Managing reactive and predatory dog behaviour

“Alternative Behaviour Training”

Overview

Unwanted dog reactivity and predatory behaviours can be influenced and reprogrammed by classical conditioning[1] and – at the right time - superimposing operant conditioning[2]. This is achieved, for example, as the dog reacts fearfully to a trigger (or spies a prey) that response is replaced with a positive (and distracting) experience – such as high value food (classical conditioning). Then, once the emotional response is on the “reprogramming path”, if the dog is taught to “Sit” (or take any other simple action – “Stand; Down; Touch”) and is rewarded (operant conditioning) this is reinforcing and, eventually, the “Sit” cue elicits the expectation of the reward, which transmutes into the dog sighting the reactivity trigger and automatically sitting to receive a reward, removing focus from the perceived feared or threatening trigger. The same process can be used to moderate unwanted predatory reactions.

Note that we do not regard this type of reactive or predatory behaviour as anything to do with “dominance”, as some pundits do[3]. These behaviours are generally simply fear (of the hairy green monster – whatever the trigger is), a level of warning off a perceived threat, or in the case of predation, an innate “prey drive” which is built-in to the dog’s nature - but can be modified or harnessed, as in the case of hunting and retrieving dogs (we distinguish this from the herding instinct which herding sheep/cattle dogs display). In this context the emotional response can be thought of as an instinctual response. Note that reprogramming the terrier breeds (initially bred mostly for vermin control) to moderate their predatory drive (especially of anything small and fast moving) although programmable with the protocols described, can be particularly challenging, based on their intense drive and independent nature and will mostly require careful management, from avoiding prey situations if possible and, in particular, interupting the predatory sequence, from spotting to capture, before the stalking and chasing phases are initiated.

We assume, in the following, that the guardian has already engaged in some basic “manners/obedience” training, such that in the house, in the yard or on a walk without severe distractions, the dog will follow basic cues such as “Wait”, “Sit”, “Look-at-me”, etc. That is, the dog already has some essential “control” factors in place and can behave reasonably sensibly (i.e. with appropriate manners) in the calm environment of the home, yard or uninterrupted walk. If this is not the case, then start by referring to the Training section technique description articles on this website to see what might be done to put in place a stepwise training program so that you have an established common language with your dog – you know what you want the dog to do and the dog understands your directions/cues.

Countering reactivity and predation

How it works – desensitization and counterconditioning

Remember that, in any canine behaviour change protocol, there are two essential parts –

·       Managing the behaviour (“management strategy”) and

·       Training alternate behaviour (“training strategy”).

In the simplest case, the management strategy would be identifying the trigger (the item or event that triggers the reactivity/predation drive) and either avoiding it completely or, as a step towards the reprogramming as part of the training strategy, managing the distance, duration, intensity and frequency of exposure to avoid going “over threshold” (where the dog can no longer “think” but is controlled/driven by that fear or drive). The training strategy would be the protocol and practice of implanting an alternate emotional reaction (the classical conditioning part) and behaviour (the operant conditioning part) through teaching over several controlled exposures and then (at the right time) overlaying a desired behaviour (such as the “Sit”, “Touch” or “Look-at-Me”).

Note that certain trainers advocate forcing the dog to “work through the fear/reactivity and make its own choice”, as if this is a better strategy to pursue in overcoming the dog’s emotional reactions to a trigger. The flaw in that thesis should be obvious, in that this is simply a version of flooding, which is excessive forced exposure to the trigger and which not only reinforces/heightens the continued fear/horror of the trigger, but potentially temporarily suppresses the reaction, delivering an illusion of success – when in fact the dog has “shut down”. That is, the dog is “scared to death” – or at least, “immobility”. This supposed cure is worsened when a punishing correction is added to the fear/reactivity event to stop the dog responding to the trigger, because it feels there is no escape and no alternative emotion, action, nor behaviour open to it, leaving a future “explosion of reactivity” on the horizon[4]. Without providing/conditioning an alternative emotional response to the dog, how can this help it to “make its own choice”? The dog has only the one ingrained pathway to a response, which is literally fear layered on fear.

The preferred reprogramming protocol is commonly referred to as “desensitisation and counterconditioning” - gradually increasing the intensity/closeness/duration of the trigger over several sessions so the dog can remain under threshold and become habituated to it and building in a desired/preferred emotional response. To help make descriptions of this protocol clear, some terminology bears explanation:

·       Desensitisation (somewhat similar to habituation) is getting used to something gradually (literally “becoming less sensitive”) - the dog learns that the trigger is not to be feared but rather ignored as of no consequence – or at least not reacted to. This is very different to the “flooding” of forced overexposure referred to above.

