Critical Periods in Puppyhood: The Key to Raising a Confident Dog

For many years, early puppy training was viewed with scepticism. Puppies under six months of age were often described as too excitable, too distracted, or simply incapable of focusing enough to benefit from structured training. As a result, pet guardians were frequently advised to wait before enrolling their puppies in any form of formal education.

Recent research has fundamentally changed this view. Far from being ineffective, early puppy training is now understood to be a crucial component of healthy behavioural development. The science behind this shift highlights the importance of early experiences and explains why what happens in a puppy’s first months can shape behaviour for a lifetime.

This article explores the research underpinning early puppy training, focusing on critical periods in development and the long-term impact of early socialisation and learning. It is based on a Psychology Today article by Stanley Coren and draws on decades of behavioural science.

The origins of critical period research in dogs

One of the most influential bodies of research into canine development was conducted between 1945 and 1965 at Jackson Laboratories in Bar Harbor, Maine. Led by John Paul Scott, this extensive programme examined hundreds of purebred and crossbred dogs to understand how genetics and early experiences interact to shape behaviour.

Scott and his colleagues demonstrated that experiences during the first year of a dog’s life can have profound and lasting effects. Their work identified specific developmental windows—now commonly referred to as critical periods—during which experience has a disproportionate influence on future behaviour (Scott’s work is discussed extensively in Dewsbury, 2012; Bolman, 2022).

These findings challenged the idea that behaviour is largely fixed or that learning should be delayed. Instead, they showed that early exposure, even in small amounts, can produce long-term behavioural changes.

The three critical periods in puppy development

Scott’s research described three key developmental periods that are particularly relevant for puppy guardians and trainers.

The first is the primary period, which begins around day three of life and lasts until approximately day sixteen. During this stage, puppies are highly sensitive to environmental input, even though their awareness appears limited. Experiences during this time contribute to neurological development and help shape how puppies respond to stimulation later in life.

The second and most widely discussed phase is the socialisation period, which begins at around three weeks of age and continues until approximately twelve to fourteen weeks. This is when puppies learn how to interact with other dogs, humans, and the broader environment. At around three weeks of age, puppies show low levels of fear and a strong tendency to approach unfamiliar people and situations. As this period progresses, that natural openness gradually declines.

Experiences during this phase play a major role in determining how comfortable a dog will be with novelty, handling, and social interaction as an adult. Insufficient or poorly managed exposure during this time has been linked to increased fearfulness and anxiety later in life.
​
​The third stage is the juvenile period, which extends roughly from the end of the socialisation period until six to nine months of age. During this phase, puppies become increasingly capable of influencing their environment and adapting their behaviour to different situations. It is a particularly important time for reinforcing desirable behaviours through training and structured learning experiences.
​

References

​Coren, S. (2024, October 17). Do puppy classes have a lasting effect on dog behaviour? Psychology Today.
https://www.psychologytoday.com/us/blog/canine-corner/202410/do-puppy-classes-have-a-lasting-effect-on-dog-behavior
Dinwoodie, I. R., Zottola, V., & Dodman, N. H. (2021). An investigation into the impact of pre-adolescent training on canine behavior. Animals, 11(5), 1298.
https://doi.org/10.3390/ani11051298
Bolman, B. (2022). Critical periods in science and the science of critical periods: Canine behavior in America. Berichte zur Wissenschaftsgeschichte, 45(1–2), 112–134.
https://doi.org/10.1002/bewi.202200004
Dewsbury, D. A. (2012). John Paul Scott: The study of genetics, development, and social behavior. In W. E. Pickren, D. A. Dewsbury, & M. Wertheimer (Eds.), Portraits of pioneers in developmental psychology (pp. 229–248). Psychology Press.

If you’ve ever watched your dog suddenly freeze mid-walk, drop into a crouch, lock eyes on a pigeon… and then explode forward like a missile, you’ve seen prey drive in action.

It can look intense. Sometimes even alarming. But it isn’t bad behaviour, defiance, or a lack of training. It’s instinct — deeply wired into what it means to be a dog.

Prey drive is the natural motivation to orient toward, chase, and sometimes capture moving things. It evolved to help animals survive. In modern dogs, it still shows up — just in less helpful contexts, like squirrels, cyclists, skateboards, cats, or wildlife across the park.

And importantly, prey drive isn’t limited to “hunting dogs”. Terriers, herding breeds, sighthounds, companion breeds, and yes, even tiny dogs can all show strong chase behaviours. The difference isn’t whether prey drive exists, but how it’s expressed.


Prey drive is a sequence, not a single behaviour

One of the most helpful ways to understand prey drive is to see it as a behavioural sequence rather than a single action.

​In its full form, that sequence looks like this:
​
• Orienting — scanning, sniffing, tracking movement
• Stalking — freezing, crouching, creeping forward
• Chasing — rapid pursuit
• Capture and shake — often seen when dogs shake toys
• Consumption — a phase most pet dogs never reach

​Not every dog completes every step. Herding breeds often orient and stalk but don’t grab. Terriers tend to chase, grab, and shake. Some dogs explode straight into the chase without much warning at all.

Understanding where your dog tends to “enter” this sequence is key. The earlier you can intervene, the more successful and humane your training will be.


Why prey drive can become a problem

Prey drive itself isn’t a flaw. In fact, it fuels many behaviours we actively encourage: fetch, tug, toy play, scent work, and structured sports.

The problem arises when the chase becomes unsafe.

A dog who is fully locked into the sequence may ignore cues, traffic, people, or other dogs. That intense arousal — the dopamine rush of the chase — is powerfully reinforcing. The more often a dog practises chasing, the more efficient and motivated they become at doing it again.
​
This isn’t stubbornness. It’s learning.
​
That’s why simply telling dogs to “stop” rarely works once they’re already in full pursuit. Management and prevention matter just as much as training.
​

Managing prey drive safely and ethically

When working with dogs who love to chase, there are five broad principles that consistently help.

1. Prevent rehearsal

Chasing is fun — enormously fun. So the first step is limiting opportunities for your dog to practice it.

That might mean using fixed-length leads rather than retractables, secure fencing, baby gates, or careful separation from smaller animals. If a dog doesn’t get reinforced for chasing, the habit weakens over time.

