Behavioral Synchronization Between Dogs and Humans: What Studies Show

1. What Behavioral Synchronization Is

1.1 The Same Thing, at the Same Time, in the Same Place

Behavioral synchronization is the temporal coordination of behavior between individuals. In the definition used by the research group that has produced most of the dog studies, being synchronized means doing the same thing, at the same time and in the same place as others (Duranton & Gaunet, 2016). The phenomenon is widespread in animals that live in groups or pairs, and their review discusses its adaptive value within species before proposing that it should also be studied between species that live together.

The definition is deliberately behavioral. It says nothing about what the animal intends, feels or understands. Two individuals are synchronized when their behavior is coordinated in time more closely than would be expected if each acted independently. Whether that coordination reflects attention, attachment, habit, a shared trigger in the environment or a physical constraint such as a leash is a separate question, and much of this article is about keeping those questions apart.

1.2 Three Components and an Extra One

The experimental papers distinguish three components (Duranton et al., 2017). Temporal synchrony means switching actions at the same moment, whether or not the actions are identical. Location synchrony means being in the same place at the same time, for example staying close to the partner and moving to the same part of a room. Activity synchrony means showing the same behavior at the same time, such as walking while the partner walks and standing still while the partner stands. Later studies added body orientation, the extent to which a dog faces the same direction as the person (Wanser et al., 2021), and gaze direction, the extent to which a dog looks toward the same area as the person (Duranton et al., 2017).

These components are not interchangeable: a dog can stay close without matching pace, or match activity changes from a distance. The general problem of turning everyday descriptions into measurable variables applies here in full (how behavior is operationalized for research).

1.3 What Synchronization Is Not

Several neighboring phenomena are often merged with synchronization. Emotional contagion concerns the transfer of an emotional state from one individual to another (what is known about emotional contagion from humans to dogs). Synchronization concerns the coordination of observable behavior and can occur without any shared emotion. Social referencing means using another individual's reaction to evaluate an ambiguous situation (how dogs use their owners' reactions in uncertain situations). Imitation means copying the form of an action, often after a delay. Unintended cueing, the Clever Hans effect, means that a person's small movements guide a dog's response without anyone intending it.

Physiological covariation is a further category. When the cortisol or heart rate values of dogs and owners are correlated, this is often called "synchronization" in the literature, and the title of one well-known study uses exactly that word (Sundman et al., 2019). But a correlation between two values measured over weeks or months is a different kind of observation from a dog stopping three seconds after its owner stops. This article treats physiological covariation separately and as the weaker category, for reasons explained in Section 7.

2. How Synchronization Is Measured

2.1 Standardized Walks

Most behavioral studies use a simple structure. A person, usually the owner, follows a fixed script of standing still, walking at normal pace and walking fast, often for periods of 20 to 30 seconds, while the dog moves freely. The person is instructed not to talk to, touch or look at the dog, so that the dog's behavior cannot be explained by direct requests. Video or tracking data then show how much time the dog spent moving while the person moved, standing while the person stood, and close to the person.

The first indoor study used five conditions of 30 seconds each after ten minutes of exploration (Duranton et al., 2017), the puppy study 20-second phases (Duranton et al., 2022), an enclosure study 30 seconds of walking followed by 30 seconds of standing (Heurlin et al., 2024), and the child study walks along color-coded tape lines (Wanser et al., 2021).

2.2 Tracking Distance and Speed

More recent studies use position data. In a comparison of walking alone with the owner versus walking in a group, the dogs and humans wore GPS devices outdoors and were measured during standing, slow walking and fast walking (Lamontagne et al., 2023). A later study fitted harnesses with GPS receivers corrected by a fixed base station, which gave position data with centimeter-level precision every quarter of a second, and calculated the distance between dog and partner, the difference in their speed and their body alignment (Lamontagne et al., 2025). These measures are more precise than coding video by eye, but they still measure coordination of movement, not its cause.

2.3 Why Measuring It Is Hard

Synchronization is easy to see and surprisingly hard to demonstrate. Four problems recur across the literature.

The first is the shared environment. Two individuals passing the same gate and the same distraction will show coordinated behavior even if neither attends to the other, so correlated movement records do not show a response unless common triggers are ruled out.

The second is the leash. A dog on a short leash cannot do much else than follow the person's pace and direction. The first indoor study was designed explicitly to remove this problem: its authors noted that the earlier dog–human walking studies had observed most dogs on leash, and they therefore tested dogs moving freely (Duranton et al., 2017). Any synchrony measured on leash mixes the dog's behavior with mechanics.

The third is the baseline. Dogs stand and walk for a good part of any observation, so some overlap with a person's standing and walking would occur by chance. Studies handle this differently, for example by comparing behavior across conditions in which the person does different things (Duranton et al., 2017) or by testing whether overlap exceeds chance levels (Wanser et al., 2021). A percentage of "synchronized time" without a comparison is difficult to interpret.

The fourth is the time scale. Most test phases last seconds to a few minutes, in rooms or enclosures. Whether the same pattern holds over a normal walk of half an hour, with other dogs, smells and people around, is largely untested.