·       Counter-conditioning: a classical (respondent) conditioning technique in which the trigger is paired with something pleasant (such as food). This unconscious association makes the trigger transition from a negative or (with habituation) a neutral to a positive stimulus and may be referred to as a “conditioned emotional response”.

·       Conditioned emotional response: an animal associates a particular emotion with a stimulus; when the dog sees/hears/feels a particular stimulus, it immediately experiences that positive/negative emotion (as Pavlov’s dog’s reaction to the bell).

Action steps

The protocol can be implemented in two sequential, linked in desirable outcome, ways:

Firstly: when the trigger (the scenario/event that cause a reactive response in the dog) occurs, the dog is delivered high value treats (this is generally small food treats, but for certain dogs, could be the much-loved ball or tug-toy) to associate a positive outcome (the food/toy) with that trigger – and also distract from the trigger. The procedure is implemented at the dog’s own pace, avoiding going “over-threshold” (see further below). This is the essence of “desensitisation and counterconditioning”, viz., controlled exposure to the trigger but “conditioning the emotional response” to that trigger (through association – that is, classical conditioning) to mean something positive, thereby working on the dog’s reaction (and, hence, the observed unwanted behaviour - be that lunging, barking, growling, chasing) to become a positive/desired reaction and hence behaviour.

Secondly: once that positive association is established when the trigger occurs, the dog may be asked/cued to perform an action such as the "Come", “Sit; Stand; Down; Look at Me; Touch” (as taught through operant conditioning) and rewarded for that action – adding the dog’s brain focus on the action to the positive association in the sequence. This latter process of implanting operant conditioning (coupled with/overlayed on classical conditioning) is commonly referred to as implementing a “positive interrupter” - particularly important in managing predatory behaviour, interupting the predatory sequence as soon as the prey is spotted, but before the stalk-chase-capture sequence is initiated (once that sequence is in full flow, the dog is not listening and not thinking, the primal drive has taken over and there is little chance of stopping it). That is, to consolidate the desired outcome of changing the dog’s reaction, we are working on the emotional response through the “good association” (as above with classical conditioning) then overlaying the work on changing “behaviour” (which may be separated from the spontaneous emotional reaction, be it fear/predation/whatever).

Changing behaviour, indirectly changes the “temperamental reaction” to a trigger. Yes – it is a circular process/reaction, each part complimenting/reinforcing the other. In certain situations, this may also be helpful in changing the “perception of other dogs and humans/guardians (if they are the trigger) causing them to feel less threatened by your dog, so they in turn act less threatening to your dog”[5].

This “alternative behaviour training”[6] activates the thinking part of the dog’s brain. The challenge being addressed is that immediately the “reactive behaviour” initiates, or in predation during the stalking of the prey, the dog is very likely to switch to a primitive part of the brain that handles emotions and instincts. By performing a learned behaviour, the dog keeps thinking and responding and is hence less likely to fall into instinctual/emotional behaviour[7]. In addition, if already been trained with (for example) a reliable "Recall" and/or “Sit-Stay”, then there is further positive pressure for the dog to remain focused on the task rather than the reactive/predatory temptation, as it is just that little bit more difficult to chase or lunge if in a well implanted sit-stay or dashing back to the guardian (which it already understands it must perform and will be well rewarded for). The added advantage of this “positive interrupter training” is that when the dog and guardian are “surprised” by a trigger, the interrupter cue can be implemented to distract from that trigger if quickly applied and the dog physically guided (by encouragement and gentle leading) away.

Calming on a leash walk

Teaching the dog to enter a calm state on cue is a helpful overlay on these protocols and can be practiced on a street/park walk or at home, possibly inviting the dog to go in its crate with a kong/food puzzle for just 5 – 10 minutes (no need to lock the crate door).

During an on-leash walk, if the dog commonly reacts to passing dogs/people, then after 10-15 minutes of the “sniff and toilet” stage of the walk (preferably without distractions or “triggers” and probably on-leash) find a park bench just off/beside the track/walking zone and cue the dog with something like “Let’s sit”/ “Quiet time”/ whatever works for you. Sit quietly with the dog, focused on its body language and maintaining its attention with occasional treats to keep it focused on you and away from passers-by (no – stay off the mobile phone!). After, say, 5 minutes, cue “OK – let’s go” and proceed. Do this exercise every 15 minutes or so of the walk. Interspersing some basic manners/obedience training (either during the walk or in the calm seated periods) will help keep the dog in its “thinking brain” as the outing proceeds. If the dog’s reaction is going to occur as a “trigger” approaches (and there is no handy bench close by) then step out of the way with the dog focused away from the potential trigger and cue its attention (“Sit”/ “Look-at-me”) and reward until that trigger has passed by (i.e. the “positive interrupter” technique).