When introducing a new pet to a prey-driven dog, supervision isn’t optional. Good management protects everyone involved.

2. Teach incompatible behaviours

While preventing unwanted behaviour, we actively teach behaviours that can’t happen at the same time as chasing.

This could be orienting to you, a strong “watch me”, a disengage-and-reorient game, or even moving toward you instead of away. The key is choosing behaviours your dog can perform easily and finds genuinely reinforcing.

You’re not suppressing instinct — you’re giving it a different outlet.

3. Practice with mild versions of triggers

Once your dog can respond reliably in low-distraction environments, you can carefully introduce controlled versions of the trigger.

Distance matters. Leashes matter. Reinforcement matters. Success might initially be nothing more than a brief glance away from the chase target — and that’s worth reinforcing.

Progression should be slow and intentional.

4. Have a backup plan

If your dog doesn’t respond to a cue, calmly guide them away and reinforce that choice. Then reassess your setup.

In some cases, mild space-blocking — calmly stepping between your dog and the trigger — can interrupt the sequence. But this should only be used after you’ve built strong reinforcement histories for alternative behaviours, not as a primary strategy.

​5. Be patient and consistent

Chasing habits don’t disappear overnight. Expect weeks or months, not days. Consistency, repetition, and realistic expectations make all the difference.
​
Emergency skills like a strong “leave it” or recall can also interrupt the sequence before it fully escalates — but they work best when paired with good management and reinforcement, not pressure.
​

You don’t need to eliminate prey drive

Trying to suppress prey drive entirely often backfires. Frustration builds. Behaviour leaks out elsewhere.

A far more effective approach is to channel it.

Dogs who love sniffing often thrive in scent work or nose games. Dogs obsessed with chasing may find healthy outlets in flirt poles or lure coursing. Terriers who dig? Earthdog trials were designed for them.

When dogs are allowed to express their instincts safely and appropriately, they’re often calmer, more responsive, and easier to live with.

Prey drive isn’t a problem to fix. It’s a trait to understand, manage, and work with.
​
If your dog has a bit of wild in them, you’re not doing anything wrong. With the right setup, clear training, and ethical support, prey drive doesn’t have to put anyone at risk — and it doesn’t have to take the joy out of life with your dog.

If you’d prefer to watch rather than read, I’ve also made a full YouTube video breaking down prey drive, the chase sequence, and practical management strategies.

You can watch it here:

References:
​
McConnell, P. B. (2010, June 15). Chase this, not that. The Other End of the Leash. https://www.patriciamcconnell.com/theotherendoftheleash/chase-this-not-that
​Schade, V. (2023, October 3). Prey drive in dogs. PetMD. https://www.petmd.com/dog/behavior/prey-drive-dog

When people talk about dog behaviour, the conversation often turns quickly to breed.

“She’s a Border Collie — of course she’s intense.”
“He’s a Labrador, they’re always friendly.”
“That breed is just aggressive.”
​
These statements feel intuitive, but how accurate are they?
​
To answer that properly, we need to look at what science actually says about genetics, breed differences, and behaviour, and — just as importantly — what genetics doesn’t explain.

Behaviour is always the result of an interaction between genes and environment. Genetics matter, but they don’t operate in isolation, and they don’t lock dogs into a single behavioural outcome.

Understanding this balance is essential for guardians, trainers and behaviour professionals who want to work with dogs rather than against them.

Genetics as Predisposition, Not Prescription

Genetics influence behaviour by shaping predispositions — tendencies toward certain responses, motivations, or sensitivities. They do not function as rigid instructions that dictate exactly how a dog will behave in real life.

A useful way to think about genetics is as a set of probabilities. A dog may be more likely to find movement reinforcing, more likely to show persistence in tasks, or more likely to be sensitive to unfamiliar stimuli — but likelihood is not certainty.

This distinction is crucial.

Two dogs of the same breed can share broad genetic tendencies and still behave very differently depending on their early experiences, socialisation, learning history, and daily environment. Genetics load the dice, but experience determines how the dice are rolled.

​Why Breeds Show Behavioural Patterns at All

Breed differences in behaviour didn’t arise by accident.
For generations, humans selectively bred dogs for specific functional roles: herding livestock, retrieving game, guarding property, controlling vermin, or working closely with people. Over time, this selective breeding increased the frequency of certain behavioural traits within breed populations.
​
As a result, some breeds are more likely to show behaviours such as:

• Chasing and orienting to movement
  
  
• Willingness to work closely with humans
  
  
• Persistence when faced with obstacles
  
  
• Sensitivity to environmental changes
  
  
• Responsiveness to cues and reinforcement
  
  
  

For example, Border Collies often show strong herding-related behaviours, while Labradors frequently display retrieving tendencies. These patterns are real and observable at a population level.

However, population trends do not define individuals.

What Large-Scale Research Tells Us

A large behavioural genetics study examined 101 dog breeds across 14 behavioural dimensions, including traits such as trainability, aggression, fearfulness, attachment and chasing behaviour.

The findings showed that genetics accounted for a substantial proportion of behavioural variation across breeds, with heritability estimates often falling between 60 and 70 percent when comparing breed averages.

Traits with particularly strong genetic signals included:

• Trainability
  
  
• Stranger-directed aggression
  
  
• Chasing behaviour
  
  
• Attachment and attention-seeking
  
  
  

This means that, on average, breeds differ from one another in predictable ways for certain behaviours.

But there’s an important caveat.

High heritability across breeds does not mean that behaviour is fixed within individual dogs. It simply means that when you compare breed averages, genetics explains a meaningful portion of the differences observed.

​Within a single breed, individual variation remains substantial.

Introduction
​ 
Modern dog training has evolved significantly, with scientific research consistently supporting the use of positive reinforcement over aversive methods. While punishment-based techniques may sometimes produce immediate results, they often come with serious long-term consequences for both behaviour and welfare. 
In this article, we’ll explore why aversive methods—such as physical corrections, verbal corrections, and intimidation—are not only less effective but also harmful, while positive reinforcement fosters better learning, stronger bonds, and improved emotional wellbeing. 
 
