Abstract

I propose that reinforcing full-body shake-off behaviour may support emotional regulation and behavioural recovery in dogs, drawing on ethological theory, physiological and neurobiological perspectives, and repeated observations during applied behaviour practice. The hypothesis remains untested in controlled studies. In this article the shake-off is defined as a fast side-to-side full-body shaking movement that begins at the head and travels towards the tail in a wave-like pattern. Observations were drawn from pet dogs seen in training and behavior consultation sessions. The movement was most often observed after moments of social or environmental intensity, such as fixating on, barking and lunging at another dog; chasing wildlife; jumping and mouthing, or grabbing at a leash or handler’s clothing; stopping play or other intense activities. Shake-offs were first reinforced during reactive dog training, and later in other contexts as well. This article outlines a two-level strategy for reinforcing spontaneous shake-offs in neutral and early-arousal contexts, along with a complementary conditioned-relaxation approach used in selected cases. I begin by defining several key expressions used in dog training and applied behaviour consulting, as these terms are used inconsistently across disciplines. I then summarize how shake-offs have been described in the literature and outline the functions proposed in dogs and other species. After that, I will describe the strategies in which I reinforced shake-offs and present short case studies with video examples. Finally, I will discuss the limitations of these observations and suggest directions for future research.

Introduction

Background

Earlier literature described shaking behaviour broadly, often framing it as a stress-related, displacement, or cut-off action without a consistent operational definition. Trembling and shivering were generally treated as separate categories, but different types of shaking were not examined independently of their assumed emotional function. In these sources, stress-related labels referred to shaking observed during or after situations involving uncertainty, frustration, or mild social tension. “Displacement” and “cut-off” behaviors were used to describe brief self-directed movements, such as shaking, scratching, and yawning that appeared to interrupt or redirect an ongoing behaviour sequence. In ethological terms, self-directed movements are brief actions the animal performs on their own body or in close proximity to it, including grooming and scratching, and may occur during ambiguous or mildly arousing situations. More recent observational work has begun to distinguish the form and context of shaking, specifically naming shake-off with greater precision, creating space for clearer definitions. The following section outlines this historical development in more detail.

Historically, dog training and veterinary behaviour literature has framed shake-offs as stress- or conflict-related behaviours. They have been described as stress signals (Rugaas, 2006, p. 56), cut-off signals (Stewart, 2016, p. 68), or displacement behaviours linked to frustration, conflict, or uncertainty (Handelman, 2008, p. 238; Horwitz, 2018, p. 396; Overall, 1997, p. 210). Early physiological studies documented increases in cortisol and heart rate during acute and chronic stressors, and reported body shaking among several behaviours observed during these challenges (Beerda et al., 1997, 1998, 1999). However, these studies did not analyse shake-offs as a distinct behaviour, which limits how their findings can be interpreted in relation to this specific movement. Later observations have broadened the picture. Pastore et al. (2011) reported that shake-offs occurred most often after agility runs rather than during the competitive event itself, indicating that the behaviour may arise at points of transition rather than solely in response to acute distress.

Preliminary practice-based data offer additional context. A small exploratory field project by DogFieldStudy (2024) paired real-time heart rate tracking with video during off-leash walks. In several clips, shake-offs were observed in close proximity to peak heart rate moments. Although not peer-reviewed and based on a limited sample, this biometric pairing offers early indications that shake-offs may occur near shifts in physiological arousal. In close-contact interactions with both humans and conspecifics, both dogs and human-socialized wolves displayed self-directed behaviours including head and body shaking, indicating that shake-like movements can emerge during mild social stress or arousal adjustments (Wirobski et al., 2021). A recent study on restrictive veterinary handling found that dogs frequently performed body shaking immediately after release from mild restraint, interpreting the movement as part of a short transition period following mild stress (Cisneros et al., 2025).

The most detailed and methodologically focused work to date is the 2024 study by Bryce and colleagues, the first peer-reviewed investigation to examine the canine shake-off directly. Observing 120 shake-offs across 96 dogs in public and daycare settings, the authors described the movement as “a rapid side-to-side movement of the body or head” (2024, p. 1). Shake-offs in dry dogs occurred most frequently after greetings, interruptions, and other activity transitions. Affective indicators such as posture and proximity remained stable, and in most cases a different behaviour occurred immediately before and after the shake-off. Bryce et al. found no empirical support for interpreting shake-offs as stress signals; instead, they proposed that the behaviour functions as a transitional or regulatory action. As an observational study without physiological measures and limited to two environments, the findings warrant cautious generalisation but currently provide the clearest empirical description of the behaviour’s form and context.

