Approach–Avoidance Conflict in Dogs: Behavioral Outcomes Under Competing Motivations

2. Theoretical Foundations: Kurt Lewin and Conflict Types

The systematic study of motivational conflict originated with Kurt Lewin’s field theory (1931). Lewin distinguished three fundamental types of conflict:

• Approach‑approach conflict – Between two desirable goals. The dog must choose between two equally attractive options (e.g., two different treats). This is the easiest type to resolve.

• Avoidance‑avoidance conflict – Between two undesirable outcomes. The dog must choose the lesser of two evils (e.g., approach a feared stranger or remain near a loud noise). More difficult to resolve.

• Approach‑avoidance conflict – A single goal that has both attractive and aversive qualities. The dog is simultaneously drawn toward and repelled by the same stimulus. This is the most common and behaviorally most destabilizing form of conflict in dogs.

Lewin proposed that AA conflict generates a force‑field of opposing gradients, typically with avoidance gradients increasing in strength more steeply than approach gradients as the subject moves closer to the goal. This is why a dog that seems calm at a distance may suddenly show fear or aggression as it gets closer.

More recent research using hand movement trajectories has empirically confirmed Lewin’s predictions that AA conflict elicits behavioral oscillations and freezing prior to resolution, reflecting the underlying force‑field dynamics (Enisman et al., 2024).

For a deeper understanding of decision‑making under uncertainty, see cognitive abilities in dogs – why our canine companions are smarter than we think.

3. Neurobiological Substrates – Insights from Animal Models

The neural circuits underlying AA conflict have been studied primarily in rodent models, but these findings provide a useful framework for understanding similar processes in dogs, given the conserved nature of basic motivational circuits across mammals. The following structures have been implicated.

3.1 Prefrontal Cortex – Prelimbic and Infralimbic Subregions

In rodent studies, the prelimbic (PL) region of the medial prefrontal cortex (mPFC) plays a causal role in mediating risky decision‑making under AA conflict. Using a paradigm in which rats retrieved memories of shock‑ and food‑paired cues, researchers identified two distinct conflict phenotypes: rats that continued to press a lever for food (“Pressers”) and rats that completely suppressed food seeking (“Non‑pressers”). Single‑unit recordings showed that increased risk‑taking in Pressers was associated with persistent food‑cue responses in the PL cortex and reduced spontaneous activity in PL glutamatergic neurons during conflict. Activating these neurons in Pressers attenuated food seeking, whereas inhibiting them in Non‑pressers reduced defensive responses and increased approach behavior (Fernandez‑Leon et al., 2021).

The infralimbic cortex (IL), by contrast, is involved in the inhibition of fear responses and the retrieval of extinction memories, helping to resolve conflict by down‑regulating defensive output. While direct evidence in dogs is lacking, the conserved role of the mPFC in executive control across mammals suggests that similar mechanisms may operate in canines.

3.2 Hippocampus, Perirhinal Cortex, and Stimulus Type

For decades, the hippocampus was considered the central structure for arbitrating AA conflict. However, recent optogenetic studies in rodents have refined this view. The ventral hippocampus appears preferentially involved in context‑associated conflict, whereas the perirhinal cortex is critically involved when conflicting motivational values are associated with objects rather than contexts (Dhawan et al., 2023). This distinction is relevant for dog behavior, where conflicts often involve specific objects (food, toys, humans) versus contexts (rooms, outdoor spaces).

3.3 Central Amygdala (CeA)

In a naturalistic foraging task where rats faced a trade‑off between food and a robotic predator, rats displayed two conflict‑like behaviors: diagonal approach and stretched posture. Electrolytic lesions of the CeA significantly reduced these conflict behaviors, accompanied by a decrease in hiding and an increase in approach. Conflict‑like behaviors were not tightly coupled with either approach or avoidance, indicating that the CeA plays a distinct modulatory role (Kimm et al., 2023).