1. Positive Reinforcement Is More Effective 
Scientific studies and practical experience consistently show that positive reinforcement is the superior training method. Here’s why: 

• Faster learning – Dogs learn more quickly when rewarded for desired behaviours rather than punished for mistakes. 

• Fewer behavioural issues – Positive reinforcement reduces aggression, fear, and anxiety. 

• Stronger human-dog bond – Pleasant interactions increase trust and cooperation. 

• Higher satisfaction for guardians– Trainers using reward-based methods report greater enjoyment and success. 

In contrast, aversive techniques often create additional problems, making training more difficult in the long run. 
 
2. Aversives Increase Aggression 
Physical punishment can trigger defensive aggression in dogs. Research shows that dogs subjected to corrections (such as hitting, lead jerks, or shock collars) may: 

• Become hand-shy – Some dogs start flinching or snapping when a hand moves toward them, even if no correction is intended. 

• Develop pain-induced aggression – A dog in pain may lash out, even at their owner. 

• React to minor triggers – Aversive training can make dogs hypersensitive, leading to overreactions in normal situations. 

Aggression is rarely the intended outcome of punishment, yet it is a common side effect. 
 
3. Escape and Avoidance Behaviours 
Dogs trained with aversives often learn through escape and avoidance—meaning they perform behaviours simply to stop something unpleasant. The problem? 

• Fear becomes linked to the situation – The presence of the trainer, a lead, or even a training environment can trigger stress. 

• Behaviours become persistent – Escape responses are self-reinforcing because they reduce fear in the moment. 

• Defensive aggression may emerge – If a dog feels trapped, they may resort to aggression as a last resort. 

Instead of teaching a dog what to do, aversives teach them what to avoid—often at the cost of their emotional wellbeing. 
 
4. Negative Emotional Responses 
Emotions play a crucial role in learning. Just as humans associate negative experiences with fear or resentment, dogs do the same. 

• Fear inhibits learning – A stressed dog cannot focus or retain new information effectively. 

• Chronic stress leads to long-term issues – Repeated punishment can cause depression, hypervigilance, or apathy. 

• Positive reinforcement creates enthusiasm – Dogs trained with rewards are more eager to participate and learn. 

We should strive for training methods that build confidence, not fear. 
 
5. Learned Helplessness: When a Dog Gives Up 
One of the most severe consequences of aversive training is learned helplessness—a state where a dog stops trying to avoid punishment because they believe nothing they do will change the outcome. 
Signs include: 

• Shut-down behaviour – The dog becomes passive and unresponsive. 

• Depression and apathy – Loss of interest in play, food, or interaction. 

• Compromised welfare – The dog’s quality of life is severely diminished. 

This is a tragic outcome that no responsible trainer should risk. 
​​

6. Aversives Don’t Address the Root Cause 
Punishment suppresses symptoms rather than solving the underlying issue. For example: 

• A fearful dog that growls may stop growling after punishment—but their fear remains, potentially leading to a sudden bite later. 

• A dog that barks at other dogs may become quiet under threat but still feel intense anxiety. 

Without addressing the root cause (fear, frustration, or lack of training), the problem persists and may resurface in worse ways. 
 
7. They Don’t Teach an Alternative Behaviour 
Punishment tells a dog what not to do but fails to teach what they should do instead. 

• If a puppy chews shoes, punishing them doesn’t show them what’s acceptable to chew. 

• If a dog jumps on guests, shouting at them doesn’t teach them to sit politely. 

Effective training involves redirecting to appropriate behaviours and reinforcing those instead. 
 
8. Health Risks of Aversive Methods 
Physical corrections can cause injuries, especially around the neck (from check or prong collars). Potential issues include: 

• Tracheal damage 

• Spinal misalignment 

• Increased stress hormones (negatively impacting overall health) 

Positive reinforcement, on the other hand, carries very little or virtually zero physical risk. 
 
9. Proper Application of Punishment Is Nearly Impossible 
For punishment to be effective (without causing fear or aggression), several strict criteria must be met: 

• It must be immediate (within seconds of the behaviour). 

• It must be consistent (every single time the behaviour occurs). 

• The intensity must be just enough to stop the behaviour (but not so strong as to cause trauma). 

In real-world training, this precision is almost impossible to achieve—making punishment a poor choice compared to reward-based alternatives. 
 
10. Aversives Reinforce the Trainer, Not the Dog 
Ironically, punishment can be reinforcing for the trainer because it often produces an immediate (though temporary) change in behaviour. This leads to: 

• Increased reliance on punishment – The trainer repeats the method, often escalating intensity. 

• False confidence – The problem appears "fixed," even though the underlying issue remains. 

• Missed opportunities for better training – Positive reinforcement is overlooked in favour of quick fixes. 

 
Conclusion: The Clear Choice for Modern Dog Training
​ 
The science is clear: Positive reinforcement is safer, more effective, and more ethical than aversive methods. It builds trust, prevents aggression, and fosters a happy, well-adjusted dog.
​ 
For a more detailed discussion on this topic, be sure to watch my companion YouTube video where I break down these concepts with practical examples. ​

​Training a dog can sometimes feel frustrating, especially when it seems like your dog simply isn't grasping what you are trying to teach. In many cases, the problem isn't the behaviour or even the learner — it's the timing. Understanding and applying the Law of Contiguity can make a remarkable difference in how quickly and effectively your dog learns.

The Law of Contiguity is a fundamental principle of learning which states that an association is formed between two events when they occur closely together in time. In the context of dog training, this means a dog is much more likely to associate a behaviour with a consequence if the consequence happens immediately after the behaviour.

Timing is absolutely critical. If you delay reinforcement — even by a few seconds — your dog may not understand what they did correctly. Instead, they might associate the reward with whatever they happen to be doing at the time the consequence occurs. This can cause confusion and slow learning, or worse, lead to associations you did not intend.

The same principle applies when attempting to discourage unwanted behaviour. Unfortunately, many well-intentioned dog guardians inadvertently violate the Law of Contiguity when using punishment or corrections. For instance, punishing a puppy for a toilet mistake hours after the event is not only ineffective but can also create unnecessary anxiety and fear. In such cases, because there is no close temporal proximity between the behaviour and the consequence, the puppy does not understand what they are being punished for. They may display submissive behaviours that look like 'guilt,' but these are simply reactions to the guardian’s current emotional state and body language — not an understanding of past actions. Importantly, even with impeccable timing, punishment procedures are better avoided when kinder and more effective ways are available.