These dog-specific findings are complemented by physiological and biomechanical research that sheds light on how the shake-off movement is produced. Dickerson et al. (2012) showed that wet and dry shake-offs share the same underlying oscillation pattern. In mammals, rhythmic and reflexive movements of this kind are typically coordinated by brainstem and spinal motor systems (Longstaff, 2005, pp. 217–238). These circuits are influenced by neuromodulators such as serotonin, produced in the brainstem, which is closely linked to arousal and alters the excitability of motor networks (Longstaff, 2005, pp. 96–99). This makes it plausible that shake-offs may shift with short-term changes in arousal, although this has not been tested directly in dogs.

Beyond domestic dogs, comparable patterns appear across several species. In primates, body shaking is included among self-directed behaviours that increase during socially tense or ambiguous situations and are sensitive to pharmacological manipulation of anxiety (Reamer et al., 2010). In greylag geese, brief whole-body shaking occurs after agonistic interactions and is associated with elevated heart rate, indicating a post-conflict arousal shift (Wascher et al., 2008). Ungulates show similar patterns: This work identified displacement grooming behaviours, including headshakes, in antelopes and related species during ambiguous or tension-related contexts (Luo et al., 2022). In horses, shake-like release behaviours have been documented among other stress-related or transitional responses, particularly in therapeutic handling contexts (Kaiser et al., 2006). A 2025 study of Asian elephants by Vilela documented shake-like self-directed behaviours, including rapid head shakes, increasing after socially tense interactions. While species-specific in form, these findings collectively suggest that brief shaking or analogous self-directed movements often appear during transitional or post-conflict shifts in arousal.

Taken together, the canine-focused and cross-species literature reviewed above is consistent with interpreting shake-offs as naturally occurring transitional behaviours rather than reliable indicators of stress. In my observational data, shake-offs frequently appeared between behavioural states, with the preceding state typically higher arousal than the state that followed, raising the possibility of a behavioural regulatory role. Although shake-offs are both common and easily observed, no research has examined whether reinforcement influences their timing or availability. The hypothesis explored in this article is that reinforcement may increase the accessibility of shake-offs during moments of behavioural transition.

Strategies

For the purposes of this article, reactive dogs and reactive (behaviour) were defined earlier.

The core strategy described here — reinforcing spontaneous full-body shake-offs — was first applied with reactive dogs in Behavior Adjustment Training (BAT) (Stewart, 2016), where environmental management was used to keep the dog under threshold so that information processing (noticing), choice-making (engaging and disengaging) , and behavioural recovery (going or staying under reactivity threshold) remained possible. Within this framework, the shake-off is described as a natural “reset” behaviour suitable for reinforcement (Stewart, 2016, p. 68).

Spontaneous shake-offs were later reinforced in dogs who showed other forms of heightened arousal, including adolescent dogs who pulled, chased, mouthed, nipped, or jumped during emotionally intense moments. Across cases, shake-offs appeared during state transitions, particularly when dogs shifted from alertness or intensity toward under-threshold behaviours. The aim of reinforcing these transitions was to increase the availability of shake-offs and to encourage the behaviour to appear earlier in emotionally challenging sequences, when it might support behavioural recovery. Notably, several dogs did not offer shake-offs during reactive dog training when they remained consistently on threshold, but began to offer them once the behaviour was reinforced in low-arousal (under-threshold) contexts outside of training. In those instances, shake-offs were followed by shifts toward lower-arousal behaviour, indicating that reinforcement may increase the availability of the behaviour during arousal transitions. Reinforcement appeared to increase the availability of shake-offs and to shift them earlier in emotionally challenging sequences, potentially supporting behavioural recovery. From a learning-theory perspective, the shake-off could function as a differentially reinforced alternative behaviour (DRA) and, in many contexts, an incompatible behaviour (DRI). Shake-offs were captured and reinforced under threshold, most often with food and, in fear-reactive cases, functionally by increasing distance from the trigger. In practice, shake-offs frequently appeared as transitional behaviours between brief on-threshold engagement and subsequent disengagement. Reinforcement appeared to make shake-offs more available during these transitions and to support earlier behavioural down-regulation. Because the shake-off is a naturally occurring movement, it could be captured in real time without prior training.

Core Strategy: Reinforcing spontaneous shake-offs

This approach distinguishes between shake-offs occurring in neutral or low-arousal contexts (under threshold) and those occurring in contexts where emotional regulation is challenged (on  threshold, but not over threshold).

Neutral-context shake-offs, including shaking after waking, shaking off water, or shifting from play back to walking, were acknowledged with calm verbal praise or gentle petting. In pet dogs these typically function as lower-value reinforcers, as they are frequently used in daily interaction and only occasionally, but not consistently, paired with primary reinforcement. These lower-value reinforcers were used to avoid increasing arousal or altering the context of the behaviour, and delivered on a continuous schedule.