Warning

There is an important caveat in both elements of the protocols - they will be unlikely to work if the dog is already “over threshold”, that is, that the dog has already fallen into the primitive instinctual reaction of fear or predation. This can be determined very easily – if the dog will not perform the alternate behaviour (which you know it has learned and can do in ordinary undistracted circumstances) or if it refuses/shows no interest in the high value food or favourite toy being presented. Other indicative behaviours of the “over threshold state” are lunging/leaping/intense barking (fight), attempting to escape/hiding and avoiding interactions (flight), other general stress indicators, such as freezing, fainting (extreme – collapsing into coma-like state), fidgeting (self-licking/spinning, scratching, tail chasing and yawning – which dogs do for stress release when under even a little pressure), fawning (“cow-towing”/supplication), flocking (looking for protection from their flock/family/guardian)[8].

The options then are to immediately move away from the trigger, implement simple exercises or behaviours (like “Sit/Down”, “Go find” or “Touch”) to get its focus on you rather than the trigger and help to calm the dog. Alternatively, to regain the dog’s attention you may remind it that you are there and offer protection with a hand touch[9] (not the violent poke often seen implemented by TV dog handlers – nor the hiss they seem to favour – neither of which tells the dog the guardian is on their side, but rather is something else to fear – administered by the guardian/trainer). Remember, you are still maintaining the “dog companion” attitude in your training – and see below on “Comforting a dog in stressful situations”.

A comprehensive method of managing and training through reactivity is covered in the Behaviour Adjustment Training (“BAT”) described by Stewart (2016)[10] and another methodology using Constructional Aggression Treatment (“CAT”) is described by Snider (2017)[11]. Note that both protocols, although simple enough to understand and both roughly compatible with the protocols described herein, do require time and effort both in comprehending the protocols and in setting up many, many “staged dummy run scenarios”. Generally, a level of help from others (preferably with experience in planning the protocols and in setting up the practice scenarios) will be needed.

The time, effort, persistence and patience required is why the TV and YouTube coercive/punishment-based trainers scoff at these protocols, even claiming that they don’t work at all (because those trainers are incapable of, or can’t be bothered, understanding them and spending the time and effort) and that their punishment-based remedies “work in just one session of 10 or 15 minutes”. Do they? Well, yes, to the extent that the subject dog quickly fears the guardian more than the trigger and shuts down, uncertain what to do (refer again Footnote 4). That changes nothing in the dog’s mentality and emotional reactions but does leave the guardian with a fearful and unpredictable dog. The guardian/trainer is no longer the dog’s companion and so the dog is no longer the guardian’s companion either. Companionship is built on mutual trust, respect and affection[12].

One further warning. Despite best efforts at implementing these protocols (and be sure not to give up because of lack of time and patience and avoid jumping to a conclusion that the dog is “beyond help”) there will be certain dog’s whose emotional reactivity state, level of environmental reactions and hence maladaptive behaviours, are so high that they really need more help “to carry the load in their brain”. Dogs in this condition might be thought of as having such high arousal/fear that it can be likened to a mental condition from which they need relief, just so they can come down from that state of high alertness - sometimes referred to as being highly strung or "constantly wired" - and “think”. This is when consulting an experienced and properly qualified Veterinary Behaviourist or Veterinary Behaviour Specialist[13] would be advisable, rather than only a dog trainer. The training you have implemented will not be thrown out as a waste of time, in fact, it will be a critical foundation and continue to play an important role in conjunction with the Veterinary Behaviourist’s advice.

For extreme cases, a Veterinary Behaviourist might suggest the guardian considers targeted medication to “help the dog carry the mental load”. This is not just drugging a dog into “euphoric stupidity” – actually just the opposite, it is providing more opportunity for the dog to stay longer “in the thinking brain” when faced with a trigger and so being able to respond to the management and training strategies described above. Here we are referring to dogs for which the desensitization and counterconditioning techniques alone are not getting where the guardian hopes to be – but neither will pure medications alone deliver the desired outcome. The two protocols must work in concert and a good Veterinary Behaviourist will explain this to the guardian, including how to use any proposed medication sequence to enable the training strategies described take effect, then how to scale down and wean off any medication in a sensible time frame.