Another example highlights the importance of timing when teaching new behaviours. Imagine teaching your dog to sit. If your dog sits and you deliver a reinforcer — like a treat or praise — within a second or two, they are likely to associate the sitting behaviour with the reward. However, if you wait 10, 30, or even 60 seconds before offering reinforcement, the opportunity for a clear association is lost. By then, your dog might be sniffing the floor, looking around, or engaging in a completely different behaviour, and the link between the sit and the reward disappears.

Tools like clickers in clicker training are designed to help bridge this gap by marking the correct behaviour at the precise moment it occurs, even if the reward follows slightly later. This way, the dog clearly understands which behaviour earned the reinforcement.

Timing also plays a crucial role when using negative punishment strategies. For instance, if a dog jumps on people, immediately withdrawing attention — such as turning away or leaving the room — can help the dog associate the jumping behaviour with the loss of social interaction. However, if attention is withdrawn only after a delay, the association becomes unclear, and learning is compromised. Importantly, while negative punishment can be an effective tool, using management strategies and teaching alternative appropriate behaviours are often more humane and effective long-term solutions.

Interestingly, the Law of Contiguity can be traced back to ancient philosophy. It was first proposed by Aristotle and has since been studied and validated by numerous philosophers and researchers. It applies not just to dogs, but to any being with a brain.

The take-home message is clear: the Law of Contiguity underscores the importance of immediate consequences in shaping behaviour. Whether reinforcing desired behaviours or discouraging unwanted ones, success depends on your ability to ensure that consequences closely follow the behaviours you want to address.

For a deeper dive into this important principle and practical tips on applying it in your training, check out the full video below.
​

At the turn of the 20th century, a horse named Hans captivated the German public—and eventually, the world. Dubbed Clever Hans, this seemingly unremarkable horse was believed to possess astonishing intellectual abilities. Under the guidance of Wilhelm von Osten, a German mathematics instructor and amateur horse trainer, Hans was presented as an animal prodigy capable of solving arithmetic problems and even answering calendar-related questions.

According to von Osten, Hans could add, subtract, multiply, and divide. He could work with fractions, and his understanding extended beyond numbers into concepts related to dates and sequences. Whether questions were posed verbally or in writing, Hans would respond with a series of hoof taps—stopping, remarkably, at the correct number every time.

Von Osten toured with Hans across Germany, drawing large crowds wherever they went. Spectators were eager to witness what seemed like undeniable proof that animals, under the right circumstances, could possess intelligence rivalling that of humans. Hans’s performances were met with awe and curiosity, inspiring admiration as well as scepticism.

Enter Oskar Pfungst, a psychologist with a growing interest in animal behaviour and the scientific method. Sensing that there was more to Hans’s abilities than met the eye, Pfungst set out to investigate. In 1907, he conducted a series of careful and controlled experiments designed to isolate the true source of the horse's apparent intelligence.

Pfungst’s experimental design was ahead of its time. He tested Hans with multiple questioners, sometimes allowing them to know the answers and other times deliberately keeping them in the dark. He varied the presentation of questions and observed the horse under different environmental conditions. Over time, a clear pattern emerged.

Hans only performed well when the questioner knew the answer. When the person posing the question was unaware of the correct response—or if visual cues were blocked—Hans could no longer produce the right number of taps. This discovery led to a breakthrough: Hans wasn’t doing maths at all. Instead, he was reading incredibly subtle signals from humans—tiny changes in posture, shifts in facial expression, even minute muscle movements.

Pfungst’s findings dismantled the myth of Hans the mathematician but revealed something arguably more fascinating: the horse’s incredible sensitivity to human body language. This became known as the Clever Hans Effect, a term that remains relevant in scientific and training communities to this day. It describes the way an animal (or human) can unconsciously respond to involuntary cues provided by another individual, especially during testing or training.

What made Pfungst’s contribution so valuable wasn’t just the outcome—it was the methodology. His meticulous attention to experimental controls set a precedent for behavioural science. By demonstrating how unintentional cues could skew results, he highlighted the critical importance of controlling for observer bias in any investigation involving live subjects.

​

The Clever Hans Effect has wide-ranging implications. In scientific research, particularly in the study of behaviour and cognition, it has prompted the widespread use of double-blind procedures. These approaches ensure that neither the experimenter nor the subject knows which variables are being tested, helping prevent the kind of unconscious cueing that fooled the world in Hans’s case.

Even outside the lab, the effect plays a significant role. For example, when training drug-sniffing dogs, it’s essential that handlers are unaware of which containers contain contraband. If they know, they might—without realising—give away the location through a glance, a shift in stance, or a change in breathing. The dog, attuned to its human partner, may pick up on that signal and indicate a "find" based on human behaviour rather than scent detection.

This story also holds important lessons for those of us working closely with animals, whether as trainers, behaviour consultants, or curious observers. It reminds us to reflect carefully on what our animals are responding to and whether we might be shaping behaviour unintentionally. It challenges us to be more precise in our training, more thoughtful in our observations, and more humble in our assumptions.

Above all, the tale of Clever Hans is a powerful example of the scientific process in action. What began as a sensation built on anecdotal performance became, through careful investigation, a case study in critical thinking and experimental rigour. It urges us to meet extraordinary claims with healthy scepticism and to ask deeper questions about the mechanisms behind what we see.

So, the next time someone shares a story about an animal with seemingly supernatural abilities, take a moment to think of Hans. Let curiosity lead the way—but don’t forget the value of cautious inquiry and the importance of good experimental design.

To see the full story brought to life, check out the video on my YouTube channel, Train Me Please.