3. Dogs Moving With Their Owners: The Core Experiments

3.1 Indoors, Off Leash

The study that started the experimental program tested 48 pet dogs, 24 molossoid and 24 shepherd dogs with equal numbers of males and females, aged one to eleven years, in an unfamiliar empty room of 23 square meters at a veterinary school in France (Duranton et al., 2017). The owners followed a sequence of conditions in which they stood still, walked, or switched between the two, without speaking to their dogs.

The results were consistent across measures. The dogs spent on average about 79 percent of the testing time close to their owners, and this proportion did not depend on condition, breed group, sex or age. The time dogs spent in the center of the room was correlated with the time their owners spent there (r = 0.51), the time dogs spent stationary with the owners' stationary time (r = 0.60), and the time dogs spent gazing toward the front of the room with the owners' gaze toward the front (r = 0.47). When owners switched from standing to walking or the reverse, dogs switched on average 3.4 seconds later (Duranton et al., 2017).

Shepherd dogs switched faster than molossoid dogs (about 2.2 versus 4.7 seconds), which the authors attributed to body weight rather than social responsiveness. They also discussed alternatives: proximity seeking in a stressful room, which they argued against because the dogs looked relaxed while conceding that physiological arousal was not measured, and the fact that owners decide when and where walks go in daily life, so following a decision maker could produce the same pattern (Duranton et al., 2017).

3.2 Outdoors

A follow-up study observed pet dogs moving freely in a familiar outdoor area (Duranton et al., 2018). The dogs stayed close to their owners, moved when their owners moved and at the same pace, and stood still when their owners stood still. The abstract does not report the sample size, so the size of this study cannot be stated here. The main contribution is that the indoor pattern was not limited to a small unfamiliar room.

3.3 Walking With One Person or a Group

Using GPS tracking, a study with 30 shepherd dogs and 30 molossoid dogs compared two situations: the owner walking alone with the dog, or the owner walking together with two familiar people (Lamontagne et al., 2023). The dogs synchronized with their owner at the same level in both situations. The group did change something else: dogs looked at their owners more often when the owner was part of a group. The authors interpreted this as the same level of locomotor coordination combined with a difference in how often the dogs looked at their owner; describing this as “checking in” is already one possible interpretation of the gaze behavior. According to a later summary by the same group, the level of obedience training did not affect the dogs' synchronization with humans (Lamontagne et al., 2025).

3.4 A Familiar Dog or a Familiar Person

Dogs also synchronize with other dogs. A study summarized by the same research group found that familiar dogs stayed closer to each other than unfamiliar dogs during off-leash walks (Duranton, 2020, as summarized by Lamontagne et al., 2025). The first direct comparison between the two partners observed 20 adult dogs from one agility club, each moving around a 25 by 25 meter enclosed park, once with a familiar dog and once with a familiar person from the same household (Lamontagne et al., 2025).

The dogs stayed closer to the familiar dog than to the familiar person. They also matched the familiar dog's speed more closely, but only if they did not live with other dogs themselves. Body alignment and gaze did not differ between the two conditions. The authors explained the closer speed match with another dog mainly physically: two quadrupeds move in similar ways, which makes matching easier. They also noted the limits of their sample: dogs used to training and to frequent contact with other dogs, half of them herding breeds (Lamontagne et al., 2025).

Synchronization with humans is real in these data, but not stronger than with a familiar dog. Humans are one kind of social partner among others.

4. Synchrony and Affiliation

4.1 Do Dogs Prefer People Who Synchronize With Them?

In humans, synchrony and affiliation are linked in both directions: people who like each other tend to move more in synchrony, and moving in synchrony increases liking. The Duranton and Gaunet review states this as the starting point of the dog research (Duranton & Gaunet, 2018). One experiment tested the second direction in dogs. Pet dogs met two unfamiliar people: one walked in synchrony with the dog, the other walked randomly. In a subsequent preference test, molossoid dogs showed a clear preference for the person who had synchronized with them. Shepherd dogs did not (Duranton et al., 2019a).

The authors concluded that dogs show greater affiliation with humans who mimic their walking and that genetic selection modulates this propensity. That conclusion rests on one of two breed groups. The sample size and preference measure are not given in the abstract and are not reported here. The design was experimental, which is a strength, but it is a single study without a replication found for this review.

4.2 Shelter Dogs as a Study Population

Familiarity and bond are expected to matter if synchrony reflects affiliation. A study tested 30 shelter dogs with their principal caregivers in the dogs' regular walking area, while the caregivers stood still, walked normally or walked fast (Duranton et al., 2019b). The shelter dogs synchronized their locomotor activity with their caregivers, but less strongly than pet dogs had done with their owners in an earlier study, and they kept greater distances. The authors proposed that the strength of the social bond explains most of the difference.

Two limits apply. The comparison with pet dogs was made across studies, and shelter dogs differ from pet dogs in history and living conditions, not only in their bond. Synchronization is present but weaker in this population; that a closer bond would raise it was not tested.

