Environmental Enrichment in Dogs: What It Is, What It Changes and How to Measure It

1. What Enrichment Means

1.1 A Functional Definition

In everyday use, "enrichment" has come to mean a product category: anything that keeps a dog busy. The scientific term is narrower and more demanding. In the definition most often cited in the applied ethology literature, environmental enrichment is "an improvement in the biological functioning of captive animals resulting from modifications to their environment" (Newberry, 1995).

The important word is improvement. Under this definition, a change to the environment only counts as enrichment if it produces a measurable benefit for the animal. Putting a toy into a kennel is a modification. Whether it is enrichment is an empirical question, and the answer can be no. Newberry also argued that the goal of enrichment is to increase the biological relevance of the environment, so that the animal can perform behavior that matters to it, rather than to fill time or to make the environment look more interesting to the people who care for it.

This distinction runs through the whole literature. Many studies test a modification and then ask whether it changed behavior. Fewer ask whether the change was an improvement in welfare, and fewer still can answer that question with confidence.

1.2 The Usual Categories

Reviews of enrichment for kennelled dogs sort the options into broad groups: social enrichment, which means contact with other dogs or with people; object enrichment, mainly toys and chews; food-based enrichment, such as food-dispensing toys and scatter feeding; sensory enrichment, which covers sound and scent; and changes to the physical space, such as kennel size, furniture or visual access (Wells, 2004). Cognitive enrichment, meaning tasks that require learning or problem solving, is often treated as a separate category, although in practice it overlaps with food-based enrichment and training.

The categories are useful for organizing studies, but they hide an important point. Each category bundles very different experiences. A chew toy, a ball and a puzzle toy are all "objects", and a five-minute play session with a stranger and a week of foster care are both "social contact". The outcome of a study depends on the specific version that was tested, not on the category label.

1.3 Enrichment and Positive Welfare

Modern welfare frameworks give enrichment a more specific role. The Five Domains Model assesses welfare through nutrition, environment, health and behavior, which feed into a fifth domain, the animal's mental state (Mellor, 2017). An extension of the model shifted attention from minimizing negative states toward providing opportunities for positive ones, such as engagement, exploration and play (Mellor & Beausoleil, 2015). Seen through this framework, enrichment is one of the main ways to create such opportunities (how welfare and quality of life are assessed).

A related line of work emphasizes agency, the capacity of an animal to make choices and to act on its environment, and argues that opportunities to exercise agency should be part of how positive welfare is assessed (Littlewood, Heslop & Cobb, 2023). This matters for enrichment because it shifts the question from "what does the dog receive?" to "what can the dog do?". A toy that is placed in front of a dog and a choice the dog makes about where to rest are different kinds of enrichment, even if both are counted under the same heading.

2. Where the Evidence Comes From

2.1 Kennels, Laboratories and Shelters

Most controlled research on enrichment in dogs was carried out in three settings: laboratory kennels, research colonies and rescue shelters. This is not an accident. These are the places where dogs live in environments that are clearly restricted, where housing is standardized enough to allow comparisons, and where welfare concerns created a practical need for improvement.

The setting matters for interpretation. In a barren kennel, the baseline is low. A dog with little to do, limited social contact and an unfamiliar or noisy environment has a lot of room for improvement, and almost any meaningful addition may change behavior. In a home where a dog already has walks, social contact, varied surroundings and play, the same addition starts from a different baseline. Findings from kennels describe what enrichment can do in a restricted environment. They do not automatically describe what it adds to a companion dog's life.

Throughout this article, shelter and laboratory studies are treated as what they are: studies of particular populations under particular housing conditions. Their results are informative about mechanisms and about what dogs respond to, but the size of the effects cannot be transferred directly to dogs living in homes.

2.2 What the Classic Review Concluded

A widely cited review of enrichment for kennelled dogs drew several conclusions that still shape the field (Wells, 2004). Social contact, with other dogs or with people, was described as the most important form of enrichment for dogs in kennels. Toys had more limited value: adult dogs tended to lose interest in a toy within days, and in shelter settings many dogs paid little attention to the toys they were given. Research on olfactory and auditory stimulation was described as preliminary at the time.

Much of the later research can be read as testing these conclusions. Studies on habituation to toys looked at why interest fades. Studies on sound and scent added data to the preliminary picture. Studies on social contact measured physiological effects that the early behavioral work could only infer.

