothing out the Lumpy Bits; Why it might be wise to teach the indication before the search and all that back chaining stuff

At its most basic, Scent Detecting requires your dog to search for, and locate, the source of an odour. On this simple description, your dog is engaged in Scent Detecting activities (entirely independent of you) for a very large portion of his day; finding a dropped crumb of food on the floor, discovering where a passing rabbit has left some tasty deposits, examining your clothes when you return home at night.

A little bit of independent Scent Detecting.

As Horowitz et al (2014) point out, your dog’s Umwelt or “Self-World” is primarily olfactory; “Dog noses house hundreds of millions more olfactory cells than humans’ do, and their corresponding brain regions are much more developed relative to their visual areas than in humans“.

So, what differentiates your dog’s independent Scent Detecting activities from those that involve you, his handler?

One word … Training

Well, two actually … Careful training

Ok, three … Careful, thorough, training!

Careful, thorough, training

Given your dog’s natural Scent Detecting talents, it can be hugely tempting to simply let him ‘get on with it‘. Spend a few minutes browsing any social media platform and you’ll soon find plenty of examples of this laissez-faire approach to Scent Detecting training, often with highly questionable results.

Yes, olfaction is your dog’s speciality but he isn’t Lassie! He can’t read your mind, guess what you want or speak your language. If you want him to find a specific odour for you – truffles perhaps – then you need to find some way of harnessing his natural abilities … some way of communicating your particular scent detecting requirements to your dog that might just differ from his own! And this is where careful, thorough, training comes into play.

Splitting vs Lumping

I start by splitting my Scent Detecting training into three separate, but interconnected, phases;

  1. Familiarising my dog to the target odour.
  2. Training the passive indication.
  3. Searching.

By splitting I am referring to the process of dividing the whole training task – in this case, Scent Detecting – into smaller, more manageable, training elements as opposed to lumping all three elements together and training them as a whole (Berg, 2018. Zerubavel, 1996).

According to Zerubavel (1996), “although the world in which we live is essentially continuous, we experience it as discrete chunks: strangers and acquaintances, fiction and nonfiction … Carving out of reality such ‘islands of meaning’ involves two contrasting yet complementary cognitive acts – lumping and splitting. The former entails grouping ‘similar’ things together in a single mental cluster. The latter involves perceiving ‘different’ clusters as separate from one another“.

In short, splitting makes our world, and our training, more manageable.

Why Split the training?

Well … the simple answer is … I like my Scent Detecting training to progress as smoothly as possible. I don’t like lumps!

I don’t like lumps in my Scent Detecting training!

If, as seems common practice, I were to lump my scent detecting training together and set my dog off to search an area before he’d been familiarised to the target odour and had a trained and reliable indication, how would I manage the following possible outcomes?

  1. He seemed to search the area very well but he didn’t find the target scent source.
  2. He definitely found the target scent but he pulled it out of the box, ran around the room with it in his mouth and chewed it up.
  3. He didn’t move from my side.

How do I let my dog know which bits he did well and were exactly what I could have hoped for and which bits were (to my view) just plain wrong? How can I reward his thorough searching in outcome 1 even though he didn’t actually locate his target scent? How can I reward my dog for locating his target scent in outcome 2 when he then went on to chew it up? And what on earth do I do about outcome 3?

Tricky, isn’t it? …. and this is what lumpy “training” looks like. Messy. Confused. Frustrating. Directionless. Unreliable. If our dogs learn anything from this lumpy approach to training, it would seem to be despite us rather than because of us. In contrast, splitting allows us to work on one specific training element at a time, helping to ensure competence in that area before moving on to another training element and then, eventually, putting all the elements together in one Scent Detecting whole or Lump.

But Splitting the training is so time-consuming, isn’t it?

Well, NO … it isn’t.

Splitting your training into smaller elements allows you to simplify things for your dog, aids clear communication, reduces frustration and increases your chances of getting things right in the first place without the need to add in further, remedial (Sticking Plaster), training at some later point in time.

Splitting your training results in …

  1. easily understood and manageable steps toward a larger end-goal.
  2. tiny bits of behaviour that, when successfully achieved, will allow you to reward your dog immediately.
  3. tiny bits of behaviour that, when things don’t go so well, will allow you to identify and address the problem rapidly with a revised training plan and approach.
  4. no need for your dog to “guess” what is required of him → less frustration → increased likelihood of training success.
  5. no wasted opportunity to reward a well executed behaviour because it’s been lumped together with a whole host of other, poorly executed, behaviours.

When you Split … always keep the Lumps in mind!