4.3 Puppies

If synchronization with humans depends on a bond, it should develop with the bond. A study tested 29 retriever puppies from five litters at a guide-dog breeding center, each once at one month and once at two months of age, with their familiar caregiver and with an unfamiliar experimenter (Duranton et al., 2022). The room measured about 4 by 2.4 meters, and each test lasted 80 seconds.

At both ages, puppies moved more when the experimenter walked than when she stood still: at one month about 19 versus 12 seconds, at two months about 29 versus 12 seconds. Two-month-old puppies, but not one-month-old puppies, also moved fast more when the experimenter walked fast. Familiarity made no difference: puppies synchronized with the unfamiliar experimenter as much as with their caregiver and stayed equally close to both (Duranton et al., 2022).

The authors suggested that the basic tendency may have been selected during domestication, while familiarity effects develop later; these puppies did not live in human homes, so their bond with the caregiver may have been weak. One breed type, one facility and very short tests limit the study, but the tendency clearly appears early, before extensive training.

4.4 Breed Groups, Pack-Living Dogs and Wolves

A Swedish study ran three experiments in outdoor enclosures (Heurlin et al., 2024). In the first, 20 companion dogs were tested after three kinds of interaction with their owners: being ignored, being petted and playing. The dogs adjusted their movement to the owner's movement after all three, and the owner's walking also influenced the dogs' exploration and eye contact. In the second, 24 dogs of ancient breeds and 17 solitary hunting dogs were compared with the 20 companion dogs: all groups synchronized their movement with the owner, and human walking increased eye contact and reduced exploration.

The third experiment compared six socialized pack-living wolves with six similarly socialized pack-living dogs. Neither group changed its movement in response to the handler (Heurlin et al., 2024). The authors concluded that synchronization may depend on how much a dog's everyday life is shared with humans. With six animals per group, and with pack-living dogs that differ from companion dogs in rearing as well as daily life, this is a plausible reading rather than a firm result.

4.5 What "Social Glue" Means Here

The research group behind most of these studies describes synchronization as acting as a "social glue" between dogs and humans (Duranton & Gaunet, 2018). That phrase summarizes a hypothesis: that synchrony both reflects and strengthens affiliation. The evidence so far supports parts of it. Synchrony is present in pet dogs, weaker in shelter dogs walked by caregivers and in pack-living dogs, present in puppies regardless of familiarity, and in one study associated with a preference for a synchronizing stranger in one breed group. What has not been shown is that synchronizing with a dog over time improves the relationship, or that a dog's degree of synchrony can be used to assess the quality of its bond. The links to attachment research are plausible but have not been tested directly (how attachment styles in dogs are classified).

5. Children in the Family

Almost all studies tested dogs with adults. One study asked whether family dogs also synchronize with children living in the same household (Wanser et al., 2021). According to the university's summary of the study, 30 children and adolescents aged 8 to 17 took part with their own family dog, and 83 percent of them had a developmental disability. In a large empty room, the children walked along color-coded tape lines in a standardized way while their dog was off leash.

The dogs showed all three measured components above chance: activity synchrony, proximity and orientation (Wanser et al., 2021). The university summary reports that dogs and children were in activity synchrony about 60 percent of the time, that dogs moved during about 73 percent of the time the child moved and stood still during about 41 percent of the time the child stood still, that the dogs were within one meter of the child about 27 percent of the time, and that they faced the same direction about 34 percent of the time. These values were lower than those reported in earlier studies with adults, and a later study summarized the child findings as synchronization "to a lesser extent" (Heurlin et al., 2024).

The researchers pointed out that dogs are more likely to bite children than adults, one reason why dog–child interactions deserve study (what is known about dog bites and children). The study does not link lower synchrony to risk, nor does it show why dogs synchronized less. Most children in the sample had a developmental disability, so transfer to typically developing children is open.

6. Gaze, Blinks and Attention

6.1 Looking Where the Person Looks

In the indoor study, dogs gazed toward the front of the room more when their owners did so (Duranton et al., 2017). In the Swedish enclosures, the owner's walking increased the dogs' eye contact (Heurlin et al., 2024), and in the GPS study, dogs looked at their owners more often when the owner walked in a group (Lamontagne et al., 2023). Gaze is the component closest to social cognition in the narrower sense, because it suggests that dogs monitor the person rather than only following their position. Dogs' sensitivity to human gaze and pointing is well documented in other paradigms (how dogs read human gestures and gaze).

The boundary to social referencing is narrow. An earlier experiment by the same group, summarized in a later paper, found that pet dogs facing an unfamiliar person adjusted their reaction to the owner's behavior when highly familiar with the owner, and the title of the study states that the owner's direction of movement alone was enough (Duranton et al., 2016, as summarized by Duranton et al., 2022). In synchrony research, the dog does what the person does; in referencing research, it uses the person's reaction to evaluate a third thing.

6.2 Eyeblinks

A short report examined the timing of blinks between dogs or cats and humans during brief periods of mutual gaze (Koyasu et al., 2022). Owners blinked more often in the quarter second after their dog blinked, strangers somewhat later, and dogs and cats tended to blink about a second after the human. The study recruited 26 dogs but could analyze only seven dog–human pairs after exclusions. With that sample, the result is a hint that synchrony may extend to small signals during face-to-face contact, not an established finding.