An important note is that we are not addressing exhuberant puppy behaviour in the foregoing, when what is required is regular, careful, sympathetic, fun and structured training of the desired "manners" the guardian is hoping the pup will come to adopt. Obviously, this will be aided by the maturity of the pup as it ages, but don't just hope that it will grow out of any unwanted behaviours with age and ignore training. Bad habits generally get worse with age, not better.

Unwanted/excessive barking at the window or door

The protocols described can be extended to a way to change the common behaviour of over-reaction exhibited as panic barking in the house in response to a stimulus, such as seeing a passer-by through the window, or hearing a noise from the neighbor. This procedure is not immediately intuitive but makes sense of utilising the combination of classical conditioning and operant conditioning and so might also be described as “calming via desensitization and counterconditioning” but really focusses mostly on the counter-conditioning part. “Over-reaction” is emphasised because it is generally part of the dog’s job description – according to both the guardian and the dog – to warn the household of what the dog perceives as a possible threat. So, a level of tolerance and sensible management is required and the following protocols are for situations in which the guardian finds the dog’s reaction excessive, often to the point where neither the guardian nor the dog have any peace of mind.

As above, there are two steps to this procedure (protocol parallels that described by Hodel[14]). The first is through classical conditioning (association of an inevitable outcome with the stimulus/trigger). The second is operant conditioning (training an action to be associated with the “cue”).

STEP 1: The method relied on food treating when the stimulus (e.g., person passing the window or noise from next door neighbor) triggers barking. This is not seen as rewarding the unwanted barking behaviour, but rather replacing the fearful or stressed feeling the stimulus generates with the positive emotion associated with receiving the treats – that is, a change in the dog’s perception of what the stimulus means – i.e., this is the first step of classical conditioning in the procedure, where the inevitable outcome of the stimulus (the noise or passer-by) is pleasant. Obviously, the guardian remains calm throughout – speaking soothingly and moving slowly – cue “OK, Relax, I’ve got this”; “Fine, I see that”; Yes, Good Dog, it’s OK”; consistently calmly voiced at the event can build the word-cue to relax.

STEP 2: Then – having repeated that first step MANY times over some weeks - and you will know it is time when the stimulated dog starts to look for the treat reward on hearing/seeing the stimulus - move to the second step - of operant conditioning – when the stimulus occurs, we ask (cue) the dog to take a desirable action, like “Go to bed, Sit, whatever, and reward that behavior. Using a chew/kong/toy/food puzzle as the reward on the bed/mat will enhance and extend the calming and the calm period.

Like all dog behaviour changes – lots of time and repetitions – and always having the treats/chew/kong/toy ready. Don’t rush the teaching – make sure STEP 1 is well established for every stimulus event before moving to STEP 2.

Note on Comforting during stressful events

We are often told that a guardian shouldn’t offer comfort to their dog during stressful situations. The logic is generally that this would reinforce the behaviours associated with the response to triggers. However, a better understanding is that if there is anything that can be done to make the dog less stressed it is OK, because such comforting helps reduce the dog’s arousal and improves its ability to think and opens the door to the classical conditioning element of a positive association described above. Obviously, the guardian should avoid escalating their dog’s emotional reaction (and thence behaviour) by their own reactions/behaviours, such as screaming, jerking the dog, running away, etc., assuming these are not necessary for safety reasons.

References

Working through the steps to “reprogram” a reactive or aggressive dog, or one which has a high predatory drive (assuming this is not desirable, because the dog is not actually used for hunting) can be complex and does take patience and persistence (as does all dog training). So, in addition to the references noted in the text, further reading for those looking to understand the spectrum of training options, the following references may be of interest:

Donaldson, J (2004) FIGHT: A Practical Guide to the Treatment of Dog-Dog Aggression; Printed USA; 115pp.

Greenwood, D (2024) Training Reactive Dogs; Strategies for Positive Reinforcement, Confidence-building and Enhancing Quality of Life for Canine and Owner; Amazon Sydney; 174pp.

Gurden, J (2021) Understanding Reactive Dogs: Why Dogs React and How to Help; 235pp

Hallgren, A (2012) Stress, Anxiety and Aggression in Dogs; Cadmos Publ. UK; 143pp.

Meuller, S., et al (2022) Don't Eat That: Force-Free Food Avoidance Training for Dogs who Love to Scavenge; Independent Publ.; 190pp.