Reference
Bellows, A. (2007, February). Clever Hans the Math Horse. Damn Interesting. https://www.damninteresting.com/clever-hans-the-math-horse/

Photo reference
The “Clever Hans Phenomenon” revisited - Scientific Figure on ResearchGate. Available from: https://www.researchgate.net/figure/clever-Hans-an-Orlov-trotter-horse-1895-1916-and-his-owner-and-teacher-Wilhelm-von_fig1_260376462 [accessed 25 Feb, 2024] 

Operant conditioning is a foundational concept in modern dog training, yet its depth and complexity are often under-appreciated. This article delves into its historical development, its relationship with classical conditioning, and its crucial role in behaviour modification.


The Origins of Operant Conditioning


While Ivan Pavlov was pioneering research into classical conditioning, psychologist Edward L. Thorndike was investigating how animals solve problems. He formulated the "Law of Effect," which states that behaviours followed by pleasant consequences are likely to be repeated, whereas those followed by unpleasant consequences are likely to diminish.


Building on Thorndike’s work, B.F. Skinner introduced the concept of operant conditioning in 1937. He argued that organisms learn responses by interacting with their environment, with behaviour being modified through its consequences. This laid the groundwork for applied behaviour analysis and modern training methodologies.


The Interplay Between Classical and Operant Conditioning


Although classical and operant conditioning are often discussed separately, they are deeply interconnected. In reality, distinguishing between the two is a simplification for analytical purposes. Both processes occur simultaneously: when we shape behaviour using operant conditioning, we also influence the learner’s emotional state through classical conditioning. This underscores the importance of considering both aspects in training.


Contingencies and the ABCs of Behaviour


For operant conditioning to occur, there must be a contingency—a causal relationship between behaviour and environmental consequences. The most fundamental form is a two-component contingency: for example, a dog sits and receives a treat, or a dog barks and another dog moves away.


In applied behaviour analysis, a more comprehensive model is the three-component contingency:


Antecedent: The cue or situation prompting the behaviour.
Behaviour: The dog’s response.
Consequence: The outcome that influences future behaviour.


For example, if a dog sees another dog too close (antecedent), barks (behaviour), and the other dog moves away (consequence), the barking behaviour is likely to be reinforced. Context plays a crucial role, as external factors can influence whether a behaviour occurs. For instance, a dog’s recall response may vary depending on environmental distractions.


The Nature of Operant Behaviour


Operant behaviour consists of voluntary actions that an organism performs in response to environmental stimuli. These behaviours are defined by their consequences, which can be classified into five categories:


Positive Reinforcement: Adding a stimulus to increase behaviour frequency (e.g., giving a treat for sitting).
Negative Reinforcement: Removing a stimulus to increase behaviour frequency (e.g., releasing pressure on a dog’s bottom when they sit).
Positive Punishment: Adding a stimulus to decrease behaviour frequency (e.g., yelling at a dog for jumping, if jumping decreases).
Negative Punishment: Removing a stimulus to decrease behaviour frequency (e.g., walking away when a puppy bites, if biting decreases).
Extinction: Eliminating reinforcement to reduce a behaviour (e.g., ignoring begging at the table to decrease the behaviour).


It is important to note that reinforcement and punishment are defined by their effects on behaviour, not by intent. Additionally, the terms "positive" and "negative" refer to the addition or removal of stimuli, not to their pleasantness.

​Understanding Negative Reinforcement


One of the most misunderstood concepts in operant conditioning is negative reinforcement. It involves increasing a behaviour by removing an aversive stimulus. However, for something to be removed, it must first be introduced—this often results in a combination of positive punishment and negative reinforcement.


A classic example involves training a horse to turn using a bit. When pressure is applied to the reins, the horse experiences discomfort. As soon as the horse turns, the pressure is released. The turning behaviour is negatively reinforced (because turning removes the pressure), while the behaviour of moving forward in a straight line is positively punished (because an aversive stimulus was added when the horse failed to turn).


The Implications of Extinction


Extinction occurs when a behaviour that was previously reinforced is no longer reinforced, leading to a decrease in its occurrence. However, this process is often accompanied by an extinction burst, where the behaviour temporarily intensifies before fading. If reinforcement resumes, spontaneous recovery can occur, reinstating the behaviour. This unpredictability makes extinction a less reliable behaviour modification tool compared to reinforcement-based strategies.


The Risks of Positive Punishment


Scientific evidence strongly advises against using positive punishment as a primary training method. It carries a high risk of negative side effects, including fear, aggression, and damage to the trainer-dog relationship. Furthermore, it often fails to provide the learner with an alternative behaviour to perform, making reinforcement-based approaches more effective and ethical.


Final Thoughts


Operant conditioning is an essential framework for understanding and modifying behaviour. However, its application requires careful consideration of contingencies, reinforcement schedules, and the broader emotional impact on the learner. By leveraging reinforcement and minimising punishment, we can create ethical, effective training strategies that benefit both dogs and their trainers.


To explore these concepts further, watch our YouTube video on operant conditioning and download our digital handout for additional insights and practical applications.

​Digital Handouts available:

Operant Conditioning Quadrants and Extinction https://www.buymeacoffee.com/trainmeplease/e/130745
Flowchart of Basic Operant Conditioning Procedures
https://www.buymeacoffee.com/trainmeplease/e/130754


References:
O'Heare, J. (2010) Changing Problem Behavior. BehaveTech Publishing.

Classical conditioning is one of the foundational principles of learning that governs behaviour across species, including dogs and humans. First identified by the Russian physiologist Ivan Pavlov in the late 19th and early 20th centuries, this process explains how associations between stimuli shape behavioural responses. Understanding classical conditioning is crucial for dog trainers, pet guardians, and behaviour professionals, particularly when addressing emotional responses such as fear and excitement in dogs.


Pavlov’s Discovery and the Basics of Classical Conditioning

Ivan Pavlov’s research initially focused on physiological processes, specifically salivation in dogs. During his experiments, he observed that his dogs would begin to salivate before the food was presented, merely in anticipation of the meal. This unexpected response led him to investigate the mechanisms behind it, eventually formulating the principles of classical conditioning.

Although popular explanations often refer to Pavlov using a bell in his experiments, he actually used a metronome. The bell has become a common example because it simplifies the explanation.