7. Physiological "Synchrony": A Separate and Weaker Category

7.1 Long-Term Stress Hormones

Hair cortisol reflects stress hormone levels over weeks to months (what hair cortisol and other stress measures can show). A Swedish study measured hair cortisol in 58 dog–owner pairs, Shetland sheepdogs and border collies, in summer and in winter (Sundman et al., 2019). Owner and dog values were correlated at both time points. The dogs' activity, measured with an activity collar for a week, and the number of training sessions did not affect the dogs' cortisol. Three of the owners' personality traits, neuroticism, conscientiousness and openness, were associated with the dogs' cortisol, while the dogs' own personality had little effect. The authors suggested that dogs to a great extent mirror their owners' stress level.

That suggestion is an interpretation of a correlation. Dog and owner share a home, a routine, a neighborhood and the same disruptions. A stressful period affects both without either transmitting anything to the other. The asymmetry in the personality associations points in the direction of an owner-to-dog influence, but a correlational design cannot rule out a shared cause.

7.2 Not in All Breed Groups

A follow-up study by the same group tested 24 dogs of ancient breeds and 18 solitary hunting dogs with their owners (Höglin et al., 2021). In neither group was the owners' hair cortisol related to the dogs'. Instead, a relationship factor, the owners' perceived cost of keeping the dog, was associated with the dogs' cortisol. The authors concluded that long-term stress synchronization is probably a trait of breeds selected for close cooperation with humans. Taken together, the two studies show that the covariation is not a universal property of dog–owner pairs.

7.3 Heart Rate Variability Under Owner Stress

Heart rate variability reflects autonomic arousal on a scale of seconds. In a Japanese study that started with 34 dog–owner pairs, of which only 14 could be analyzed because of technical problems with the heart rate recordings, owners went through a modified social stress test in front of an audience on one day and a control condition on another, while the dog, on a loose leash, watched (Katayama et al., 2019). Within each pair, the researchers correlated the time courses of the dog's and the owner's heart rate variability.

The results are narrower than the title, "Emotional contagion from humans to dogs is facilitated by duration of ownership", might suggest. The correlation of one heart rate measure was only slightly and not significantly higher under stress than in the control condition. The owners' heart rate variability did not show a measurable stress effect, and their self-reported anxiety only tended to be higher, which the authors interpreted as a moderate stressor, possibly buffered by the dog's presence. The clearest result was that, in the stress condition, the dog–owner correlation in one heart rate variability measure increased with the duration of ownership (Katayama et al., 2019). The dogs also looked at their owners longer under stress. This is a finding about covariation and ownership duration, not direct evidence that stress was transferred to the dog.

7.4 Heart Rate and Activity During Interaction

A Finnish study measured heart rate variability and physical activity in dogs of cooperative breeds and their owners during resting periods and four interaction tasks: stroking, training, sniffing and playing (Koskela et al., 2024). Of 30 recruited pairs, 25 were analyzed. Across the session, dog and owner heart rate variability were correlated (ρ = 0.53), and so was their activity (ρ = 0.40). The pattern differed between tasks. Heart rate variability was correlated during the resting periods at the start and end, activity during stroking and playing, and neither during training and sniffing.

An important control: when dogs were paired at random with other dogs' owners, the heart rate correlation disappeared (ρ = −0.018), which argues against a mere effect of identical room conditions (Koskela et al., 2024). The authors noted that heart rate variability reflects general arousal rather than specific emotions and that movement affects heart function. Twenty-five pairs of cooperative breeds in one session is a small, selective sample.

7.5 Sleep

Concordance can also be measured over nights. In 56 pairs of military veterans and their service dogs, activity trackers worn for 14 days showed that dog and person were in the same sleep or wake state on average about 79 percent of the time (Leighton et al., 2025). The dogs were more sensitive to the veterans' waking than the reverse. But concordance was not consistently related to measures of the bond or of well-being. This is a specialized population, and the result is a reminder that even high concordance does not by itself indicate a better relationship.

7.6 What the Physiological Studies Can and Cannot Show

A systematic review of physiological co-modulation in human–animal interaction included 37 studies, 22 of them with dogs (Bargigli et al., 2026). Most used correlation analyses. Co-modulation was significant in 16 studies, partial in 16 and absent in 5, and sample sizes ranged from ten pairs or fewer to more than 130. Methods that analyzed coupling across time series gave more consistent results than correlations between single time points. The authors judged the evidence as supportive but not conclusive, and the heterogeneity of methods as limiting generalization.

Two distinctions follow. Correlated hair cortisol says that dogs with high values tend to have owners with high values over months; it says nothing about moment-to-moment responses. Correlated heart rate time courses come closer to synchrony, but they remain correlations of arousal, not of emotion. And "emotional mirroring" implies that the dog takes over the owner's emotional state, which none of these studies measured. The covariation they report is compatible with shared environment, shared activity and mutual influence in either direction.

None of these studies measured oxytocin, so links between synchrony and this hormone remain speculation (what oxytocin does and does not do in dogs).