3.4 Ventral Striatum and Thalamus

In a study that tracked conflict strategies across 16 days of training, rats separated into three subgroups: “Timers” (49% – approach early, avoid later), “Avoidance‑preferring” (32%), and “Approach‑preferring” (19%). Timers showed increased activity in the ventral striatum and midline thalamus, consistent with low fear and high behavioral flexibility. Avoidance‑preferring rats showed increased amygdala activity, and Approach‑preferring rats showed decreased activity in the prefrontal cortex (Bravo‑Rivera et al., 2021).

Cautionary note: These findings derive almost exclusively from rodent studies. While the conservation of basic motivational circuitry across mammals suggests plausible parallels in dogs, direct neurobiological evidence in canines remains sparse. The following sections therefore describe behavioral and assessment tools that are directly validated in dogs.

For more on prefrontal cortex function in impulse control, see prefrontal cortex and self‑control in dogs. For the role of dopamine in reward‑based learning, see dopamine and learning in canine neurochemistry.

4. Cognitive Dissonance – A Related Construct

While approach‑avoidance conflict concerns competing motivations toward a stimulus, cognitive dissonance (Festinger, 1957) refers to the psychological discomfort experienced when an individual holds two contradictory beliefs or when behavior is inconsistent with a belief. The question of whether dogs experience cognitive dissonance remains debated. Some researchers suggest that the effort justification paradigm (preferring rewards that required more effort) may be explained by a simpler mechanism such as contrast. However, if non‑human animals engage in dissonance reduction, this would support the original theory as proposed by Festinger over revisions that focus on self‑concept. For applied purposes, trainers are advised to reduce conflict by making outcomes predictable and clear, regardless of the underlying mechanism.

For a discussion of how prediction error and expectation violations drive learning (a related neurocomputational framework), see prediction error in dogs – the core mechanism of learning.

5. Behavioral Outcomes of Approach‑Avoidance Conflict in Dogs

When a dog is in AA conflict, it cannot simultaneously approach and avoid. The behavioral manifestations fall into several categories, many of which have been documented in dog‑specific research.

Important note: Not all ambiguous or mixed behaviors necessarily reflect true motivational conflict. Some may arise from incomplete information, competing learned responses, or rapidly shifting emotional states. The following list describes behaviors that are commonly associated with AA conflict in the literature, but a single behavior should never be taken as definitive evidence.

5.1 Approach‑Withdrawal Oscillations

The dog may move toward the stimulus, then back away, then move forward again – a pattern that reflects the competing motivational gradients. This is often seen in hesitant greetings and has been described in studies of dog behavior under emotional valence contrasts (Albuquerque et al., 2023).

5.2 Freezing and Stretched Posture

Freezing is the complete cessation of movement except for breathing. It is an active coping strategy in which the dog attempts to become “invisible” to a perceived threat. In conflict, freezing occurs when approach and avoidance forces cancel each other out. Stretched posture (a crouched, elongated body with the head extended forward) is a compromised posture that allows the dog to investigate while remaining ready to retreat.

5.3 Displacement Behaviors

Displacement activities are normal behaviors performed out of context, believed to occur when an animal is in conflict about two incompatible desires. Classical examples include sniffing the ground before greeting another dog, scratching, yawning, lip licking, and grooming. As noted by Hetts and Estep, these may be signs of fear or stress – but not necessarily. A dog sniffing might be avoiding social pressure, not necessarily stressed. Moreover, conflict is inherently somewhat stressful, so displacement behaviors can be understood as low‑level signals of motivational competition. However, their function is not always clear, and caution is warranted when interpreting a single displacement behavior as definitive evidence of emotional distress (see also Horowitz, 2009, for a discussion of over‑interpreting canine body language).

5.4 Ambivalent Body Signals – Simultaneous Approach and Avoidance Indicators

The dog may orient toward the stimulus (approach motivation) while simultaneously showing signs of fear or submission: ears flattened, eyes wide showing the sclera (“whale eye”), tail tucked, body lowered. This configuration – sometimes called “ambivalent posture” – is often misinterpreted as “friendly” because the dog is facing the stimulus; in reality, it indicates high conflict. The term “intentional ambivalence” should be avoided as it implies conscious intent; instead, describe the observable pattern: concurrent approach‑oriented and avoidance‑oriented body signals.

