The Use of Secondary Reinforcers in Canine Counterconditioning: A Classical Conditioning Perspective

By Will Bangura, M.S., CAB-ICB, CBCC-KA, CPDT-KA, FDM, FFCP

Introduction

Some professional guidance, including position statements by veterinary behavior organizations, has appropriately emphasized that secondary reinforcers are not required for counterconditioning and may be unnecessary or inadvisable in certain cases, particularly when working with noise-sensitive dogs or novice handlers (Todd, 2025). The present discussion does not dispute this position. Rather, it examines the conditional role secondary reinforcers may play when analyzed strictly through classical conditioning and affective neuroscience, particularly in applied contexts where perfect timing and stimulus control are difficult to achieve.

Counterconditioning is one of the most widely used and empirically supported interventions for fear, anxiety, and aggression in dogs. Its purpose is not to teach a new behavior, but to alter an existing conditioned emotional response by changing the meaning of a stimulus that previously predicted threat, discomfort, or uncertainty. At its core, counterconditioning is a Pavlovian learning process, dependent on stimulus–stimulus associations rather than consequences contingent on behavior.

Within applied canine behavior practice, secondary reinforcers such as clickers or brief verbal markers are sometimes incorporated into counterconditioning protocols. This has led to confusion, particularly when markers are discussed using operant terminology that emphasizes “reinforcing behavior.” Such framing is inappropriate for counterconditioning and risks misrepresenting both the mechanism and the evidence.

This article examines the role of secondary reinforcers in canine counterconditioning strictly through the lens of classical conditioning, focusing on timing, contingency, predictability, and emotional learning. Benefits, limitations, and appropriate applications are discussed with careful attention to what the peer-reviewed literature does and does not support.

Counterconditioning as Pavlovian Learning

In classical counterconditioning, a previously conditioned stimulus is repeatedly paired with an appetitive unconditioned stimulus, leading to a shift in the stimulus’s emotional valence as it comes to predict a positive outcome rather than threat (Pavlov, 1927; Domjan, 2018).

This process depends on two key variables. Temporal contiguity influences how readily associations form, while contingency, defined as the degree to which the conditioned stimulus predicts the unconditioned stimulus, is the primary determinant of associative strength (Rescorla, 1988). When either contiguity or contingency is degraded, emotional learning is weakened or disrupted.

Importantly, counterconditioning does not require the dog to perform any behavior. The dog’s actions during exposure are not reinforced or punished. The only requirement is that the conditioned stimulus is followed, predictably and consistently, by the unconditioned stimulus while the dog remains below threshold.

Secondary Reinforcers as Predictive Signals, Not Behavioral Reinforcers

A secondary reinforcer, also referred to as a conditioned reinforcer, is a stimulus that acquires reinforcing properties through repeated pairing with a primary reinforcer (Skinner, 1938). In operant conditioning, such stimuli function as consequences that increase the probability of behavior. In counterconditioning, however, the same stimuli are used not as behavioral consequences, but as predictive signals within a Pavlovian framework.

When used appropriately, a clicker or verbal marker in counterconditioning does not reinforce behavior. Instead, it functions as a conditioned predictor that marks the precise moment at which the conditioned stimulus is detected and confirms that a positive outcome is forthcoming. In this way, the marker acts as a bridging stimulus, preserving the temporal relationship between perception of the trigger and subsequent delivery of food.

From a Pavlovian perspective, the marker itself becomes part of the stimulus sequence, strengthening contingency detection and reducing ambiguity about what predicts the unconditioned stimulus.

Timing Precision and Emotional Association

Emotional learning is highly sensitive to timing. Research on fear conditioning and extinction demonstrates that alterations in the temporal relationship between conditioned and unconditioned stimuli can significantly influence the strength and specificity of learned associations (LeDoux, 2000; Quirk & Mueller, 2007).

In applied behavior work, immediate delivery of food following exposure to a trigger is often difficult. Handlers may need to manage leashes, create distance, or ensure safety before reinforcement can be delivered. A conditioned marker allows the practitioner or pet parent to preserve temporal contiguity even when food delivery is delayed by a second or two.

Without a marker, delayed reinforcement risks weakening the association or inadvertently pairing the unconditioned stimulus with a different stimulus or internal state. In emotionally sensitive dogs, this can slow progress or create inconsistent outcomes.

Consistency and Timing in Classical Counterconditioning

In counterconditioning, the critical learning process is the formation of a new stimulus–stimulus association, whereby a previously fear-eliciting or aversive stimulus becomes a reliable predictor of a positive outcome. The effectiveness of this process depends heavily on temporal contiguity and contingency, both of which are central principles of classical conditioning (Pavlov, 1927; Rescorla, 1988).

The use of a secondary reinforcer, such as a clicker or brief verbal marker, can support this process by improving the temporal precision with which the conditioned stimulus is paired with the unconditioned stimulus. Rather than functioning as a reinforcer for behavior, the marker serves as a bridging signal that preserves the temporal relationship between perception of the trigger and subsequent delivery of the primary reinforcer.

Empirical research in applied dog training has examined how conditioned markers function in learning, with evidence supporting their role as conditioned reinforcers rather than simple temporal bridges (Feng et al., 2016). Although this study examined task acquisition rather than emotional learning, it provides insight into how signal properties influence associative learning.