8. Contagious Yawning: Present, but Contested in Meaning

Contagious yawning is the most frequently cited example of dogs "mirroring" humans. In the first study, 21 of 29 dogs yawned when watching a person yawn, while control mouth movements produced no yawning in any dog (Joly-Mascheroni et al., 2008). A later study of 25 dogs found more yawns when watching the owner than when watching an unfamiliar person, and no difference in heart rate between conditions, which the authors read as support for a link to empathy rather than to stress (Romero et al., 2013).

Other results point elsewhere. In 60 shelter dogs watching an unfamiliar person, contagious yawning did not occur at the group level; 12 dogs yawned only in response to human yawns, these dogs were not better at following human cues, and their cortisol stayed elevated after the yawning stimuli while other dogs' cortisol fell (Buttner & Strasser, 2014). The authors interpreted this as arousal in a stressful context, and yawning is also discussed as a possible stress-related behavior (what research says about yawning as a stress signal).

The most thorough analysis combined data from six studies in a Bayesian reanalysis (Neilands et al., 2020). It found robust support that contagious yawning exists in dogs, but no evidence for a familiarity bias or a sex bias, two predictions of the empathy hypothesis. In a new experiment, dogs did not yawn more in response to a prosocial than to an antisocial demonstrator. The authors concluded that contagious yawning in dogs is present but not mediated by empathic mechanisms. Yawning is therefore a form of behavioral matching with an unclear function, not evidence that dogs share their owners' feelings.

9. Synchrony, Imitation and Unintended Cueing

9.1 Automatic Imitation

Synchronization concerns timing; imitation concerns form. The two meet in a study on automatic imitation (Range et al., 2011). Dogs were trained to open a sliding door with their head or their paw, and were then rewarded for using the same body part as their owner had just used, or the opposite one. Dogs that had to do the opposite needed more trials to learn the task and later made more errors in the direction of copying. The authors described this as automatic imitation: dogs had difficulty suppressing a tendency to match the body part they had seen used. They interpreted it as shaped mainly by dogs' developmental experience with humans rather than by domestication.

This finding is relevant because it shows a tendency to match human movements that operates even when it costs the dog a reward. It is not the same as synchronizing pace on a walk, and research on social learning (how dogs learn by observing others) and overimitation (why dogs sometimes copy unnecessary actions) describes copying behaviors with their own mechanisms.

9.2 The Clever Hans Contrast

The Clever Hans effect is the mirror image of synchronization research. In synchronization studies, the person's behavior is the variable of interest; in cognition studies, it is a confound that may guide the dog without anyone intending it (how unintended human cues influence dogs' choices).

The same sensitivity that makes synchrony possible makes cueing possible. In detection work, handlers' false beliefs about where a target was hidden changed where alerts were called, even when no target odor was present at all (Lit et al., 2011). In a pointing study, by contrast, owners' beliefs alone did not measurably change their dogs' choices, and owner knowledge without a visible point did not guide the dogs, although owners who actively directed their dogs could influence them (Schmidjell et al., 2012). The authors noted that owner behavior itself was not measured, because dogs may respond to cues observers cannot detect.

Synchronization studies ask people to stay silent and keep their hands still. This reduces deliberate signals but cannot remove the movements that are the test condition. A dog that stops when its owner stops responds to a human cue by design; whether one calls this synchrony or cueing depends on the question, not on the dog.

9.3 Alternative Explanations

Several explanations compete with, or contribute to, what is measured as synchrony. They are not mutually exclusive, and most studies can rule out only some of them.

• Leash mechanics. On leash, pace and direction are partly imposed, so off-leash findings do not describe leash walking (Duranton et al., 2017).

• Following the person. Owners decide when and where walks go; following a decision maker produces the same data as synchronization (Duranton et al., 2017).

• Shared stimuli. A noise, a gate or another dog can make both partners stop or move.

• Proximity seeking under arousal. Authors argued against stress as the main explanation but did not measure physiological arousal (Duranton et al., 2017; Duranton et al., 2022).

• Training history. One summary reported no effect of obedience training level (Lamontagne et al., 2025), but most samples did not vary training systematically.

• Physical similarity. Speed matching is easier between partners that move alike (Lamontagne et al., 2025).

10. What Follows for Practice

10.1 What the Research Supports

Some conclusions are directly supported. Pet dogs moving freely adjust their position, movement and pace to a familiar person in standardized conditions (Duranton et al., 2017; Duranton et al., 2018). This happens without verbal instructions. The person's own movement is therefore a signal for the dog, whether or not the person intends it. The tendency appears early in puppies (Duranton et al., 2022), is weaker in some populations with less everyday contact with humans (Duranton et al., 2019b; Heurlin et al., 2024), and is weaker with children than with adults (Wanser et al., 2021). Physiological covariation exists in some studies and some breed groups, but its meaning is unresolved (Bargigli et al., 2026).

No study reviewed here tested whether deliberately using synchrony in training improves leash walking, cooperation or the relationship. Everything in the following subsections that goes beyond these findings comes from practice and is marked as such.

10.2 Walking Together

In practice: Many dogs adjust to the handler's pace more readily when the handler walks steadily and signals changes of speed and direction with the body before using the leash; slowing down over a few steps is easier to follow than an abrupt halt. This is an observation from training practice, not a finding from the synchronization studies, which tested dogs off leash and briefly.