2.3 An Early Laboratory Comparison

One of the earliest systematic comparisons was carried out with 48 laboratory beagles housed under different conditions, including increased social contact, brief daily handling and the provision of toys and chews (Hubrecht, 1993). Dogs that received toys and chews used them, and their behavior changed noticeably compared with controls, including less chewing on parts of the kennel. At the same time, some forms of interaction with kennel mates decreased.

This study is often cited as evidence that toys help kennelled dogs, but it shows something more interesting. Adding an object did not simply add a new activity on top of existing behavior. It redistributed behavior. Time spent with the toy was time not spent on something else, in this case partly on social interaction. Whether that trade is an improvement depends on what the dog gave up, which the behavioral data alone cannot settle.

2.4 Food Toys in Laboratory Dogs

A later pilot study compared eight laboratory dogs that received a food-filled rubber toy with nine control dogs (Schipper, Vinke, Schilder & Spruijt, 2008). The dogs with the toy showed more appetitive behavior, more activity and more frequent transitions between behaviors, and they barked less. These effects were short-term: they were measured around the time the toy was available and did not show that the dogs' overall welfare had changed.

The study is small, but its design is typical. Many enrichment studies measure behavior during or shortly after the intervention and find a change. Fewer follow dogs over weeks, and fewer still measure outcomes that are clearly linked to welfare rather than simply to activity.

3. Novelty and Habituation

3.1 Why Interest in a Toy Fades

Anyone who has bought a dog a new toy knows the pattern: intense interest on the first day, less on the second, and soon the toy lies in a corner. The review literature already noted that adult kennelled dogs lose interest in toys over several days (Wells, 2004). An experimental study looked at this more closely.

Sixteen kennel-housed Labrador retrievers were given repeated play sessions with toys (Pullen, Merrill & Bradshaw, 2012). Play declined across sessions, the classic pattern of habituation. Lengthening the interval between sessions, from seconds to minutes, did not prevent this decline. What restored play was a change to the toy itself: when the toy was replaced with one that differed in color or odor, the dogs played again, and this happened regardless of which type of contrast was used. The authors concluded that dogs habituate to the toy as a whole rather than to a specific feature.

3.2 What Follows from Habituation

Habituation is not a defect in the dog or in the toy. It is a basic learning process that allows animals to stop responding to stimuli that turn out to have no consequence. A toy that never does anything new becomes, for the dog, part of the background.

Two practical consequences follow. First, the effect of an object on behavior is largely an effect of novelty, and novelty decays. Second, the effect can be renewed by change, which is the rationale for rotating toys rather than leaving all of them out permanently. The study on Labradors supports rotation as a principle. It does not establish an optimal rotation schedule, and it was carried out in kennel-housed dogs, so the speed of habituation in a home with many other sources of stimulation is unknown.

Food-dispensing toys can maintain engagement longer than an unchanged empty object because the dog's behavior can continue to produce consequences and reinforcement. This does not amount to a general direct comparison showing that all active tasks hold attention longer than static objects. In learning terms, a food toy is less a simple object than a small operant task (how reinforcement schedules maintain behavior).

3.3 Novelty Is Not Always Welcome

Novelty has a second side. For a confident dog, a new object is an invitation to explore. For a fearful dog, it can be a source of uncertainty. The enrichment studies discussed here rarely report individual differences in response to novelty, but the principle matters in practice: an intervention that relies on novelty assumes that the dog experiences novelty as attractive. For some dogs, especially in a new environment, the more useful enrichment may be predictability and the chance to withdraw rather than another new object.

4. Sensory Enrichment: Sound and Scent

4.1 Auditory Stimulation

Sound is one of the most studied forms of sensory enrichment in kennelled dogs, partly because it is cheap and easy to deliver. In an early shelter study, dogs exposed to classical music spent more time resting and less time vocalizing, while heavy metal music was associated with more agitated behavior (Wells, Graham & Hepper, 2002). Audiobooks were later associated with more resting in dogs at a rehoming kennel (Brayley & Montrose, 2016).

These findings are consistent, but they share limits: shelter populations, short observation periods and behavioral outcomes that measure resting and vocalizing rather than welfare directly. The research on sound is covered in more detail elsewhere (what the studies on music and auditory enrichment show).

4.2 Olfactory Stimulation

Scent is the sensory channel most closely tied to how dogs explore the world (how olfaction shapes dog behavior). It was therefore a natural candidate for enrichment research.