5.5 Redirected Behavior

When approach and avoidance are balanced, the dog may redirect behavior toward a neutral target. For example, a dog in conflict about a stranger may suddenly scratch its own neck, bark at nothing, or even redirect a snap toward the leash. Redirected aggression is well documented in dogs under barrier frustration or high arousal (Vieira de Castro et al., 2020; Flint et al., 2021).

5.6 Suppression and Shutdown

If conflict is chronic and inescapable, some dogs stop exhibiting any behavioral signs of conflict. They may appear calm or compliant while underlying stress remains elevated (masking). This is distinct from genuine resolution of conflict and is associated with chronic stress.

Stress‑related behavioral categories (adapted from the ethogram used in Vieira de Castro et al., 2020):

• Body turn: The dog rotates its body (or head only) to the side, away from the handler, in an attempt to avoid them. Tense or low posture; ears back; tail down.

• Move away: The dog takes one or a few steps away from the handler, attempting to avoid or escape. Tense or low posture; ears back; tail down.

• Crouch: The dog lowers its body (or head only) towards the floor, bending legs and arching its back. Ears back; tail down.

• Lying on side or back: The dog lies down on its side or back with head close to the ground, attempting to avoid confrontation. Legs may be open, exposing the ventral region. Tail still and curled between the legs.

• Paw lift: One forelimb is lifted, usually slowly, and immediately returned to rest. Dog is in a tense or low posture.

For more on how learning history and emotional state interact, see learned behavior vs. emotional response in dogs.

6. Common Dog Behavior Phenotypes Involving AA Conflict

Below is a summary of typical conflict contexts, the competing motivations, and observable signs (CMS‑ready bullet structure).

Leash reactivity

• Approach motivation: Desire to greet or investigate another dog

• Avoidance motivation: Fear of the dog, pain from leash pressure, or restraint‑related frustration or aversion

• Observable signs: Lunging, barking, stiff body, ears back, tail tucked concurrently

Resource guarding with ambivalence

• Approach motivation: Dog wants to consume resource

• Avoidance motivation: Dog fears human approach

• Observable signs: Dog eats with whale eye, ears back, body tense; stops eating if human moves

Hesitant greeting

• Approach motivation: Dog wants to sniff stranger

• Avoidance motivation: Dog fears the stranger (posture, direct eye contact, reaching)

• Observable signs: Approaches with stretched posture, tail low, ears back; retreats or shows displacement (sniffing, yawning)

Barrier frustration (fence/leash)

• Approach motivation: Dog wants to reach stimuli on other side

• Avoidance motivation: Restraint‑related frustration or aversion; possible fear of conflict outcome

• Observable signs: Frantic running, whining, barking, redirected aggression toward barrier

Touch sensitivity (vet/grooming)

• Approach motivation: Dog has learned to tolerate procedure (history of reinforcement)

• Avoidance motivation: Dog fears or experiences pain

• Observable signs: Lying on side (appeasement) with tense muscles, paw lift, whale eye, warning growl

For a deeper discussion of barrier frustration, see aversive training methods – neurological effects in dogs.

7. Assessment of Approach‑Avoidance Conflict in Dogs

7.1 C‑BARQ (Canine Behavioral Assessment and Research Questionnaire)

The C‑BARQ is a standardized, validated questionnaire developed by Hsu and Serpell (2003). It provides numerical scores for 14 behavior categories, including stranger‑directed aggression, stranger‑directed fear, dog‑directed fear, and excitability. Stranger‑directed fear, for example, is defined as “fearful or wary responses when approached by strangers” – a classic AA conflict construct. The C‑BARQ has been used in large samples of dogs across many breeds and is available for clinical and research use. Researchers should consult the original validation literature for specific sample sizes and psychometric properties.

7.2 Approach‑Avoidance Laboratory Tests for Dogs

Albuquerque et al. (2023) presented dogs with situations of different emotional valences (positive, negative, and neutral) and scored behavioral responses. Conflict behaviors (e.g., stretched posture, displacement, ambivalent signals) were more frequent in ambiguous or mixed‑valence conditions. Similarly, Flint et al. (2021) exposed dogs to common household noises and scored stress‑related behaviors, many of which reflect AA conflict (approach to investigate vs. avoidance of the sound source).