Pulling on the leash is a common reason for frustration on walks, and it is tempting to answer it with leash corrections. Aversive leash methods can stop pulling in the moment. The question is what the dog learns from them, and at what cost: whether it learns to walk with the person or to avoid the discomfort, and how the experience affects its arousal and its relationship with the person holding the leash (what side effects of punishment research has documented). The synchronization data offer no support for corrections; they describe a tendency to coordinate that appears without them.

10.3 Loose Leash and Freedom of Movement

In practice: A loose leash gives the dog room to coordinate its own movement with the handler instead of being pulled along. Where it is safe and permitted, short off-leash or long-line phases in a familiar area show more clearly how strongly a dog follows a handler's movement than walking on a short leash does. This is an inference from practice and from the logic of the off-leash studies, not a tested training method.

10.4 The Calm Handler

Advice that "a calm handler makes a calm dog" is common. The evidence for it is weaker than it sounds. The physiological studies show correlations in arousal measures between dogs and owners in some breed groups and some situations (Sundman et al., 2019; Koskela et al., 2024), but not in others (Höglin et al., 2021), and a study that manipulated owner stress produced only a weak stressor and an effect tied to ownership duration (Katayama et al., 2019). In practice: Moving slowly, breathing steadily and avoiding hectic movements around an aroused dog often seems to help, but this is an observation from practice, not a demonstrated effect of synchronization on the dog's emotional state.

10.5 Handling and Cooperative Care

Veterinary and grooming procedures put people and dogs close together under some degree of stress. Training approaches that give the dog more predictability and choice in such situations are discussed in their own literature (how cooperative care gives dogs more control during handling). The synchronization studies did not include handling. That slow, announced movements are more readable for a dog than sudden ones is a reasonable inference from them, not a tested finding.

10.6 Reading What the Dog Follows

In practice: When a dog stops, turns or looks somewhere on a walk, it helps to ask what it is following: the handler's movement, an unintended cue, a stimulus both noticed, or something only the dog perceives. Watching posture, attention (how to read canine body language) and the exact moment of change often separates these. This is a practice heuristic, not a research method.

Play between dogs and humans is another situation in which timing matters, and play research has its own findings on signals and coordination (how dogs signal and coordinate play).

11. Summary at a Glance

Behavioral synchronization means doing the same thing, at the same time, in the same place as a partner (Duranton & Gaunet, 2016). Off leash, pet dogs stayed close to their owners, matched standing and walking and switched activity about 3.4 seconds after them (Duranton et al., 2017), also outdoors (Duranton et al., 2018) and in groups (Lamontagne et al., 2023), though less closely than with a familiar dog on some measures (Lamontagne et al., 2025).

Synchrony was weaker in shelter dogs with caregivers (Duranton et al., 2019b), with children (Wanser et al., 2021) and absent in a few pack-living wolves and dogs (Heurlin et al., 2024), but present in young puppies regardless of familiarity (Duranton et al., 2022). One breed group preferred a synchronizing stranger (Duranton et al., 2019a). Physiological covariation is a separate, mostly correlational category with mixed results (Sundman et al., 2019; Höglin et al., 2021; Bargigli et al., 2026), and contagious yawning exists without clear evidence of empathy (Neilands et al., 2020).

12. Research Gaps and Critical Appraisal

12.1 One Research Program

Most behavioral evidence comes from one French research group and its collaborators, using similar paradigms (Duranton & Gaunet, 2016, 2018). Consistency within a program is valuable, but independent replications with different procedures are few. The Swedish enclosure study and the American child study are important exceptions (Heurlin et al., 2024; Wanser et al., 2021).

12.2 Sample Size and Breed

Sample sizes range from seven analyzed dog pairs (Koyasu et al., 2022) and six wolves (Heurlin et al., 2024) to 60 dogs in the largest behavioral study (Lamontagne et al., 2023). Several studies used only two breed groups, molossoids and shepherds (Duranton et al., 2017; Lamontagne et al., 2023), or cooperative breeds only (Sundman et al., 2019; Koskela et al., 2024). The breed effect on preference rests on one study (Duranton et al., 2019a). For several key papers, sample sizes could not be confirmed from accessible sources and are not stated in this article.

12.3 Short Tests in Controlled Settings

Test phases lasted 20 to 30 seconds in most studies and a few minutes in others, in rooms or enclosed areas. Everyday walks last longer, take place on leash more often than not, and include other people, dogs and smells. How much of the measured synchrony survives in such conditions is unknown.

12.4 Mechanism

Studies describe synchrony but rarely test its mechanism. Following a decision maker, proximity seeking, local enhancement, learned responses to the owner's movement and a general tendency to coordinate with social partners all predict similar data. Physiological arousal was not measured in the main behavioral experiments, although the authors named it as a possible influence (Duranton et al., 2017).

12.5 Physiology and Causality

The physiological studies are correlational, with one partial exception that manipulated owner stress but produced a weak stressor (Katayama et al., 2019). Shared environment is the main alternative, and the Koskela et al. study included a direct control by random re-pairing (Koskela et al., 2024). Hair cortisol and moment-to-moment heart rate measure different time scales, and the review found methods heterogeneous (Bargigli et al., 2026).