In a study of 55 shelter dogs, cloths scented with different essential oils were hung in the kennels (Graham, Wells & Hepper, 2005). Lavender and chamomile were associated with more resting and less moving and vocalizing. Rosemary and peppermint were associated with more standing, moving and vocalizing. The scents therefore did not simply "enrich" in a uniform way. Different odors were associated with different activity patterns, some calmer and some more active.

A later study exposed 15 shelter dogs to several odors in a fixed order (Binks, Taylor, Wills & Montrose, 2018). Vanilla, coconut, ginger and valerian were associated with less vocalizing and less movement, and coconut and ginger with more sleeping. Because all dogs received the odors in the same order, effects of time in the shelter or of habituation to the procedure cannot be separated from effects of the individual odors.

A third study compared olfactory and auditory enrichment in 60 shelter dogs (Amaya, Paterson & Phillips, 2020). Dogs exposed to a dog-appeasing pheromone product or to music spent more time lying down, and control dogs vocalized roughly three to four times as much as dogs in the treatment groups. Within the five days of the study, the authors did not observe habituation to the treatments. Pheromone products as a separate topic are discussed elsewhere (what is known about pheromones and supplements).

4.3 What the Sensory Studies Measure

Taken together, the sensory studies show that sound and scent in a kennel are associated with changes in activity and vocalization. They are less clear about what those changes mean. More resting can reflect relaxation, but in a kennel it can also reflect withdrawal or reduced arousal for other reasons. Less vocalizing is welcome for staff and neighbors, and it may be a sign of reduced stress, but vocalizing has many causes.

This is a recurring problem in enrichment research: the outcome measures are behaviors that are easy to count, and their link to the dog's emotional state is an interpretation (why behavior is not a direct readout of emotion). The interpretation is often plausible, but it is not measured.

For companion dogs, a further caution applies. The odors in these studies were presented passively in a kennel. Scent work, where a dog actively searches for an odor, is a different kind of experience and is discussed in section 6.

5. Social Enrichment

5.1 Contact with People

The conclusion that social contact is the most important enrichment for kennelled dogs (Wells, 2004) has been supported by physiological data. In dogs newly admitted to a shelter, sessions with a person, whether passive presence, petting or play, were associated with lower plasma cortisol than control conditions, along with less vocalizing and panting (Shiverdecker, Schiml & Hennessy, 2013). The study did not find that one type of interaction was clearly superior. The presence of a person and the interaction itself appeared to matter more than the specific activity.

A larger study of a temporary fostering program measured cortisol in 207 shelter dogs before, during and after one- or two-night stays in a home (Gunter, Feuerbacher, Gilchrist & Wynne, 2019). Cortisol ratios dropped during fostering and returned to baseline after the dogs went back to the shelter; the longest uninterrupted rest bouts were also observed during the home stays. The distinction matters: fostering changed the entire environment, not just human contact. The study therefore supports an effect of temporary foster placement, not an isolated effect of social contact.

The mechanism behind these effects is often described as social buffering, the reduction of stress responses through the presence of a social partner (what is known about social buffering in dogs).

5.2 Contact with Other Dogs

Contact with other dogs is a form of enrichment as well, but the evidence is more mixed than for contact with people. In the early laboratory comparison, social housing and contact were part of the conditions tested (Hubrecht, 1993), and later reviews list group or pair housing as a major form of enrichment in kennels (Wells, 2004). In a pilot study of dogs in an assistance-dog training program, peer play was among the activities associated with the largest increase in relaxation behaviors (Hunt, Whiteside & Prankel, 2022).

The benefit of dog-dog contact depends heavily on the partner and the setting. A compatible companion can be a source of play and security. An incompatible one can be a source of chronic tension. The research on kennelled dogs typically controls compatibility through careful pairing, which is not the same as a group of unfamiliar dogs in a park.

5.3 What This Means Outside a Kennel

For companion dogs, the social dimension of enrichment is easy to overlook because it is already present in most homes. That is precisely why it is the least likely to be missing. The studies in this section suggest that, for a dog in a restricted environment, the most powerful addition is often a person rather than an object. For a dog in a home, the question is less whether social contact exists and more what form it takes: shared activity, calm presence, and predictable interaction are all part of it.

6. Food, Foraging and Work

6.1 Contrafreeloading

Food-based enrichment is often justified by a phenomenon called contrafreeloading: the tendency, documented in many species, to work for food even when the same food is freely available. If animals prefer to work for food, then feeding them in a way that requires effort would give them something they value.