From a classical conditioning standpoint, consistent predictors enhance contingency detection and strengthen associative learning, particularly under conditions of uncertainty, where ambiguity about stimulus–outcome relationships would otherwise weaken learning (Rescorla & Wagner, 1972; Pearce & Bouton, 2001)

When no marker is used, successful counterconditioning depends entirely on the handler’s ability to deliver the unconditioned stimulus immediately and consistently following exposure to the conditioned stimulus. Any delay or inconsistency risks weakening the association or pairing the reinforcer with an unintended stimulus.

Emotional Regulation and Safety Signaling

Predictive signals play a central role in emotional regulation. Neuroscience research demonstrates that learned safety signals modulate fear expression by engaging prefrontal–limbic circuits that inhibit threat responding, allowing fear to be suppressed even in contexts where aversive stimuli were previously experienced (Christianson et al., 2012; Sangha et al., 2020).

In counterconditioning, a well-conditioned marker can acquire predictive properties that contribute to emotional regulation by functioning as a reliable predictor of positive outcomes. While it does not replace the unconditioned stimulus, it can help stabilize arousal long enough for reinforcement to occur, particularly in dogs prone to rapid escalation.

This effect is consistent with broader findings on predictability and stress modulation in animals, where clear and reliable cues are often associated with improved welfare outcomes, depending on context and the animal’s ability to anticipate events (Bassett & Buchanan-Smith, 2007).

Situations Where Secondary Reinforcers May Be Unnecessary or Inappropriate

Secondary reinforcers are not mechanistically required for counterconditioning to occur. In highly controlled settings where the conditioned stimulus is immediately followed by an appetitive unconditioned stimulus, classical counterconditioning can proceed effectively without the use of a marker (Domjan, 2018).

In addition, some dogs with extreme sound sensitivities or trauma histories may initially find clickers aversive. In such cases, a softly conditioned verbal marker or a marker-free protocol may be preferable until the dog’s baseline emotional state improves.

For low-intensity triggers and highly food-motivated dogs, simple stimulus–food pairing can be sufficient, provided timing is clean and arousal remains low. However, these conditions are relatively uncommon in clinical behavior cases involving fear or aggression.

Practical Implications for Behavior Modification

When used correctly, secondary reinforcers in counterconditioning function as precision tools, not as behavioral reinforcers. Their value lies in preserving timing, strengthening contingency, and reducing human variability, all of which support reliable emotional learning.

Mischaracterizing markers as operant reinforcers in counterconditioning protocols risks conceptual confusion and weakens scientific credibility. Clear distinction between Pavlovian and operant processes is essential for both ethical practice and effective intervention.

Conclusion

Counterconditioning is a Pavlovian process aimed at changing emotional meaning, not controlling behavior. Secondary reinforcers, when used as predictive signals rather than behavioral consequences, can enhance this process by improving timing precision, consistency, and emotional clarity. While not mandatory, they are often advantageous in real-world applications where perfect timing is difficult to achieve.

Grounding their use in classical conditioning principles ensures conceptual accuracy, aligns with the neuroscience of emotional learning, and supports humane, evidence-based behavior modification.

References

Bassett, L., & Buchanan-Smith, H. M. (2007). Effects of predictability on the welfare of captive animals. Applied Animal Behaviour Science, 102(3–4), 223–245. https://doi.org/10.1016/j.applanim.2006.05.029

Christianson, J. P., Fernando, A. B. P., Kazama, A. M., Jovanovic, T., Ostroff, L. E., & Sangha, S. (2012). Inhibition of fear by learned safety signals. Journal of Neuroscience, 32(41), 14118–14124. https://doi.org/10.1523/JNEUROSCI.3340-12.2012

Domjan, M. (2018). The principles of learning and behavior (7th ed.). Cengage Learning.

Feng, L. C., Howell, T. J., & Bennett, P. C. (2016). How clicker training works: Comparing Reinforcing, Marking, and Bridging Hypotheses. Applied Animal Behaviour Science, 181, 34-40. https://doi.org/10.1016/j.applanim.2016.05.012

LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–184. https://doi.org/10.1146/annurev.neuro.23.1.155

Pavlov, I. P. (1927). Conditioned reflexes. Oxford University Press.

Pearce, J. M., & Bouton, M. E. (2001). Theories of associative learning in animals. Annual Review of Psychology, 52, 111–139. https://doi.org/10.1146/annurev.psych.52.1.111

Quirk, G. J., & Mueller, D. (2007). Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology, 33(1), 56–72. https://doi.org/10.1038/sj.npp.1301555

Rescorla, R. A. (1988). Pavlovian conditioning: It’s not what you think it is. American Psychologist, 43(3), 151–160. https://doi.org/10.1037/0003-066X.43.3.151

Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning. In A. H. Black & W. F. Prokasy (Eds.), Classical conditioning II (pp. 64–99). Appleton-Century-Crofts.

Sangha, S., Diehl, M. M., Bergstrom, H. C., & Drew, M. R. (2020). Know safety, no fear: Neural mechanisms of safety learning and fear suppression. Neuroscience and Biobehavioral Reviews, 108, 218–230. https://doi.org/10.1016/j.neubiorev.2019.11.006

Skinner, B. F. (1938). The behavior of organisms. Appleton-Century.

Todd, Z. (2025, November 10). Should you use a secondary reinforcer (clicker) in counter conditioning? The American Veterinary Society of Animal Behavior. https://avsab.org/should-you-use-a-secondary-reinforcer-clicker-in-counter-conditioning/