7.3 The Strange Situation Procedure (SSP)

The SSP, originally developed for human attachment research and adapted for dogs, induces mild approach‑avoidance conflict by introducing a stranger and separating the dog from the owner. The dog must balance the motivation to explore the novel environment and person against the motivation to remain near the owner. Reunion behaviors after separation are particularly diagnostic of conflict resolution style.

For more on the SSP and attachment, see attachment styles in dogs – secure, avoidant, and ambivalent.

8. Practical Implications for Trainers and Owners

8.1 Recognize Conflict Before It Escalates

Learn the ethogram of conflict behaviors: stretched posture, freezing, paw lift, whale eye, displacement behaviors (sniffing, yawning, scratching), approach‑withdrawal oscillations. When these are observed, the dog is not “being stubborn” – it is likely in conflict. Stop escalating pressure.

8.2 Reduce Approach Motivation or Avoidance Motivation – Not Both

To resolve AA conflict, either reduce the perceived threat (avoidance motivation) or reduce the attractiveness of the goal (approach motivation). Do not punish the dog for showing conflict behavior. For example:

• A dog in conflict about a stranger: increase distance (reduces threat). Do not force the dog to approach.

• A dog conflicted about leaving a resource: add distance or offer an alternative resource of higher value. Do not use physical force.

8.3 Use Systematic Desensitization and Counter‑Conditioning

Desensitization gradually increases exposure to the trigger while keeping the dog below the conflict threshold. Counter‑conditioning changes the dog’s emotional response from negative to positive by pairing the trigger with a high‑value reward. This directly reduces avoidance motivation by reducing fear.

8.4 Address Masking and Suppression

If a dog no longer shows conflict behavior but you suspect the underlying conflict remains (e.g., the dog freezes into a down‑stay during triggers), be cautious. True resolution changes the internal state; masking does not. Use choice tests (the “consent test”) to assess whether the dog will voluntarily re‑engage.

8.5 Manage the Environment to Reduce Conflict

Prevent rehearsal of problematic AA behaviors by controlling distance to triggers, using visual barriers, and avoiding over‑threshold exposure. Management creates safety and allows the nervous system to down‑regulate.

8.6 Avoid Punishment of Conflict Behaviors

Punishing a dog for growling, snapping, or freezing may suppress the behavior but does not resolve the underlying conflict (Horowitz, 2009; Vieira de Castro et al., 2020). In fact, it may increase the risk of suppressed warning signals and subsequent escalation without warning. Punishment is particularly problematic in AA conflict because it adds aversive consequences to a situation already characterized by competing valences, potentially intensifying avoidance.

8.7 Consider Social Conflict – Witnessing Human Disputes

Rial et al. (2025) found that dogs witnessing a human conflict between their caregivers showed increased stress and displacement behaviors (yawning, lip licking, freezing) and a decrease in relaxed behaviors. This demonstrates that social conflict between owners can itself become a source of AA conflict for the dog (wanting to intervene vs. avoiding a threatening situation). Owners are advised to minimize overt human conflict in the presence of the dog and to provide a safe space.

For more on when and how to use extinction in conflict resolution, see extinction in dog behavior – why learned behavior disappears and returns. For a discussion of how conflict relates to frustration, see the neurobiology of frustration in dogs.

9. Summary of Approach‑Avoidance Conflict in Dogs

Definition

• Simultaneous attraction to and repulsion from the same goal stimulus

• Caused by competing approach and avoidance motivations

• Common in everyday dog training and behavior

Types of Conflict (Lewin)

• Approach‑approach: two desirable goals (easiest to resolve)

• Avoidance‑avoidance: two undesirable outcomes (moderately difficult)

• Approach‑avoidance: one goal with both attractive and aversive qualities (most common in dogs)

Neurobiological Substrates (Rodent Models)

• Prelimbic prefrontal cortex (PL) – mediates risk‑taking; individual variability