Pavlov identified several key components in this process:

• Unconditioned Stimulus (US): A stimulus that naturally elicits a response (e.g., food).
• Unconditioned Response (UR): The instinctive reaction to the unconditioned stimulus (e.g., salivation when food is present).
• Neutral Stimulus (NS): A stimulus that initially has no effect on behaviour (e.g., the sound of a bell before conditioning).
• Conditioned Stimulus (CS): The formerly neutral stimulus that, after repeated association with the unconditioned stimulus, elicits a response on its own (e.g., the bell after conditioning).
• Conditioned Response (CR): The learned response to the conditioned stimulus (e.g., salivation at the sound of the bell, even in the absence of food).

This process highlights how an originally neutral stimulus can acquire meaning through association, leading to predictable behavioural outcomes.


​The Role of Classical Conditioning in Dog Training

Classical conditioning is highly relevant in dog training, particularly in shaping emotional responses and addressing behavioural issues. Many of the associations dogs form with their environment stem from classical conditioning, influencing how they react to people, objects, and experiences.

​Common Examples in Everyday Life

1. Leash Excitement: If a dog consistently experiences a walk after seeing their guardian pick up the leash, the sight of the leash alone will begin to elicit excitement.
2. Feeding Cues: The sound of a can opener may prompt a cat to run to their food bowl, anticipating their meal due to repeated associations.
3. Fear of Nail Clippers or Syringes: If a dog experiences discomfort every time their nails are clipped or they receive an injection, they may begin to fear these objects even before anything happens, as the clippers or syringe become conditioned stimuli for discomfort.

These examples illustrate how classical conditioning operates in daily interactions with animals, often shaping their emotional states without deliberate training efforts.
​

Emotional Responses and Behaviour Modification

Classical conditioning plays a significant role in addressing behavioural concerns rooted in emotions, such as fear and anxiety. A fearful reaction to a raised hand, for example, may result from past punishment, leading the dog to associate a raised hand with negative experiences. Similarly, a dog who receives treats and praise when meeting new people may develop a positive emotional response toward social interactions.
Understanding and applying classical conditioning principles can help modify problematic behaviours by replacing negative associations with positive ones. This process, known as counterconditioning, is often used in behaviour modification strategies to reduce fear-based responses in dogs.


Classical vs. Operant Conditioning

It is important to distinguish between classical and operant conditioning. Classical conditioning involves associations between stimuli and involuntary responses, while operant conditioning deals with voluntary behaviours and their consequences. However, in real-world training, these two learning processes often occur simultaneously.
For example, when teaching a dog to sit, the act of sitting is reinforced (operant conditioning), but the emotions associated with training—whether positive or negative—are influenced by classical conditioning. A positive training experience leads to a dog feeling relaxed and eager to participate, while aversive methods may induce anxiety or fear, creating negative associations with training sessions.


Conclusion

Classical conditioning, also known as Pavlovian or respondent conditioning, is a powerful mechanism that shapes behaviour in dogs and humans alike. From basic reflexes to complex emotional responses, this scientific principle provides invaluable insight into learning and behaviour. By applying classical conditioning thoughtfully in training, pet guardians and trainers can create positive associations that enhance their dogs' well-being and foster better communication.
​
To explore this topic further and see how classical conditioning shapes behaviour, watch the full video below.
​

Flooding as a dog behaviour modification procedure is almost always a terrible idea. Yet, it remains a popular approach in dog TV shows and online content. Unfortunately, people searching for information on dog training—particularly about reactivity and aggression—are frequently exposed to methods that rely on flooding.
This is both disappointing and unsurprising.


The Influence of Media and Popular Culture

In the early 2000s, the TV show Fear Factor captivated audiences by forcing participants to confront their deepest fears—whether that involved snakes, spiders, or extreme heights. The show had no real therapeutic value; in fact, the psychological impact of these forced exposures ranged from negligible to outright harmful. However, it made for commercially appealing content.
The same principle applies to many popular dog training shows. The promise of a dramatic transformation in just minutes makes for compelling viewing, but what is really happening?
In many cases, these so-called transformations fall under the umbrella of flooding—a procedure that is not only ineffective but also often misapplied.


What Is Flooding in Dog Training?

Flooding is a technique in which a dog is exposed to a feared stimulus at full intensity while being prevented from escaping or avoiding it. The idea is that the dog will eventually stop exhibiting fear responses.
For example:

• A dog afraid of loud noises might be confined to a room while loud sounds play, with no option to leave.
• A dog fearful of water might be forced into a pool and prevented from escaping.
• A reactive dog may be placed close to other dogs while escape behaviours like barking and lunging are suppressed.

In theory, once the dog stops reacting, it has "overcome" its fear. In reality, this process often leads to significant behavioural and emotional harm.


The Problems with Flooding

More Ethical and Effective Alternatives Exist
Ethical and scientifically sound methods, such as desensitisation and counterconditioning, achieve long-lasting behaviour change without unnecessary distress.

Incorrect Application
Even in human clinical settings, flooding is rarely used, and when it is, it is carefully controlled. The aversive stimulus must be removed once the fear response ceases. In dog training, this condition is almost never met. Instead, dogs are often forced into overwhelming situations where their fear is intensified rather than reduced.

Risk of Learned Helplessness
Flooding can lead to a state known as learned helplessness, where the dog stops responding because it has learned that its actions have no impact on the environment. This can generalise beyond the specific fearful situation, leading to an overall suppression of behaviour.

Potential for Lasting Psychological and Physiological Damage
Studies indicate that exposure to high-stress situations can cause long-term emotional and physiological side effects. Since animals cannot verbally communicate their distress, it is difficult to assess when emotional harm has occurred, making flooding a highly risky approach.


Flooding in a Clinical Context

In human psychology, exposure therapy is sometimes used for PTSD, OCD, and anxiety disorders. However, this process is overseen by highly trained professionals in controlled environments. The same level of expertise is rarely present in dog training, where flooding is often applied by self-taught trainers with no formal education in behaviour science.
​
If you needed help overcoming a serious fear, would you trust someone without formal training? The same consideration should apply to dogs.


What’s Really Happening in Popular Dog Training Shows?