On-threshold shake-offs, such as those observed after noticing a trigger or in contexts where the dog had previously escalated to barking, lunging, or grabbing, were marked with a conditioned verbal marker (“Yes”) and followed immediately with high-value food. Food value was assessed individually by comparing options and confirming that the dog would eat under threshold. Continuous capture-based reinforcement maintained clarity for caregivers and, in reactive cases, food was paired with functional reinforcement (increasing distance) when relevant. We used BAT-style setups in which reactive dogs were worked in controlled environments with calm decoy dogs. Distance and positioning were arranged so the dog could notice the trigger while remaining under threshold. In these setups, handlers reinforced a range of naturally occurring under-threshold behaviours that were incompatible with escalation, such as reorienting, pausing, moving away, sniffing, and shake-offs (examples only, not an exhaustive list). Functional reinforcement was provided by increasing distance when relevant. Escalation was avoided through environmental management, including adjusting distance, briefly blocking visual access when needed, and delivering food reinforcement to support under-threshold behaviour. Across contexts, the two-level structure remained simple for caregivers to apply consistently, using social reinforcement for neutral-context shake-offs at home and on calm walks, and marker-based high-value food (and, when relevant, increased distance) for on-threshold shake-offs during training sessions.

Complementary strategy: Teaching a safety-sign relaxation cue

A conditioned relaxation cue was incorporated with two dogs as a complementary strategy. Relaxed behaviour on a mat was first shaped following standard relaxation protocols (Overall, 1997), and once the dogs consistently showed softened muscles, slower breathing, and settled postures, a verbal cue “relax” was added. This cue was then practised in calm indoor contexts until the behaviour became reliable.

Drawing on the phenomenon that cues associated with safety can later function as conditioned inhibitors of fear (Rescorla, 1969), the verbal cue was introduced outdoors during situations where the dogs were on threshold. For the purposes of this article, on threshold refers to observable patterns such as forward weight shift, fixed gaze, stiffened musculature, closed mouth, disinterest in high-value food rewards, and sustained orientation toward a stimulus with limited spontaneous disengagement, while the dog still remained responsive to the handler (e.g., follows previously learned cues). This state typically precedes clusters of reactive behaviours as defined in this article (e.g., barking, lunging, accelerating toward stimulus, etc.)

The cue did not evoke relaxation directly. Instead, it frequently preceded a full-body shake-off, followed by the dog moving back under threshold, indicated by resuming food intake, responding to familiar cues, or disengaging from the stimulus. By contrast, the cue did not elicit shake-offs when the dogs were fully relaxed. The cue was trained within a relaxation protocol and then extended by pairing a verbal cue (“relax”) with the relaxed, low-arousal state the protocol aimed to build. The cue was not intended to produce a specific posture; rather, it was intended to signal and help elicit the underlying relaxed state, which can be expressed through multiple behaviours. As you will see in the case examples, the cue did elicit other behaviours consistent with the relaxation protocol and under-threshold functioning (for example, softer body language, resuming food intake, and increased responsiveness to familiar cues). The notable point reported here is that, in these two dogs, the cue also frequently preceded a full-body shake-off and subsequent behavioural recovery. The shake-off itself was not separately trained as a cued response; it was an observed transitional behaviour that appeared in association with the cued relaxed state. Although based on only two dogs, and clearly preliminary, this pattern suggests that a conditioned “relax” cue may increase the likelihood of a shake-off during early arousal transitions.(on threshold) The interaction between a conditioned “relax” cue and the likelihood of shake-offs emerged unexpectedly and may warrant future study.

The next section presents a detailed case study with integrated video references to support the behavioural descriptions, followed by a shorter example demonstrating how these strategies were applied in a different context.

Mini case study

Black Mouth Cur Mix (“Hector”)

“Hector,” a neutered male, 80-pound, black mouth cur mix adopted from a rescue organization in Texas at approximately 18months of age, worked with the author between 2020 and 2023 in a program that included behaviour modification, predictable structured walks, indoor relaxation work, and enrichment. A second dog walker trained by the author supported the case from 2022 onward. Medication initially included fluoxetine 30 mg daily, which was transitioned to venlafaxine 37.5 mg daily in late 2023 following increased arousal, at the same time Hector was diagnosed with knee-ligament pathology. Complications following surgical intervention led to amputation and, ultimately, loss of the dog in early 2025.

Hector demonstrated reactive behaviour toward dogs, people, wildlife, moving objects, and mechanical sounds such as garage doors. On stepping outside, his baseline arousal was consistently high: scanning, air-sniffing, stiff tail carriage, shallow panting, facial tension, intermittent freezing, and reduced food engagement. Indoors and in the yard, his body language softened and he engaged with food more reliably. Prior to intervention, shake-offs were rare during walks, and appeared mainly indoors after arriving home and having all gear removed.