12.6 Bond and Outcome

Familiarity effects are inferred from comparisons between populations, such as pet and shelter dogs, rather than from manipulating the bond (Duranton et al., 2019b). Where bond was measured directly, concordance did not track it (Leighton et al., 2025). The reviewed evidence does not establish whether synchrony predicts welfare, training success or the quality of the relationship over time.

13. Conclusion

Dogs coordinate their movement with familiar people. Off leash, in rooms, in enclosures and in open areas, pet dogs stay close, walk when their person walks, stop when their person stops and look where their person looks. The tendency is already present in young puppies, is weaker in some dogs that share less of their daily life with humans, and is weaker with children than with adults. This is a solid and useful finding.

It is also a narrow one. Synchronization describes timing and position, not shared feelings. Dogs and owners can covary in stress hormones and heart rate, but not in all breed groups and not in a way that demonstrates emotional mirroring, and contagious yawning does not appear to reflect empathy. The same sensitivity that makes synchrony possible also makes dogs responsive to unintended cues. For everyday life, the most defensible lesson is simple: a dog reads how its person moves, so the person's movement is part of the communication, whether or not anyone means it to be.

Key Insights (Takeaways)

• Behavioral synchronization means doing the same thing, at the same time, in the same place; it describes coordination, not emotion (Duranton & Gaunet, 2016).

• Off leash, 48 pet dogs stayed close to their owners about 79 percent of the time and switched activity on average 3.4 seconds after them (Duranton et al., 2017).

• In one study, molossoid dogs, but not shepherd dogs, preferred a stranger who had walked in synchrony with them (Duranton et al., 2019a).

• Puppies synchronized with familiar and unfamiliar people alike, while six pack-living wolves and six pack-living dogs did not synchronize with a handler (Duranton et al., 2022; Heurlin et al., 2024).

• Family dogs synchronized with children above chance, but less than reported for adults (Wanser et al., 2021).

• Cortisol and heart rate covariation is a separate, weaker category: correlational, not found in all breed groups, and not evidence of emotional mirroring (Höglin et al., 2021; Bargigli et al., 2026).

• Contagious yawning exists in dogs, but a reanalysis of six studies found no familiarity bias supporting an empathy link (Neilands et al., 2020).

• The sensitivity behind synchrony also makes unintended cueing possible, so a dog's response to human movement is part of the communication whether or not it is intended (Lit et al., 2011).

References

Bargigli, G., Frassineti, L., Baragli, P., Scopa, C., Vignoli, A., & Lanata, A. (2026). Evidence of physiological comodulation during human–animal interaction: A systematic review. Annals of the New York Academy of Sciences, 1560(1), e70299. https://doi.org/10.1111/nyas.70299

Buttner, A. P., & Strasser, R. (2014). Contagious yawning, social cognition, and arousal: An investigation of the processes underlying shelter dogs' responses to human yawns. Animal Cognition, 17(1), 95–104. https://doi.org/10.1007/s10071-013-0641-z

Duranton, C. (2020). Local synchrony as a tool to estimate affiliation in dogs. Journal of Veterinary Behavior, 36, 48–53. https://doi.org/10.1016/j.jveb.2019.10.007

Duranton, C., Bedossa, T., & Gaunet, F. (2016). When facing an unfamiliar person, pet dogs present social referencing based on their owners' direction of movement alone. Animal Behaviour, 113, 147–156. https://doi.org/10.1016/j.anbehav.2016.01.004

Duranton, C., Bedossa, T., & Gaunet, F. (2017). Interspecific behavioural synchronization: Dogs exhibit locomotor synchrony with humans. Scientific Reports, 7, 12384. https://doi.org/10.1038/s41598-017-12577-z

Duranton, C., Bedossa, T., & Gaunet, F. (2018). Pet dogs synchronize their walking pace with that of their owners in open outdoor areas. Animal Cognition, 21(2), 219–226. https://doi.org/10.1007/s10071-017-1155-x

Duranton, C., Bedossa, T., & Gaunet, F. (2019a). Pet dogs exhibit social preference for people who synchronize with them: What does it tell us about the evolution of behavioral synchronization? Animal Cognition, 22(2), 243–250. https://doi.org/10.1007/s10071-019-01241-w

Duranton, C., Bedossa, T., & Gaunet, F. (2019b). When walking in an outside area, shelter dogs (Canis familiaris) synchronize activity with their caregivers but do not remain as close to them as do pet dogs. Journal of Comparative Psychology, 133(3), 397–405. https://doi.org/10.1037/com0000171

Duranton, C., Courby-Betremieux, C., & Gaunet, F. (2022). One- and two-month-old dog puppies exhibit behavioural synchronization with humans independently of familiarity. Animals, 12(23), 3356. https://doi.org/10.3390/ani12233356

Duranton, C., & Gaunet, F. (2016). Behavioural synchronization from an ethological perspective: Overview of its adaptive value. Adaptive Behavior, 24(3), 181–191. https://doi.org/10.1177/1059712316644966