When we see a dog that appears "cured" after a short, intense exposure, it is often not because the fear has been resolved. More likely, the dog has entered a state of learned helplessness. To the untrained eye, the dog looks calm, but in reality, it has simply stopped trying to escape because it has given up.
​

​A Better Alternative: Desensitisation and Counterconditioning

For almost every case, desensitisation and counterconditioning are far superior to flooding. These methods involve gradually exposing the dog to a feared stimulus at a low intensity while pairing it with positive experiences. This allows the dog to form new, positive associations and adapt without overwhelming fear.
​
In my years of experience training dogs, I have yet to encounter a case where flooding was the best or even a reasonable option. Not only that, but I frequently consult with or refer cases to professionals who specialise in behaviour modification, ensuring that dogs receive the best possible care.


Final Thoughts

While dramatic, flooding-based transformations may make for good TV, they do not make for ethical or effective dog training. If you are facing behavioural challenges with your dog, seeking guidance from a qualified professional with formal education should be your first step. Ethical, science-based approaches not only produce better results but also protect your dog’s well-being.

If you’d like to learn more about how to use desensitisation and counterconditioning effectively, I have a video on my YouTube channel covering these techniques in detail. I’ll link it below.

By choosing humane, science-based methods, we can ensure that dogs receive training that is both effective and compassionate.

​There are several options in terms of reinforcement schedules that can be used for behaviour modification. In this text I will provide you with a quick description of each of the different simple schedules and a couple of examples for each (one human example and one animal training example). I will also offer a couple of considerations for people debating the idea of which schedule to use for a given situation.
 
Early in my career I was told that, in general, a good way to go about training animals would be to use a continuous schedule of reinforcement for teaching a new behaviour and to then maintain the behaviour using a “variable schedule of reinforcement”. This is a very broad statement and one that seems to make sense to someone being introduced to animal training. However, is this really the best option to go about when training animals? And what do people mean when they mention “a variable schedule of reinforcement”? Let’s start by defining the most common types of simple schedules of reinforcement according to Paul Chance’s book Learning and behavior (2003; figure 1).

Figure 1 – The most common types of simple reinforcement schedules
 
​
The simplest type of reinforcement schedule is a Continuous reinforcement schedule. In this case every correct behaviour that meets the established criteria is reinforced. For example, the dog gets a treat every time it sits when asked to do so; the salesman gets paid every time he sells a book.
 
Partial Schedules of reinforcement can be divided into Fixed Ratio, Variable Ratio, Fixed Interval and Variable Interval.
 
In a Fixed Ratio reinforcement schedule, the behaviour is reinforced after a certain amount of correct responses has occurred. For example, the dog gets a treat after sitting three times (FR 3); the salesman gets paid when four books are sold (FR 4).
 
In a Variable Ratio reinforcement schedule, the behaviour is reinforced when a variable number of correct responses has occurred. This variable number can be around a given average. For example, the dog gets a treat after sitting twice, after sitting four times and after sitting six times. The average in this example is four, so this would be a VR 4 schedule of reinforcement. Using our human example, if the salesman gets paid after selling five, fifteen and ten books he would be on a VR 10 schedule of reinforcement, given than ten is the average number around which his payments are offered.
 
In a Fixed Interval reinforcement schedule, the behaviour is reinforced after a certain behaviour has happened, but only when that behaviour occurs after a certain amount of time. For example, if a dog is in a FI 8 schedule of reinforcement it will get a treat the first time it sits, but sitting will not produce treats for the next 8 seconds. After the 8 second period, the first sit will produce a treat again. The salesman will get paid after selling a book but then not receive payment for each book sold for the next 3 hours. After the 3-hour period, the first book he sells results in the salesman getting paid again (FI 3).
 
In a Variable Interval reinforcement schedule, the behaviour is reinforced after a certain variable amount of time has elapsed. The amount of time can vary around a given average. For example, instead of always reinforcing the sit behaviour after 8 seconds, that behaviour could be reinforced after 4, 8 or 12 seconds. In this case the average is 8, so it would be a VI 8 schedule of reinforcement. The salesman could be paid when selling a book after 1, 3 or 5 hours, a VI 3 schedule of reinforcement.
 
The next question would be “How do the different schedules of reinforcement compare to each other?”. Kazdin (1994) argues that a continuous schedule of reinforcement or at the very least a “generous” schedule of reinforcement is ideal when teaching new behaviours. After a behaviour has been learned, the choice of which type of reinforcement schedule to use becomes somewhat more complex. Kazdin also mentions that behaviours maintained under a partial schedule of reinforcement are more resistant to extinction than behaviours maintained under a continuous schedule of reinforcement. The thinner the reinforcement schedule for a certain behaviour, the more resistant to extinction that behaviour is. In other words, the learner presents more responses for less reinforcers under partial schedules when compared to a continuous schedule of reinforcement.
 
According to figure 2 we can see that, in general, a variable ratio schedule produces more responses for a similar or lower number of reinforcers than other partial schedules of reinforcement. In many situations it also seems to produce those responses faster and with little latency from the individual. This information, along with my own personal observations and communication with professionals in the field of animal training, makes me believe that when trainers use the broad term “Variable Schedule of Reinforcement” they usually mean a variable ratio schedule. 

​Figure 2 – Behaviour responses under the most common types of partial schedules of reinforcement (Chance, 2003; Kazdin, 1994; Schunk, 2012).
 

​A variable ratio schedule might elicit the highest response rate, a constant pattern of responses with minimal pauses and the most resistance to extinction. A fixed ratio has a slightly lower response rate, a steady pattern of responses and a resistance to extinction that is dependent on the ratio used. A fixed interval schedule produces a moderate response rate, a long pause in responding after reinforcement followed by gradual acceleration in responding and a resistance to extinction that is dependent on the interval chosen (the longer the interval, the more resistance). A variable interval has a similar response rate, a steady pattern of responses and is more resilient to extinction than a fixed interval schedule. These characteristics of partial schedules of reinforcement are summarised in table 1.
 
Table 1 – Characteristics of the most common types of partial schedules of reinforcement (Wood, Wood & Boyd, 2005).

​With all these different types of schedules, each with different characteristics you might be wondering: “Do I need to master all of these principles to successfully train my pet at home?” The quick and simple answer is “No, you don’t”. For most animal training situations, a continuous schedule of reinforcement will be a simple, easy and effective tool that will yield the results you want.
 