Behaviour work already in place included low-arousal street management (predictable, quiet places, keeping him under threshold), classical relaxation work indoors, and structured wildlife-watching routines. Shake-off reinforcement was layered into this framework beginning in 2021. Spontaneous shake-offs were captured and reinforced with scatter feeding, brief food-search opportunities, or increased distance when he noticed triggers without escalating. For example, spontaneous shake-offs were reinforced in both neutral contexts and on-threshold contexts. At first, he mostly offered shake-offs at home after returning from walks, and I consistently acknowledged those with calm praise and often followed with food. In parallel, I trained a relaxation protocol and paired the relaxed state with a verbal cue (“relax”). Later, he began to offer shake-offs outdoors in situations where he would previously stiffen, fixate, and often escalate into a reactive episode. When he showed on-threshold body language after noticing a trigger (most often dogs or wildlife), I used the trained “relax” cue and reinforced any subsequent shake-off immediately while he remained under threshold. I observed an increase in shake-offs beyond the original home context, including their occurrence in trigger-related situations where I had not previously seen them. This conditioned relaxation cue frequently elicited a shake-off followed by visible body language softening (transitioning to under threshold) including resuming food intake, responding to familiar cues, and disengaging from environmental stimuli or triggers.

Two video examples illustrate this work. In a 2022 wildlife context (video 2), a deer stood up suddenly on the other side of a fence. Hector oriented toward the deer and stiffened. The handler kept the leash loose and cued a “keep going”; after some steps Hector performed a spontaneous shake-off and was able to walk away from the deer. In another 2022 clip (video 4, 9), he noticed a dog across a quiet road at a workable distance. The familiar tactile prompt and quiet “relax” cue produced a shake-off, after which he was able to perform a U-turn and search for food in the grass.

Between 2021 and 2023, Hector’s recovery after exposures became faster, freeze-and-stare sequences shortened, and spontaneous shake-offs appeared more frequently as transitional resets. Although these observations remain descriptive rather than experimentally controlled, the case suggests that reinforcing shake-offs may help support in-the-moment emotional regulation when used alongside comprehensive behaviour modification plans.

Supporting example – schipperke (author’s dog)

The author’s schipperke, adopted from a breeder at 8 weeks of age, showed early signs of fearfulness and sensory sensitivity. The puppy was hesitant to approach unfamiliar people, startled easily at sudden movements or noises, barked at outdoor sounds, and was sensitive to touch and handling. When approached or touched during sleep, she occasionally growled or snapped. At home, she engaged in repetitive self- and toy-licking and -sucking. In the car, she was restless and vocal. Outdoors, she showed strong interest in fast-moving animals and objects, often shifting to on-threshold arousal without overt reactive behaviour. She frequently attempted to chase cats and birds with stiff pursuit behaviour. These last patterns were consistent with breed-typical predatory interest rather than puppy play. She did not show shake-off behaviours outdoors, but she did indoors in several contexts, such as after play, after waking from a nap, or after returning from a walk. In addition to foundational skills, her training incorporated the same relaxation work as with Hector, observational skills using the Premack principle (reinforcing a less likely behaviour with access to a more likely behaviour, for example sitting to watch birds), and later predation-substitute training routines (Muller, 2020). Functional outlets such as scent-search games, scatter feeding, and structured tug, fetch, and flirt-pole sessions functioned as substitutional activities for chasing and supported her ability to disengage from moving objects, which later transferred to wildlife.She was trained using the same core and complementary strategies described above. By adulthood, she offered spontaneous shake-offs during everyday transitions and after mild arousal shifts.

Video examples illustrate this generalization: In 2024 she performed a spontaneous shake-off after disengaging from a crow (video 1). In 2025 she offered a spontaneous shake-off after following the author’s husband and being asked to stop and return to heel (video 7). Later in 2025 she disengaged from a dead animal she had been sniffing, performed a spontaneous shake-off, and continued the walk (video 8). These situations reflect the interrupted behaviour sequences described by Bryce et al. (2024), where shake-offs commonly occurred after an ongoing behaviour was paused or redirected, requiring the dog to shift rapidly between activities. If shake-offs contribute to down-regulation, they may help dogs shift from one activity to another. Many dogs transition naturally from sniffing to walking, greeting to walking, or from play to calmer movement, etc. However, some dogs do not show these transitions as readily, and the absence of a transitional behaviour can make activity changes difficult. Thus, when caregivers interrupt an activity (for example, pulling a dog away from play or wildlife via leash or harness), frustration can emerge and may lead to mouthing, jumping, or grabbing/biting behaviour. For these dogs, access to a transitional behaviour such as a shake-off may make it easier to down-regulate and move toward a calmer activity, reducing the need for the caregiver to physically force or remove the dog. These examples suggest that the strategy may generalize across developmental stages and motivational systems. Beyond these two dogs, similar patterns appeared across a range of client cases, which are summarized in the following section on field observations and caregiver engagement.