Although I’m a big advocate of splitting, it’s worth keeping in mind that there will always be connections between the different phases of training and the overall end goal. In other words, never lose sight of the training lumps! For example, the process of familiarising your dog to the target odour (Hall et al, 2014) acts … believe it or not … as the foundation for later searching!

Splitting … and even more splitting  … 

So, as mentioned earlier, I start by splitting my Scent Detecting training into three separate, but interconnected, phases;

  1. Familiarising my dog to the target odour.
  2. Training the passive indication.
  3. Searching.

… but, my splitting doesn’t end there! Each of these three phases will be split further.

For example, the apparently simple phase of familiarising my dog to the odour – which takes approximately 30 minutes to conclude – is split, again, into the following elements;

  1. Introduction to the environment – Physical and Psychological. If you’d like to know a little bit more about this particular element, follow this link to Scent Detecting and the Enriched Environment
  2. Choice of equipment – Harness and Line.
  3. Rewards – Food vs Toys (play?). Quantity. Timing. Look out for my forthcoming Blog post – Food, Glorious Food.
  4. Selection of the Target Odour – Safety. Cost and Availability. Previous use. Single Odour vs Configural Odour. Quantity (Concentration) of Odour. There’ll be more guidance in my forthcoming Blog post – Selecting a Training Odour for your Scent Detecting Dog. Some Considerations.
  5. Mechanics of Training – Classical (Pavlovian) conditioning. Development of a positive Conditioned Emotional Response (CER).

The devil is, very definitely, in the detail.

… and then there’s Back Chaining

According to Orr (2015), back chaining is a really effective way to build reliable behaviours. It’s a very efficient way to teach a specific behaviour, limits any potential errors and “leads to fluency with less training time” (Pryor, 2012).

Scent Detecting requires a whole series of individual behaviours that, together, form a particular behaviour chain. My dogs’ Scent Detecting behaviour chain looks something like this;

Cue to start work (Context. Harness. Scent in the Environment. Word) → Searching the Area → Locates the Target Scent Source → Passive Indication → Verbal Cue to move away from the scent source → Leaves the Scent Source → Click → Returns to me → Rewards

Your dog’s Scent Detecting behaviour chain may look very different to mine. It all depends on your training approach and your particular requirements … but I’m sure you get the idea; it’s complex and, for success, careful, thorough, training is required.

I teach Scent Detecting by starting at the end of the behaviour chain and working my way back to the beginning. As Orr (2015) states, “by teaching the last part first the learner [my dog] is always moving toward the part of the skill that he learned first and with which he is most confident“. For my dogs, this means learning about the target odour first – the familiarisation process.

Familiarising your dog to the target odour – the mechanics

Familiarising your dog to a specific (target) odour involves Classical, or Pavlovian,  Conditioning. It’s about helping your dog form positive associations with the scent. What starts out as a completely irrelevant odour to your dog, or neutral stimulus (NS), will, after conditioning, start to predict that good things (rewards) will follow (McLeod, 2013).

Familiarisation does not mean leaving the target scent with your dog (amongst his bedding, perhaps) for a few weeks until he becomes “used to”, or “familiar with”, it … unfortunately, “mere exposure [has] no effect on the acquisition of an odor discrimination in dogs” (Hall et al, 2014). If only it were that simple.

Scent : Detect : Find uses a very adapted form of Hall et al’s (2014) familiarisation protocol. Unit 13 is not a laboratory! Firstly, your dog is introduced to his “working” environment – he’s allowed to explore (or not) as he wishes – and then the familiarisation process begins. Over a 20 – 30 minute time-frame, as your dog continues to investigate the environment, he’s given the opportunity to “sniff” the target odour followed, immediately, with food rewards. No other behaviour is required … no need to sit, stand, down or similar. This procedure is repeated approximately 6 – 10 times within the 20 – 30 minute time period.

As for the “sniff” … well … some trainers place huge emphasis on hearing the dog inhale – sniff – during this familiarisation process … nothing less will do. This view probably has its origins in the work of Craven et al (2010) who state that, during normal respiration, approximately 12-13% of inspired air will reach the chemosensory area of the nose – the ethmoturbinate region – and that during active sniffing this may increase by 2-3%.

I’m not worried about hearing your dog sniff. That 12-13% of normally inspired air will do just fine for me! Why? Well let me give you an everyday – olfactory – example. If you walk along a road … breathing normally … you’re likely to become aware of a multitude of different odours, from the smell of bad drains, to the odour of frying chips, to diesel fumes, to … no need to sniff. Interestingly, you’re only going to start sniffing when you encounter a very faint hint of an odour, one that you’re trying to identify and locate. As Mainland and Sobel (2006) state, “when the olfactory system encounters a concentrated odorant, sniff vigor is reduced in real time; when it encounters a diluted odorant, sniff vigor is increased in real time“. In short, as your dog explores the environment, he’ll be aware of the target odour regardless of whether he chooses to sniff or not!