A study in companion dogs tested this directly (Rothkoff, Feng & Byosiere, 2024). Thirty-eight pet dogs were offered a choice in their own homes between a tray with freely available food and a snuffle mat containing the same amount of food. Only one dog showed a strong preference for the snuffle mat. Twenty-two dogs preferred the free food. At the same time, 30 of the 38 dogs were willing to use the snuffle mat even while free food was available. Dogs with a higher body condition score chose the snuffle mat first more often, a result the authors linked to differences in motivation for food.

The authors summarized this as willingness without preference. Companion dogs in this study did not show the strong contrafreeloading reported in many other species. They were, however, willing to engage with the foraging device. This matters for how food-based enrichment is justified. The evidence does not support the claim that dogs generally prefer to work for their food. It supports the more modest claim that many dogs will engage with such devices.

6.2 What Food Toys Do and Do Not Show

Food toys change behavior in kennelled dogs, at least in the short term (Schipper et al., 2008). In a companion-dog pilot study, however, food-based activities were among those with the smallest effect on relaxation behavior, compared with peer play and a playhouse (Hunt et al., 2022). In a controlled companion-dog study discussed below, short-term toy provision did not significantly change the measured outcomes; because of an experimental error, assignment to toy-complexity levels was only pseudo-randomized.

Food toys are therefore not useless, but the claim that they are the default solution for an under-stimulated dog is stronger than the evidence. Their main effect seems to be that they extend eating time and add a small task. Whether that is an improvement depends on the dog, the device and what the dog would otherwise be doing.

6.3 Scent Work as Active Enrichment

Scent work, in which dogs actively search for a target odor, differs from passive exposure to odors. In a study with companion dogs, two weeks of nosework training shifted a cognitive bias measure in an optimistic direction compared with a control group that practiced heelwork (Duranton & Horowitz, 2019). Cognitive bias tasks are used as an indirect measure of affective state, and their interpretation has its own limits (what cognitive bias tests can and cannot show).

A scoping review of scent activities found a growing but uneven literature, and concluded that the mechanisms through which scent work might benefit dogs are incompletely understood (Fountain, Fernandez, McWhorter & Hazel, 2025). A later study measured heart rate and behavior in 21 dogs during and after scent work (Fountain et al., 2026). Heart rate increased during the search, without an increase in stress-related behavior, and there was no difference in the period after the task. This fits the view that active scent work is engaging rather than relaxing while it is happening, and that the claim of a lasting calming effect is not established by these data.

Scent work has some initial evidence of possible benefits in companion dogs, but the evidence base remains narrow. Existing studies use small samples and different outcome measures, and they do not establish a general or lasting calming effect.

7. Cognitive Enrichment and the Value of Achievement

7.1 Solving a Problem as a Reward

A study of 12 beagles tested whether solving a task is rewarding in itself, beyond the reward obtained (McGowan, Rehn, Norling & Keeling, 2014). Dogs that learned to operate a device in order to gain access to a reward were compared with conditions in which the dogs received the same reward without having to solve the task. Dogs in the learning condition showed more tail wagging and higher activity, which the authors interpreted as signs of positive emotional arousal linked to achievement.

The study is small and the behavioral indicators require interpretation, but its central point is important for enrichment: what may matter is not only what the dog receives, but whether the dog's own behavior made it happen. This links cognitive enrichment to the broader concept of control (why control and predictability matter for stress).

7.2 Enrichment and Aging

The most extensive long-term data on cognitive enrichment come from research on aging dogs. In a two-year longitudinal study, aged beagles received a fortified diet, behavioral enrichment, both, or neither (Milgram et al., 2005). Behavioral enrichment and dietary fortification each improved performance on learning tasks compared with control dogs, and the combination produced the best results.

These are among the strongest long-term data showing that a structured, ongoing enrichment program can have measurable effects in dogs over time, but they come from a laboratory population with structured cognitive testing as part of the enrichment package. They show that a combined program of stimulation can slow age-related decline in learning ability in that setting. They do not show that a specific toy or game does the same for a companion dog. The research on age-related cognitive decline is covered elsewhere (what is known about cognitive dysfunction in aging dogs).

7.3 Training as Enrichment

Training is rarely described as enrichment in the research literature, but it shares many of the features that the enrichment studies identify as important: social interaction, problem solving, control over outcomes and food or play as reinforcement. In the study on nosework and cognitive bias, the control group practiced heelwork, a form of obedience training, and did not show the same shift (Duranton & Horowitz, 2019). This does not show that training lacks enrichment value. It shows that, in that comparison, the type of activity mattered.