Description: Video 1. The schipperke is cued to perform “gyere” (“come”) while closely observing a crow and performs a voluntary shake-off before responding to the cue.

Description: Video 2. A black mouth cur mix performs a voluntary shake-off after a deer suddenly appears at close distance. The dog is cued to “keep going,” performs the shake-off, and leaves the deer. The behavior is then reinforced with food. The dog has a history of reactive behavior.

Description: Video 3. A schipperke growls and barks after exiting the yard. The handler cues a positive interrupter (“cheese”), a marker predicting a specific hand-delivered reward. The dog performs a voluntary shake-off before executing the cue.

Description: Video 4. A black mouth cur mix performs a voluntary shake-off following a hand-over-shoulder prompt from the handler, a cue previously trained during relaxation protocols.

Description: Video 5. During reactivity training, a Labradoodle observes an unusual silhouette of a passerby and performs a voluntary shake-off. The trainer marks the behavior verbally, and the guardian delivers a food reward.

Description: Video 6. A Mastiff mix watches sheep at a distance, performs a voluntary shake-off, and disengages from the stimulus.

Description: Video 7. A schipperke investigates a small dead animal. The handler cues “keep going” (in Hungarian), and the dog performs a voluntary shake-off before executing the cue.

Description: Video 8. A schipperke searches for a person. The handler stops the task and cues a “heel” position. The dog performs a voluntary shake-off before executing the cue.

Description: Video 9. People and dogs pass at a distance on the road while the handler increases distance from the triggers using food drops. The dog continues to display behaviors consistent with being at reactivity threshold. When the handler cues “relax,” the dog performs a shake-off behavior.

Field observations and caregiver engagement

Across behaviour consulting and training cases over a five-year period, spontaneous full-body shake-offs were frequently observed at natural transition points in reactive dogs (as defined earlier) and in dogs showing intense adolescent behaviours such as jumping, lunging, pulling on leash, mouthing, nipping, and barking. These observations were not collected through formal coding or blinded scoring but represent repeated, practice-based patterns across unrelated cases and varied environments.

Shake-offs often appeared after brief impulse control (a dog’s ability to pause or inhibit an immediate response to a stimulus long enough to perform an alternative behaviour) moments such as turning away from a trigger, leaving an interesting scent, or disengaging from a valued stimulus. With consistent reinforcement, many dogs began to offer shake-offs earlier, while arousal was rising, such that they either remained under-threshold or did not progress to over-threshold reactivity. Caregivers commonly reported shorter recovery periods from on-threshold state and faster  returns to lower-arousal activities as reinforcement histories developed.

In the two dogs that received the complementary strategy, a consistent pattern emerged. The verbal “relax” cue did not elicit shake-offs during fully relaxed states, but reliably preceded them when the dogs were at mild arousal or on threshold. This effect was first seen in the black mouth cur mix (“Hector”) and later appeared in the author’s schipperke trained with the same routine. In Hector’s case, the sequence—cue, shake-off, softening—was also reproduced by a second handler trained to use the same verbal cue. Although anecdotal, the reproducibility across two dogs and two handlers suggests that shake-offs may be particularly available during early arousal transitions when paired with a conditioned safety cue.

As caregivers became more consistent in noticing and reinforcing shake-offs, many reported a broader increase in awareness of their dog’s communication signals. Some described the dog as shaking off more often; others commented that their dogs shook off more often. This increase may partly reflect reinforcement effects and partly reflect increased observer sensitivity, as caregivers became better at noticing the behaviour. In both scenarios, caregivers described greater confidence and a shift toward reinforcing adaptive transitions rather than suppressing unwanted behaviour. Many caregivers of reactive dogs described a pattern of tightening or pulling the leash immediately when a trigger appeared, even at long distances. When they began looking for shake-offs, they reported being more aware of their own handling and waited to see whether a shake-off would occur at distances where the dog noticed the trigger but was not yet reactive. They described this as reducing their impulse to pull the dog away. These patterns are consistent with findings from veterinary behavioural medicine, where Talamonti et al. (2015) reported that owner adherence improves when treatment plans are clear, manageable, and matched to the caregiver’s available time and skills. Human habit-formation work points in the same direction: Low-effort behaviours repeated in stable contexts are more consistently maintained than those requiring ongoing cognitive control (Wood, 2019). This supports the impression that capturing spontaneous shake-offs is a practical fit for everyday caregiver routines.