So, back to the familiarisation process and classical conditioning.

Before Conditioning – if your dog is offered food, an unconditioned stimulus (US), he’ll begin to salivate. It’s a physiological process resulting in an unconditioned response (UR). He just can’t help it.

Food (US) —> Salivation (UR)

Initially, if you let your dog smell what will eventually become his target odour, it’ll mean nothing to him. It’s a neutral stimulus (NS).

Target Odour (NS) —> No Salivation

During Conditioning – this requires you to allow your dog to smell the target odour and then provide him with food.

Target Odour (NS) —> Food (US) —> Salivation (UR)

After Conditioning – the target odour is now a conditioned stimulus (CS) and is associated with food. Smelling the odour in his environment will cause your dog to salivate in expectation of food rewards. Salivation in response to the target odour is now a conditioned response (CR).

Target Odour (CS) —> Salivation (CR)

According to Hall et al (2014), familiarising your dog to the target odour in this way enhances odor discrimination training and may reduce the overall training time. In short, it’s a worthwhile exercise to work through especially as it only takes approximately 30 minutes to complete.

Interestingly, and importantly, by classically conditioning your dog to the target odour, it will become what Berridge et al (2009) describe as, a “motivational magnet” – something that has to be approached, often compulsively. This is the foundation of all subsequent searching behaviour. Your dog WANTS to find the target odour. As Berridge et al (2009) explain, WANTING can apply both to unconditioned stimuli (US) such as food as well as conditioned stimuli (CS) such as the target odour after a classical conditioning procedure has taken place such as the one described here.

According to Litman (2005), WANTING involves dopamine activation in the brain and is thought to “motivate approach behaviour and to attribute incentive value to stimuli associated with reward” … the target odour!

But not only will your dog WANT to find the target odour, he’ll LIKE it too. After classical conditioning, your dog will have formed a positive conditioned emotional response (CER) to the target odour. As Litman (2005) states, LIKING involves brain opioid activity and “consequent states of pleasure“. Your dog will LIKE the target odour and WANT to find it. This phenomenon – your dog’s desire to get to the target odour – is sometimes referred to as Scent Obedience.

Familiarising your dog to the target odour and the notion of Imprinting

For clarity, in some scent detecting circles, the process of Familiarising your dog to the target odour (as described here, but often using very different training protocols) is frequently referred to as Imprinting. I prefer to avoid using this term in the context of scent detecting as it has its origins within the field of Ethology and is associated with the work of Konrad Lorenz and the rearing of Geese. Beware, terminology can be confusing and misleading!

and in conclusion …

If you’d like your scent detecting to progress a little more smoothly, consider splitting your training to avoid the lumps. Start where you’d like to finish – Back Chain!

Final Note

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

  1. Berg J (2018) Lumping and splitting. Science. 359. 6382. 1309
  2. Berridge KC, Robinson TE and Aldridge JW (2009) Dissecting components of reward; ‘Liking’, ‘Wanting’ and Learning’. Current Opinion in Pharmacology. Feb. 9(1): 65-73
  3. Craven BA, Paterson EG and Settles GS (2010) The fluid dynamics of canine olfaction: unique nasal airflow patterns as an explanation of macrosmia. Journal of the Royal Society. Interface. 7. 933-943
  4. Hall NJ, Smith DW, Wynne CDL (2014) Effect of odor preexposure on acquisition of an odor discrimination in dogs. Learning and Behavior. Jan
  5. Horowitz A and Hecht J (2014) Chapter 9. Looking at Dogs: Moving from Anthropocentrism to Canid Umwelt. IN: Horowitz A (2014) Ed. Domestic Dog Cognition and Behavior. The Scientific Study of Canis Familiaris. Springer.
  6. Litman JA (2005) Curiosity and the pleasures of learning: Wanting and liking new information. Cognition and Emotion. 19(6). 793-814
  7. Mainland J and Sobel N (2006) The Sniff is part of the Olfactory Percept. Chem. Senses. 31: 181-196
  8. McLeod S A (2013) Pavlov’s Dogs. accessed; 22.12.2015
  9. Orr J (2015) Back Chaining: The top secret teaching tool that is the key to professional success. TAGteach International. 
  10. Pryor K (2012) Back-Chaining “Retrieve”. Karen Pryor Clicker Training. Accessed: 3.8.2020
  11. Zerubavel E (1996) Lumping and Splitting: Notes on Social Classification. Sociological Forum. 11.3. 421-423.