Can You Reinforce Dog Anxiety & Fear?
As a Certified Dog Behaviorist and Trainer, I want to debunk a common dog myth that a dog’s anxiety and fear can be reinforced. It can not. Only behavior can be reinforced.
So when you hear a dog trainer proclaim, “don’t comfort your dog when they are afraid, it will make them more fearful and will reinforce their fear.” Or “don’t hold your dog when they are scared.” This is nonsense. Ignore that advice as it’s incorrect.
Comfort your dog when they are scared just as you would your 2-year-old child or any other family or friend. For example, if your dog is scared of thunder or fireworks, comfort them to soothe their fears.
When your dog is scared, see what would be helpful for your dog in that moment. It’s compassionate of you to pet your dog, speak to them in soft loving tones, and hold them in a way that’s pleasing to your unique dog.
Here are some other dog myths such as dogs are pack animals, Alpha, or carnivores.
Influencing Is Not The Same As Reinforcement
Reinforcement and punishment are used by behaviorists to describe whether a behavior gets stronger or weaker as a consequence of the presence of a stimulus.
We might influence a dog’s emotion by adding or subtracting a reinforcement or punishment. But the emotion itself does not get reinforced or punished, the behavior does.
Influencing an innate process is subjective and not objective, observable, measurable, and predictable.
Operant Conditioning
Operant conditioning is contingent upon behavior. In the world of behavioral science, behaviorists focus on observable behaviors and don’t/can’t measure a dog’s inner state of thought, emotion, feeling, state of mind, or digestion. Additionally, the four quadrants of operant conditioning1 (Skinner and Thorndike) do not apply to emotion and feelings, but only measurable behavior change.
Even before Skinner’s time, his predecessor, Edward Thorndike’s (1898) law of effect2 also stated that a behavior must be present and not internal processes. Although internal processes are always present, they are subjective and not observable.
However, that’s not to say a dog is not feeling emotion while it is performing a behavior. Because as sentient beings, with limbic systems just like you and I, they are. It simply means that only behavior is capable of reinforcement and punishment, not the underlying inner state, thought, or emotion of an animal.
For example, you cue your dog to sit and reinforce them for sitting. Therefore your dog’s behavior of sitting gets reinforced and sitting behavior is more likely to occur in the future. However, your dog’s inner thought process and/or emotion(s) that they were experiencing when they sat will not be reinforced as it is impossible to reinforce or punish something we can’t see and don’t know.
Classical Conditioning
On the opposite side of the spectrum, some people might mistakenly attribute classical conditioning, AKA respondent conditioning or Pavlovian conditioning, to reinforcing a feeling because respondent conditioning is not contingent upon any behavior and can change a dog’s emotion.
Classical conditioning may elicit a change in a dog’s emotion but the emotion is not subject to reinforcement.
In classical conditioning, you may pair a stimulus to elicit a different conditioned emotional response (CER) from a dog, however, don’t get confused. You don’t use respondent conditioning or counter-conditioning to reinforce a dog’s emotion or behavior, you elicit new or changed emotion through pairing via presenting stimuli together quickly in-time producing a conditioned response.
Typically, pairing a conditioned stimulus (CS) then an unconditioned stimulus (US) produces a conditioned response (CR), (CS >US = CR). Where the CS acts as a signal or cue for the US. As you probably already assumed the US is typically a high-value food item.
As mentioned earlier, classical conditioning procedures and protocols such as desensitizing and counterconditioning (D/CC) are not contingent upon a dog’s behavior at all. The behavior is irrelevant in the sense you are not asking your dog to do anything. You are simply pairing stimuli in a quick time frame to form different neurochemical outputs and stimuli/cue associations.
Hence, paired conditioned stimuli signal the unconditioned stimulus and thus change a dog’s conditioned emotional response. However, remember that an emotion is not said to be reinforced because it is not an operant behavior, observable or measurable, and therefore not capable of being reinforced.
A CER that is elicited through classical conditioning is called a conditioned reflex that occurs because of prior conditioning and a dog’s biological history. But it doesn’t reinforce anything. Keep reading for real-life examples.