A recurring practical observation was that dogs often struggled to complete a full-body shake-off when the leash was tight. Some initiated the movement with a brief head shake but were mechanically unable to complete the sequence, and tense body language frequently persisted until slack was restored. Coaching caregivers to maintain consistent leash slack therefore became an important component of enabling shake-offs to occur naturally and be reinforced during walks. The aim is for caregivers to handle the leash in a way that allows the dog to notice, explore, disengage, and recover without being physically pulled off balance or pushed over threshold. For this, I rely heavily on Grisha Stewart’s long-leash skills and “silky leash” technique, as well as games and exercises from Simone Müller’s Walking Together book that reduce pulling by training both the dog and coaching the caregiver. Because leash mechanics and structured loose leash training are extensive topics in themselves and not specific to shake-offs, I chose not to include the full coaching procedures in this article. While loose leash skills are necessary foundations for reactive dog work, detailing them here would shift focus away from the central topic of shake-offs and their potential contribution to emotional downregulation. I believe that level of detail would be more appropriate for a separate article specifically on leash work or reactive dog training.

Together, these practice-based impressions suggest that reinforcing spontaneous shake-offs may help dogs down regulate from higher arousal levels more effectively while simultaneously increasing caregiver engagement, observational skill, and responsiveness to their dog’s changing behavioural state. These observations create the foundation for the upcoming sections, which consider their practical meaning in the Discussion and then examine the limitations and future research needs of this approach.

Discussion

The observations in this article suggest that shake-offs may play a more nuanced role than traditional labels such as displacement or cut-off behaviour describe. Across the cases reviewed, shake-offs occurred at natural transition points and were often followed by visible behavioural softening. When reinforced consistently, several dogs began offering shake-offs earlier in behavioural sequences that previously led to escalation, raising the possibility that the behaviour becomes more accessible during early arousal shifts.

Recent empirical findings support this broader interpretation. Bryce et al. (2024) reported that shake-offs were most common between behavioural categories and following interruptions, a pattern consistent with their appearance in the present field observations. A restrictive-handling study (Cisneros, 2025) extends this further: Dogs showed body shaking immediately after brief restraint, which the authors interpreted as part of a short transition period following mild stress. Taken together, these studies offer converging evidence that body-shaking behaviours often emerge during shifts out of heightened arousal or following momentary disruption, complementing the practice-based patterns described here. At the same time, these parallels remain descriptive rather than explanatory; the mechanisms underlying shake-offs—whether tied to arousal regulation, biomechanical reset, reinforcement history, or a combination of factors—are not yet understood.

Within the LIFE (Least Inhibitive, Functionally Effective) framework, this approach draws on increasing meaningful choice and supporting functional, welfare-relevant behaviours (Fernandez, 2024). Parallel principles in the Joint Standards of Practice (IAABC, 2025) emphasise non-aversive procedures along with attention to emotional well-being, choice, control, and agency. Reinforcing a naturally occurring transitional behaviour such as a shake-off may support welfare-relevant priorities by providing access to reinforcement without requiring obedience cues, physical restraint, or suppression of communicative behaviour. Shake-offs frequently appeared during arousal transitions across reactive and non-reactive dogs. Reinforcement increased behavioural options at moments when dogs were processing sudden environmental change or elevated arousal, offering choice with less frustration and less caregiver interference. While the functional role of shake-offs remains hypothetical, the procedure provides a welfare-relevant option making mechanistic claims beyond the evidence.

These observations also highlight the practical value of a behaviour that is easy for caregivers to recognise and simple to reinforce. Shake-offs may offer an accessible point of intervention that helps caregivers respond more consistently to emerging arousal shifts. These interpretations sit within several constraints, described in the following Limitations section.

Limitations

These interpretations rely on practice-based observations rather than controlled research. Data were not collected using formal coding, blinded scoring, or comparison groups, which limits reliability and generalisability. Individual variability was substantial: Dogs differed in how often they shook off, the contexts in which the behaviour appeared, and the behavioural sequences that followed. Not every shake-off reflects a regulatory process, and interpreting the movement without considering context would be inappropriate, particularly in cases involving fear, pain, or underlying medical issues.

Both the core and complementary strategies were implemented within multi-component behaviour modification plans that included environmental management, reinforcement of alternative behaviours, caregiver coaching, and, in some cases, medication adjustments and physical health considerations. These overlapping influences make it impossible to identify causal contributions from shake-off reinforcement alone. Recognising these limitations helps clarify where future research is most needed.