Doing a training session with your dog in which you ask for behaviours on cue when the dog is in front of you (sit, down, stand, shake, play dead) could be very well maintained using a continuous schedule. A continuous schedule of reinforcement would be an efficient and easy approach and it would allow you to change the cue or stop a behaviour easily (faster extinction) if you change your mind about a given behaviour later. One could argue that a variable ratio schedule would possibly produce more responses with less reinforcement, and a higher resistance to extinction for these behaviours. One of the disadvantages of this option would be the possibility of a ratio strain (post-reinforcement pauses or decrease in responding).
 
Some specific situations might justify the maintenance of a behaviour using partial schedules of reinforcement. For example, when a dog has learned that lying down on a mat in the living room results in reinforcement, the dog’s carer could maintain this behaviour using a variable interval schedule of reinforcement, in which the dog only gets reinforced after varying amounts of time for lying on the mat. Martin and Friedman (2011) offer another example in which partial reinforcement schedules could be helpful. If a trainer wants to train a lion to make several trips to a public viewing window throughout the day, the behaviour should be trained using a continuous schedule to get a high rate of window passes in the early stages. The trainer should then use a variable ratio schedule of reinforcement to maintain the behaviour. They do advise however, that this would require “careful planning to keep the reinforcement rate high enough for the lion to remain engaged in the training”.
 
The process of extinction of a reinforced behaviour means withholding the consequence that reinforces the behaviour and it is usually followed by a decline in the presentation of that behaviour (Chance, 2003). Resistance to extinction can be an advantage or a disadvantage depending on which behaviour we are considering. For example, one could argue that a student paying attention to its teacher would be a behaviour that should be resistant to extinction, and so, a good option to be kept on a partial schedule of reinforcement. On the other hand, a dog that touches a bell to go outside could be kept on a continuous schedule of reinforcement. One of the advantages of this approach would be that, if in the future the dog’s owner decides that she no longer wants the dog to touch the bell, by not reinforcing it anymore, the behaviour could cease to happen relatively fast.
 
While I do believe that for certain specific situations, partial schedules of reinforcement might be helpful, I would like to take a moment to caution against the use of a non-continuous pairing of bridge and backup reinforcer. Many animal trainers call this a “variable schedule of reinforcement” when in practical terms this usually ends up being a continuous reinforcement schedule that weakens the strength and reliability of the bridge. For more information on this topic check my blog post entitled “Blazing clickers – Click and always offer a treat?”.
 
When asked about continuous vs. ratio schedules, Bailey & Bailey (1998) have an interesting general recommendation: “If you do not need a ratio, do not use a ratio. Or, in other words, stick to continuous reinforcement unless there is a good reason to go to a ratio”. They also describe that they have trained and maintained numerous behaviours with a wide variety of animal species using exclusively a continuous schedule of reinforcement. They raise some possible complications when deciding to have a behaviour maintained on a ratio schedule. The example given is of a dog’s sit behaviour being maintained on a FR 2 schedule of reinforcement: “You tell the dog sit – the first response is a bit sloppy, the second one is ok. You click and treat. What have you reinforced? A sloppy response, chained to a good response.”
 
Karen Pryor (2006) also has an interesting view on this topic. She mentions that during the early stages of training a new behaviour you start by using a continuous schedule of reinforcement to get the first few responses. Then, when you decide to improve the behaviour and raise criteria, the animal is put on a variable ratio schedule, because not every response is going to result in reinforcement. This is an interesting point, because the trainer could look at this situation and still read it as a continuous schedule of reinforcement, when in reality the animal is producing responses that are not resulting in reinforcement. At this point in time, only our new “correct responses” will result in reinforcement. From the learners’ point of view the schedule has become variable at this stage. Pryor concludes that when the animal “is meeting the new criterion every time, the reinforcement becomes continuous again.”
 
Pryor (2006) suggests that the situations in which you should deliberately use a variable ratio schedule of reinforcement are: “in raising criteria”, when “building resistance to extinction during shaping” and “for extending duration and distance of a behaviour”. Regarding the situations in which we should not use it, she starts by saying that we should never use a variable ratio schedule purely as “a maintenance tool”. She adds that “behaviours that occur in just the same way with the same level of difficulty each time are better maintained by continuous reinforcement”. Pryor also advises against the use of a variable ratio schedule for maintaining chains, because “failing to reinforce the whole chain at the end of it would inevitably lead to pieces of the chain beginning to extinguish down the road.” Finally, she does not recommend using such a schedule of reinforcement for discrimination problems such as scent, match to sample tasks, or any other training that requires choice between two or more items.
 
In conclusion, there are a few possible schedules of reinforcement that can be effectively used to train and maintain trained behaviours for our pets. Each has its own set of characteristics, but for most training situations, a continuous schedule of reinforcement is a simple, efficient and powerful tool to effectively communicate with our pets. Some specific training situations might be good candidates for partial schedules of reinforcement. In those situations, you should remember to follow each bridge with a backup reinforcer, plan your training well and keep the reinforcement rate high enough for the animal to remain engaged. Have fun with your training!
 


Bailey, B., Bailey, M., (1998). "Clickersolutions Training Articles - Ratios, Schedules - Why And When". Clickersolutions.com. N.p., Accessed 2 February 2018.

Chance, P. (2003). Learning and behavior (5th ed.). Belmont: Thomson Wadsworth.

Kazdin, A. (1994). Behavior modification in applied settings (5th ed.). Belmont: Brooks/Cole Publishing Company.

Martin, S., Friedman, S.G., (2011, November). Blazing clickers. Paper present at Animal Behavior Management Alliance conference, Denver. Co.

Pryor, K. (2006). Reinforce Every Behavior?. Clickertraining.com. Retrieved 2 February 2018, from https://clickertraining.com/node/670

Schunk, D. (2012). Chapter 3: Behaviorism. In Learning theories: An educational perspective (6th ed., pp. 71-116). MA: Pearson.

Wood, S., Wood, E., & Boyd, D. (2005). The world of psychology (5th ed., pp. 180-190). Boston: Allyn & Bacon. Retrieved from http://www.pearsonhighered.com/samplechapter/0205361374.pdf

​Picture: www.morguefile.com