Parsons, E (2014) Teaching the Reactive Dog Class; Leading the Journey from Reactivity to Reliability: Sunshine Books, USA; 163pp.

Phenix, A (2023) Positive Training for Aggressive and Reactive Dogs: Proven Techniques to Help Your Dog Overcome Fear and Anxiety; Fox Chapel Publishing; 256pp.

[1] Classical conditioning (also known as Pavlovian conditioning” or Respondent Conditioning) occurs when two stimuli are repeatedly paired: a response which is at first elicited by the second stimulus is eventually elicited by the first stimulus alone.

[2] Operant conditioning (“instrumental conditioning” or” Skinnerian conditioning”) is a method of learning that uses rewards (likely to be repeated) and punishment (prone to happen less) to modify behavior. Note that the punishment part of the equation is not advocated in this discussion – rather the techniques involve “redirection” of the dog’s focus and emotional response. See companion articles on this website: “Reinforcement vs punishment: A dog trainer’s perspective”; and “AVSAB vs Balabanov analysis”.

[3] See article on this site “Canine dominance – is it a real thing?” for this author’s views on this.

[4] Commonly referred to as “learned helplessness”: see Overmier, J. B. & Seligman, M. E. (1967) Effects of inescapable shock upon subsequent escape and avoidance responding; Journal of Comparative and Physiological Psychology; 63(1):28-33. For an expose of more recent articles ranging from 2002 to 2021 see URL: https://www.sciencedirect.com/topics/medicine-and-dentistry/learned-helplessness#: Note that a number of experiments described read as unpleasantly as the aversive techniques being discussed – to be avoided if you have a sensitive disposition.

 [5] From Dunbar, I (2023) Barking Up the Right Tree: The Science and Practice of Positive Dog Training; New World Library, Cal., USA: 363pp; esp. P. 46.

[6] After Mueller, S (2023) Hunting Together: Predation Substitute Training V 1: Harnessing predatory chasing in family dogs through motivation-based training. Germany; 159pp; esp. P. 87: “Predation Substitute Training (PST) is a motivation-based and need-oriented training program, designed to stop uncontrolled predatory chasing and to provide safe outlets for your dog's natural drive”.

[7] Note: The thinking brain (frontal cortex) is responsible for problem solving, memory, language, judgment, impulse control, and reasoning. The primal brain (hindbrain and medulla) is responsible for survival, drive, and instinct. When the primal brain is engaged (sympathetic response), the thinking brain is not working much: https://www.crisisprevention.com/blog/education/how-the-primal-brain-affects-behavior/ 

[8] These are the “Seven F’s”: Fight; Flight; Freeze; Faint; Fidget; Fawn; Flock; common reactions of dogs to stress or threat.

[9] Being cognisant of the dog’s arousal state and likely reaction to be sure the dog does not associate the touch/pat you deliver as proof of an imminent threat it was already reacting to (refer to Article on this site regarding reading dog body language “Dog Body Language” for some direction on this understanding).

[10] Stewart, G (2016); Behavior Adjustment Training [BAT]: New Practical Techniques for Fear, Frustration and Aggression in Dogs; Dogwise Publishing USA; 290pp.

[11] Snider, K. (2017) Turning fierce dogs friendly using Constructional Aggression Treatment (“CAT”) to rehabilitate aggressive and reactive dogs; Fox chapel publishing; 220pp. Note that this method is often criticised by positive-only training advocates, as it relies on “negative reinforcement” (something unpleasant/aversive is applied for the undesirable behaviour – such as simply leash pressure - then removed for good behaviour).

[12] Refer discussions in Article on this site “Reinforcement vs punishment: A dog trainer’s perspective”.

[13] These are veterinary professionals with experience in general practice who have undergone a subsequent higher level of formal tertiary study and examination into animal behaviour and psychology. Whilst cognisant of and promoting the training methods described herein, Vet Behaviourists are additionally able to diagnose and treat animal medical conditions – both physical and mental. In Australia see, for example,  https://www.petbehaviourvet.com.au/aboutus/morethanadogtrainer/ : These Vet Behaviourists are importantly distinguished from dog trainers who refer to themselves as “Dog Behaviourists” (who, incidentally, you can often identify by their practice of drawing psychological parallels between humans – often children – and dogs. Hopefully they don’t do the reverse).

[14] Hodel, B. (2020) How to Love and Survive Your Teenage Dog: The Complete Guide to your Teenage Dog; Green Hill (Australia): 189pp.