Duranton, C., & Gaunet, F. (2018). Behavioral synchronization and affiliation: Dogs exhibit human-like skills. Learning & Behavior, 46(4), 364–373. https://doi.org/10.3758/s13420-018-0323-4

Heurlin, J., Barabás, G., & Roth, L. S. V. (2024). Behavioural synchronisation between different groups of dogs and wolves and their owners/handlers: Exploring the effect of breed and human interaction. PLOS ONE, 19(5), e0302833. https://doi.org/10.1371/journal.pone.0302833

Höglin, A., Van Poucke, E., Katajamaa, R., Jensen, P., Theodorsson, E., & Roth, L. S. V. (2021). Long-term stress in dogs is related to the human–dog relationship and personality traits. Scientific Reports, 11, 8612. https://doi.org/10.1038/s41598-021-88201-y

Joly-Mascheroni, R. M., Senju, A., & Shepherd, A. J. (2008). Dogs catch human yawns. Biology Letters, 4(5), 446–448. https://doi.org/10.1098/rsbl.2008.0333

Katayama, M., Kubo, T., Yamakawa, T., Fujiwara, K., Nomoto, K., Ikeda, K., Mogi, K., Nagasawa, M., & Kikusui, T. (2019). Emotional contagion from humans to dogs is facilitated by duration of ownership. Frontiers in Psychology, 10, 1678. https://doi.org/10.3389/fpsyg.2019.01678

Koskela, A., Törnqvist, H., Somppi, S., Tiira, K., Kykyri, V.-L., Hänninen, L., Kujala, J., Nagasawa, M., Kikusui, T., & Kujala, M. V. (2024). Behavioral and emotional co-modulation during dog–owner interaction measured by heart rate variability and activity. Scientific Reports, 14, 25201. https://doi.org/10.1038/s41598-024-76831-x

Koyasu, H., Goto, R., Takagi, S., Nagasawa, M., Nakano, T., & Kikusui, T. (2022). Mutual synchronization of eyeblinks between dogs/cats and humans. Current Zoology, 68(2), 229–232. https://doi.org/10.1093/cz/zoab045

Lamontagne, A., Legou, T., Brunel, M., Bedossa, T., & Gaunet, F. (2025). Familiar dog or familiar person: Who do pet dogs best synchronize with? Animals, 15(4), 505. https://doi.org/10.3390/ani15040505

Lamontagne, A., Legou, T., Rauchbauer, B., Grosbras, M.-H., Fabre, F., & Gaunet, F. (2023). Behavioural synchronization and social referencing of dogs and humans: Walking in dyad vs in group. Animal Cognition, 26(3), 1021–1034. https://doi.org/10.1007/s10071-023-01750-9

Leighton, S. C., Rodriguez, K. E., Jensen, C. L., MacLean, E. L., Baus, J. J., Ashbeck, E. L., Brown, E. A., Mehl, M. R., Taylor, D. J., van Marle, K., Schwichtenberg, A. J., Bedrick, E. J., & O'Haire, M. E. (2025). Sleep–wake dynamics in veteran–service dog dyads: An exploration of overnight concordance. Sleep, 48(11), zsaf177. https://doi.org/10.1093/sleep/zsaf177

Lit, L., Schweitzer, J. B., & Oberbauer, A. M. (2011). Handler beliefs affect scent detection dog outcomes. Animal Cognition, 14(3), 387–394. https://doi.org/10.1007/s10071-010-0373-2

Neilands, P., Claessens, S., Ren, I., Hassall, R., Bastos, A. P. M., & Taylor, A. H. (2020). Contagious yawning is not a signal of empathy: No evidence of familiarity, gender or prosociality biases in dogs. Proceedings of the Royal Society B, 287(1920), 20192236. https://doi.org/10.1098/rspb.2019.2236

Range, F., Huber, L., & Heyes, C. (2011). Automatic imitation in dogs. Proceedings of the Royal Society B, 278(1703), 211–217. https://doi.org/10.1098/rspb.2010.1142

Romero, T., Konno, A., & Hasegawa, T. (2013). Familiarity bias and physiological responses in contagious yawning by dogs support link to empathy. PLOS ONE, 8(8), e71365. https://doi.org/10.1371/journal.pone.0071365

Schmidjell, T., Range, F., Huber, L., & Virányi, Z. (2012). Do owners have a Clever Hans effect on dogs? Results of a pointing study. Frontiers in Psychology, 3, 558. https://doi.org/10.3389/fpsyg.2012.00558

Sundman, A.-S., Van Poucke, E., Svensson Holm, A.-C., Faresjö, Å., Theodorsson, E., Jensen, P., & Roth, L. S. V. (2019). Long-term stress levels are synchronized in dogs and their owners. Scientific Reports, 9, 7391. https://doi.org/10.1038/s41598-019-43851-x

Wanser, S. H., MacDonald, M., & Udell, M. A. R. (2021). Dog–human behavioral synchronization: Family dogs synchronize their behavior with child family members. Animal Cognition, 24(4), 747–752. https://doi.org/10.1007/s10071-020-01454-4