• Infralimbic prefrontal cortex (IL) – inhibition of fear; extinction retrieval

• Ventral hippocampus – context‑associated conflict

• Perirhinal cortex – object‑associated conflict

• Central amygdala (CeA) – modulates conflict behaviors

• Ventral striatum and midline thalamus – flexible conflict resolution (“Timers”)

• Note: Direct evidence in dogs is sparse; these are plausible mechanisms based on conserved mammalian circuitry

Behavioral Outcomes

• Approach‑withdrawal oscillations

• Freezing

• Stretched posture (crouched, elongated body, head extended)

• Displacement behaviors (sniffing, yawning, scratching, grooming) – not always stress; function often ambiguous

• Ambivalent body signals: whale eye, ears back, tail tucked while orienting toward trigger

• Redirected behaviors (snapping at leash, scratching self)

• Suppression / masking – no visible signs; internal conflict remains

Assessment Tools (Dog‑Specific)

• C‑BARQ (Hsu & Serpell, 2003) – validated questionnaire

• Approach‑avoidance tests (Albuquerque et al., 2023; Flint et al., 2021)

• Strange Situation Procedure (SSP)

• Ethogram‑based behavioral coding (adapted from Vieira de Castro et al., 2020)

Training and Management Strategies

• Recognize early conflict signs; do not ignore or punish

• Reduce threat (increase distance, safety) or reduce goal attractiveness

• Systematic desensitization (gradual exposure below threshold)

• Counter‑conditioning (pair trigger with high‑value reward)

• Management (visual barriers, controlled distance)

• Avoid punishment – may suppress without resolving conflict; may increase escalation risk

• Assess for masking – use consent tests

• Consider social conflict sources (owner disputes)

Key Insights (Takeaways)

• Approach‑avoidance conflict occurs when a dog is simultaneously drawn toward and repelled by the same stimulus. This is a common, normal phenomenon, not a sign of “mixed signals” or inconsistency. However, not every ambiguous behavior reflects true motivational conflict; some may arise from incomplete information, competing learned responses, or rapidly shifting emotional states.

• Conflicts can be resolved by reducing the strength of avoidance motivation (e.g., by increasing distance or counter‑conditioning fear) or by reducing the strength of approach motivation (e.g., by offering an alternative reward). Do not attempt to suppress behavior through punishment.

• Displacement behaviors (sniffing, yawning, scratching) are classic correlates of conflict, but their interpretation requires caution. They do not always indicate stress; they may simply reflect motivational competition. Single displacement behaviors are not diagnostic of emotional state.

• The neurobiology of AA conflict is well studied in rodents, revealing distinct roles for prefrontal cortex, hippocampus, amygdala, and striatum. While direct evidence in dogs is lacking, these findings provide a plausible framework for understanding canine conflict.

• The C‑BARQ is a validated, standardized questionnaire for assessing conflict‑related behaviors. It can help quantify problem severity and track progress.

• Ambivalent body signals (whale eye, ears back, tail tucked while orienting toward the trigger) are indicators of conflict, not conscious intent. Avoid terms like “intentional.”

• Masking (suppression of conflict signs) is not resolution. A dog that has stopped growling but still shows whale eye or muscle tension is still in conflict.

• The goal of behavior modification is to change the underlying emotional state, not just the observable behavior. True resolution requires reducing fear or increasing the positive value of the situation.

Conclusion

Approach‑avoidance conflict is a universal, often unavoidable feature of decision‑making in dogs. Every time a dog wants something that also frightens it, or wants to escape something that also offers a potential reward, the brain’s motivational systems can enter a state of competition. The behavioral outcome – freezing, stretched posture, displacement, withdrawal, or sudden aggression – depends on the relative strengths of the competing drives, as well as on learning history, context, and the individual’s neurobiological capacity to resolve conflict.

For trainers and owners, the most important lesson is that conflict behaviors are not “mixed signals” to be ignored or punished. They may indicate an underlying motivational imbalance. Punishing the behavior may suppress the signal but does not resolve the underlying conflict, and may increase the risk of sudden escalation without warning. The scientifically sound approach is to reduce the perceived threat (e.g., by increasing distance, using systematic desensitization) or reduce the attractiveness of the goal (e.g., by offering an alternative reward), thereby allowing the dog to resolve the conflict without welfare compromise.