Future research

Future work should examine whether reinforcing shake-offs influences their timing or availability during arousal changes. Studies incorporating physiological measures, such as heart rate variability, pupil dilation, or salivary cortisol, may help clarify whether shake-offs correspond to identifiable autonomic changes. Structured observations in shelters, veterinary contexts, and controlled exposure environments could refine understanding of when shake-offs occur and how they relate to behavioural transitions. Longitudinal caregiver reports and client-submitted video may further illuminate generalisation and day-to-day variability. Mechanistic work exploring whether shake-offs share properties with other fixed motor patterns may also clarify the contributions of neuromodulatory systems.

Another promising line of inquiry concerns the use of a conditioned “safe” cue, as described in the section on complementary strategies. Although anecdotal practice exists in various training traditions, its function has not been tested scientifically in dogs. Research could evaluate whether a cue trained as a conditioned inhibitor of threat (Rescorla, 1969) can be established reliably in applied settings, whether it contributes to behavioural recovery, and whether it increases the likelihood of a shake-off during early arousal transitions. Such work would help determine whether this tool is a useful evidence-based addition to behaviour modification plans or whether its effects are limited to specific contexts or individuals.

Together, these directions would help determine whether shake-offs reliably support emotional regulation, how they interact with arousal processes, and how they might be incorporated into evidence-informed training. These considerations lead directly to the Conclusion, which reflects on the potential value and boundaries of these practice-based impressions.

Conclusion

The practice-based observations reviewed here suggest that shake-offs may mark meaningful transition points during shifts in arousal and can be followed by clearer behavioural recovery. When reinforced, some dogs began to shake off earlier in sequences that had previously escalated, creating a naturally occurring moment that supported disengagement from triggers and a return to under-threshold behaviours. Reinforcing shake-offs can support agency because the behaviour is initiated by the dog, its timing is chosen by the dog rather than imposed by cues or physical intervention, and it allows the dog to return to voluntary movement and information-gathering. Reinforcement increases access to a dog-selected transitional behaviour without suppressing communication signals or requiring restraint or obedience compliance from the handler. This aligns well with choice-based, non-aversive approaches that prioritise agency and emotional safety. Here, emotional safety refers to conditions in which the dog remains under threshold, and able to observe, disengage, and recover.

It is striking that such a simple, naturally occurring behaviour may hold practical value in helping dogs navigate changes in arousal. A fuller scientific understanding of shake-offs—and of how they relate to anxiety, arousal, and recovery—could help refine how this behaviour is integrated into training and behaviour modification plans. Controlled studies will be essential to determine whether these patterns generalise across individuals and contexts, and to clarify the mechanisms that underlie them.

Ethics Statement

This article reports practice-based observations from dogs seen in routine training and behaviour consultation sessions, along with the author’s own dog, and is intended as hypothesis-generating rather than a controlled research study. No procedures were conducted beyond standard training, management, and behaviour-support work. Caregivers provided informed consent for the inclusion of case material, and additional permission was obtained for any video content shared.

Acknowledgments

Krisztina Harasztosi developed this article based on original techniques and practice-based field observations. Language refinement, citation formatting, and structural editing were supported in collaboration with ChatGPT-4. Kevin Hamilton generously provided editorial and scientific phrasing advice.

References

Aloff, B. (2002). Aggression in dogs: Practical management, prevention, and behaviour modification. Dogwise Publishing. https://search.worldcat.org/title/52150138

Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., & de Vries, H. W. (1997). Manifestations of chronic and acute stress in dogs. Applied Animal Behaviour Science, 52(3–4), 307–319. https://doi.org/10.1016/S0168-1591(96)01131-8

Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., de Vries, H. W., & Mol, J. A. (1998). Behavioral, saliva cortisol and heart rate responses to different types of stimuli in dogs. Applied Animal Behaviour Science, 58(3–4), 365–381. https://doi.org/10.1016/S0168-1591(97)00145-7

Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., & de Vries, H. W. (1999). Chronic stress in dogs subjected to social and spatial restriction. Physiology & Behavior, 66(2), 233–242. https://doi.org/10.1016/S0031-9384(98)00289-3

Bryce, C. M., Nurkin, P., & Horowitz, A. (2024). Shake it off: Investigating the function of a domestic dog behavior in social contexts. Animals, 14(22), 3248. https://doi.org/10.3390/ani14223248

Cisneros, A., Carroll, A. D., Moody, C. M., & Stellato, A. C. (2025). Handle with care: Dogs show negative responses to restrictive handling restraints and tools during routine examinations. Applied Animal Behaviour Science, 286, 106601.https://doi.org/10.1016/j.applanim.2025.106601

Dickerson, A. K., Mills, Z. G., & Hu, D. L. (2012). Wet mammals shake at tuned frequencies to dry. Journal of the Royal Society Interface, 9(75), 3208–3218. https://doi.org/10.1098/rsif.2012.0429