Here Are A Few Other Important Reasons Why An Emotion Can’t Be Reinforced
For starters, if a dog is scared (shaking, hiding, lowering their posture, barking, lunging, growling, etc.) the dog’s sympathetic nervous system is engaged and activated. When this occurs, a dog’s brain is not capable of thinking and learning and the dog can not be reinforced for any behavior.
Not sure? Try asking your dog to sit when they are lunging to attack a cat or squirrel. Or try recalling your dog when your dog is fighting with another dog. Not going to happen. It’s impossible to reinforce something when a dog is not physically, mentally, or biologically capable of learning.
The applied behavior analysis definition (or one of them) for positive reinforcement dog training definition is that reinforcement is based on the action of the dog–their behavior–and it produces an increase in the strength of behavior due to its consequences. And, in order to be considered reinforcement (or punishment) the experience must possess these three characteristics to qualify as a reinforcement (or punishment).
- The behavior must have a consequence
- The behavior must increase/decrease in strength
- The increase/decrease in the strength of the behavior must be the result of the consequence
(Chance 7th addition Learning and behavior page 133)
If you substituted the word “fear/anxiety/emotion” for behavior in the above section, it wouldn’t qualify for reinforcement (or punishment)
Again, a cue/stimulus predicts the likelihood of an action/behavior and not emotion. As mentioned, the emotion of a dog fluctuates internally and is subjective; there is no way for us to definitively see or know a dog’s emotion or thoughts.
What Does Body Language Say About Dog Anxiety?
Dog behavior and body language tell us a lot about a dog’s inner state, but not the exact emotion or cognitive thought associated at every moment in time.
We can’t reinforce what we don’t know or can’t observe. Because we can’t say for sure what a dog is feeling and thinking in their inner world, we can’t put it on a reinforcement schedule.
Other definitions also call reinforcement an operant behavior and not an emotion because as mentioned, emotion and feelings are ephemeral and impossible to measure accurately, reliably, and predictably.
Still with me? To make matters more confusing, we can measure different levels of oxytocin or cortisol, the precursors and chemicals associated with emotions such as love and stress, however measuring a chemical associated with emotion does not dictate, predict, punish or reinforce the emotion nor is there any way to measure the emotional level in any being as all sentient beings feel things very differently.
For example, ask an empath how they feel emotion vs. a non-empath. Even though the chemicals associated with each emotion might be similar the felt experience differs across all beings.3
To add more complexity to the situation, one could positive reinforce a dog while they are happy and having fun to display “typical anxiety or nervous” behaviors that are associated with fear, anxiety, and stress such as,
- Lowering of a profile
- Hiding
- Moving away from an object
- Mouth tightly shut
- Showing teeth
- Growling
- Lunging
- Barking
- Tail tucked
- Excess licking (tongue flicking)
- Grass eating
- Poop eating
And there are even cases where dogs (and other animals) learn to display these behaviors to get attention or what they want.
However, if we tried to measure chemical levels to detect an emotion or feeling at the time of these “typical anxiety or nervous” behaviors, those learned behaviors wouldn’t predict or show the stereotypical underlying emotion that is thought to be attached/associated with a dog displaying that type of body language. This is another example of how emotion and feelings can’t be assumed or reinforced simply by observing a dog’s behavior or body language.
Takeaways About Dog Anxiety
Dog behavior and body language tell us a lot about a dog’s inner state and emotions. Behaviorists and trainers use this information to guide their behavior modification and training protocols. However, correlation does not make something subject to reinforcement (or punishment) as it pertains to the science of learning and conditioning.
Emotion, feelings, thoughts, and other inner states are not subject to reinforcement and/or punishment. Comfort your dog when they are scared and/or fearful. Then work desensitization/counterconditioning (D/CC) with them to change their conditioned emotional response (CER) to the scary stimulus and you both will be happier and will strengthen your bond as you better understand how your dog communicates and the laws of learning.
The science of applied behavior analysis or “behaviorism is primarily concerned with observable behavior, as opposed to internal events like thinking and emotion.”4
Sources:
- https://en.wikipedia.org/wiki/Law_of_effect
- https://en.wikipedia.org/wiki/Operant_conditioning
- https://www.healthline.com/health/what-is-an-empath#_noHeaderPrefixedContent
- https://www.simplypsychology.org/operant-conditioning.html