Modern behavior science provides precise tools for assessing AA conflict in dogs: the C‑BARQ questionnaire, standardized approach‑avoidance tests, and ethograms of conflict‑related behaviors. Rodent neurobiology suggests that individual differences in conflict resolution explain why some dogs are “risk‑takers” and others are “avoiders.” Ultimately, the goal is not to eliminate conflict – that is impossible – but to teach the dog that safe, predictable, low‑conflict outcomes are available. When the world becomes less threatening and more rewarding, the approach gradient steepens and the avoidance gradient flattens. The likelihood of hesitation, freezing, or redirected behavior decreases, facilitating approach behavior.

Effective training changes the available options and their perceived outcomes so that the safe, prosocial behavior is also the rewarding one – rather than forcing the dog to choose under conflict.

References

Albuquerque, N., et al. (2023). How do dogs behave when presented with situations of different emotional valences? Animals, *13*(6), 1027. https://doi.org/10.3390/ani13061027

Bravo‑Rivera, H., et al. (2021). Characterizing different strategies for resolving approach‑avoidance conflict. Frontiers in Neuroscience, *15*, 608922. https://doi.org/10.3389/fnins.2021.608922

Dhawan, S. S., et al. (2023). Male rodent perirhinal cortex, but not ventral hippocampus, inhibition induces approach bias under object‑based approach‑avoidance conflict. eLife, *12*, e81467. https://doi.org/10.7554/eLife.81467

Enisman, M., et al. (2024). Hand movement trajectories illustrate the mechanism underlying Kurt Lewin‘s distinction between approach‑approach and avoidance‑avoidance motivational conflicts. Journal of Personality and Social Psychology, *127*(2), 239–258. https://doi.org/10.1037/pspa0000392

Fernandez‑Leon, J. A., et al. (2021). Neural correlates and determinants of approach‑avoidance conflict in the prelimbic prefrontal cortex. eLife, *10*, e74950. https://doi.org/10.7554/eLife.74950

Festinger, L. (1957). A theory of cognitive dissonance. Stanford University Press.

Flint, H. E., et al. (2021). Stress‑related behaviors in companion dogs exposed to common household noises. Frontiers in Veterinary Science, *8*, 760845. https://doi.org/10.3389/fvets.2021.760845

Horowitz, A. (2009). Disambiguating the “guilty look”: Salient prompts to a familiar dog behaviour. Behavioural Processes, *81*(3), 447–452. https://doi.org/10.1016/j.beproc.2009.02.014

Hsu, Y., & Serpell, J. A. (2003). Development and validation of a questionnaire for measuring behavior and temperament traits in pet dogs. Journal of the American Veterinary Medical Association, *223*(9), 1293–1300. https://doi.org/10.2460/javma.2003.223.1293

Kimm, S., et al. (2023). The central amygdala modulates distinctive conflict‑like behaviors in a naturalistic foraging task. Frontiers in Behavioral Neuroscience, *17*, 1212884. https://doi.org/10.3389/fnbeh.2023.1212884

Mariti, C., et al. (2017). Analysis of the intraspecific visual communication of adult domestic dogs (Canis familiaris): A third way of interpreting “appeasement signals”? Applied Animal Behaviour Science, *193*, 48–55. https://doi.org/10.1016/j.applanim.2017.03.003

Rial, M., et al. (2025). Affiliative and stress behaviors of domestic dogs when they witness a human conflict between their caregivers. Applied Animal Behaviour Science, *283*, 106450. https://doi.org/10.1016/j.applanim.2024.106450

Vieira de Castro, A. C., Fuchs, D., Morello, G. M., Pastur, S., de Sousa, L., & Olsson, I. A. S. (2020). Does training method matter? Evidence for the negative impact of aversive‑based methods on companion dog welfare. PLoS ONE, *15*(12), e0225023. https://doi.org/10.1371/journal.pone.0225023