DogFieldStudy. (2024). Pulse study: At the heart of the walk. https://www.dogfieldstudy.com/en/study/pulse-study-at-the-heart-of-the-walk

Fernandez, E. J. (2024). The least inhibitive, functionally effective (LIFE) model: A new framework for ethical animal training practices. Journal of Veterinary Behavior, 71, 63–68. https://doi.org/10.1016/j.jveb.2023.12.001

Handelman, B. (2008). Canine behavior : a photo illustrated handbook (First edition). Wolf and Word Press. https://search.worldcat.org/en/title/227917614

Horwitz, D. F. (2018). Blackwell’s five-minute veterinary consult clinical companion: Canine and feline behavior. Wiley-Blackwell. https://search.worldcat.org/en/title/957554882

Joint Standards of Practice Coalition. (2025). Joint Standards of Practice and Professional Code of Ethics. IAABC https://iaabc.org/standards-of-practice

Kaiser, L., Heleski, C. R., Siegford, J., & Smith, K. A. (2006). Stress-related behaviors among horses used in a therapeutic riding program. Journal of the American Veterinary Medical Association, 228(1), 39–45. https://doi.org/10.2460/javma.228.1.39

Kalueff, A. V., & Tuohimaa, P. (2005). The grooming analysis algorithm discriminates between different levels of anxiety in rats: Potential utility for neurobehavioural stress research. Journal of Neuroscience Methods, 143(2), 169–177. https://doi.org/10.1016/j.jneumeth.2004.10.001

Longstaff, A. (2011). BIOS Instant Notes in Neuroscience (3rd ed). Taylor and Francis. https://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=4634137, https://search.worldcat.org/en/title/958550087

Luo, Y., Wang, L., Yang, L., Li, X. F., Anselme, P., Wang, X., Tian, X., & Li, Z. (2022). Using a behaviour random permutation model to identify displacement grooming in ungulates. Current Zoology, 69(2), 200–207. https://doi.org/10.1093/cz/zoac035

Mueller, S. (2020). Hunting together: Harnessing predation for connection and play (Predation Substitute Training). https://mybook.to/HuntingTogether

Overall, K. L. (1997). Clinical behavioral medicine for small animals. Mosby. https://search.worldcat.org/en/title/34731581

Pastore, C., Pirrone, F., Balzarotti, F., Faustini, M., Pierantoni, L., & Albertini, M. (2011). Evaluation of physiological and behavioral stress-dependent parameters in agility dogs. Journal of Veterinary Behavior, 6(3), 188–194. https://doi.org/10.1016/j.jveb.2011.01.001

Reamer, L., Tooze, Z., Coulson, C., & Semple, S. (2010). Correlates of self-directed and stereotypic behaviours in captive red-capped mangabeys (Cercocebus torquatus torquatus). Applied Animal Behaviour Science, 124(1–2), 68–74. https://doi.org/10.1016/j.applanim.2010.01.012

Rescorla, R. A. (1969). Pavlovian conditioned inhibition. Psychological Bulletin, 72(2), 77–94. https://doi.org/10.1037/h0027760

Rugaas, T. (2006). On talking terms with dogs: Calming signals (2nd ed.). Dogwise Publishing. https://search.worldcat.org/en/title/63519566

Spruijt, B. M., van Hooff, J. A. R. A. M., & Gispen, W. H. (1992). Ethology and neurobiology of grooming behavior. Physiological Reviews, 72(3), 825–852. https://doi.org/10.1152/physrev.1992.72.3.825

Stewart, G. (2016). Behavior Adjustment Training 2.0: New practical techniques for fear, frustration, and aggression in dogs. Dogwise Publishing. https://search.worldcat.org/en/title/935989340

Talamonti, Z., Cassis, C., Brambilla, P. G., Scarpa, P., Stefanello, D., Grossi, S., Minero, M., & Palestrini, C. (2015). Preliminary study of pet owner adherence in behaviour, cardiology, urology, and oncology fields. Veterinary Medicine International, 2015, Article 618216. https://doi.org/10.1155/2015/618216

Vilela, S., Chen, P., Murakami, S., Tanabe, Y., & Yamamoto, S. (2025). Self-directed behavior reflects social stress in captive Asian elephants. Frontiers in Veterinary Science, 12, 1629664. https://doi.org/10.3389/fvets.2025.1629664

Wascher, C. A. F., Fraser, O. N., & Kotrschal, K. (2010). Heart rate during conflicts predicts post-conflict stress-related behavior in greylag geese. PLOS ONE, 5(12), e15751. https://doi.org/10.1371/journal.pone.0015751

Wood, W. (2019). Good habits, bad habits : the science of making positive changes that stick (First edition). Farrar, Straus and Giroux. https://search.worldcat.org/en/title/1082550628