In practice: Training sessions that give the dog clear options and frequent success tend to look more like enrichment than sessions that are mainly about compliance. This is an observation from training practice, not a finding from controlled studies.

8. Companion Dogs: Where the Kennel Findings Stop

8.1 A Controlled Trial in Homes

The most direct controlled test of toy enrichment in companion dogs to date comes from Chan, Arellano and Horowitz (2023). Forty-seven dogs were enrolled; seven did not complete the study and were excluded from the analyses. Dogs were initially randomized to an experimental or control group, but because of an experimental error, assignment within the experimental group to different toy-complexity levels was only pseudo-randomized. Over 12 days, dogs received five minutes of daily access to the assigned toys. Activity, eating speed and performance in a cognitive-bias task were measured.

Providing toys did not significantly change activity, eating speed or cognitive bias. More complex toys showed a trend toward a more optimistic cognitive bias, but the difference was not statistically significant. The title of the paper summarizes the result: short-term provision of toys did not improve the welfare measures used in this study.

The study has limitations. Five minutes a day over 12 days is a brief intervention, seven dogs did not complete the study, and assignment to toy-complexity levels was not fully randomized. The welfare measures may also have missed small changes. The result therefore shows that this specific short-term toy intervention did not significantly alter the selected welfare measures in the companion dogs studied; it does not show that toy enrichment is generally ineffective in pet dogs.

8.2 A Mixed Setting

A pilot study in an assistance-dog training program offers a setting between kennel and home (Hunt et al., 2022). Seven enrichment activities were tested in an office environment, including bonding activities, bubbles, peer play, interactive toys, a playhouse, food-dispensing toys and tug games. Overall, enrichment was associated with an increase in relaxation behaviors. Peer play and the playhouse had the largest effects, while food-based activities had the smallest.

As a pilot study, it cannot establish which activity is best, but it is consistent with the idea that the type of enrichment matters and that the most commonly sold product category, food toys, is not necessarily the most effective.

8.3 Association Studies

Large owner surveys add a different kind of evidence. In one Finnish survey, less daily exercise was associated with noise sensitivity and separation anxiety, and early life experiences were associated with later fearfulness (Tiira & Lohi, 2015). Such findings are often cited as evidence that more activity or enrichment prevents behavior problems.

The data do not support that conclusion directly. The surveys are cross-sectional and based on owner reports, so the direction of the association is unclear: a fearful dog may be walked less because walks are difficult, rather than becoming fearful because it is walked less (what the exercise research shows and does not show). Associations between environment and behavior are consistent with a role for enrichment, but they do not test it.

8.4 Why the Transfer Is Difficult

Several factors make the transfer from kennel to home difficult. The baseline in a home is usually richer, so the same intervention has less room to act. Companion dogs differ more from each other in age, breed, experience and daily routine than research colony dogs do, which increases variability. And owners often provide many forms of enrichment at once, so the contribution of any single element is hard to isolate.

None of this means that enrichment is irrelevant for companion dogs. It means that the effect sizes reported in kennel studies are not a guide to what a toy or food puzzle will add to a dog that already has a varied life, and that the most relevant question for a companion dog is often not "what can I add?" but "what is missing?".

9. When Enrichment Becomes a Burden

9.1 Frustration

Enrichment is usually discussed as if it could only help. But an enrichment device creates a task, and a task can fail. A puzzle that is too difficult, a food toy that releases nothing or a ball the dog cannot reach can produce frustration rather than engagement. Frustration is an emotional state with its own neurobiology and behavioral signs, including increased arousal, persistence and redirected behavior (what is known about frustration in dogs).

The enrichment studies reviewed here rarely measure frustration directly, but the principle follows from learning theory: a solvable task can support engagement and a sense of success, whereas repeated failure can produce frustration. In practice: Difficulty is best set so that the dog succeeds most of the time, and increased only when the dog solves the current level easily.

9.2 Arousal

Some forms of enrichment raise arousal. Scent work increased heart rate during the search (Fountain et al., 2026), and the beagle study on achievement described positive arousal (McGowan et al., 2014). Raised arousal during an activity is not a problem in itself. It becomes one when a dog is already highly aroused, when activities are stacked without recovery, or when the dog has difficulty calming down afterward (how arousal regulation works).

For some dogs, the most valuable enrichment is not another activity but the opportunity for undisturbed rest. A daily schedule that is filled with puzzles, games and training can become a source of load rather than relief.

9.3 Enrichment and Repetitive Behavior

Repetitive and compulsive behaviors are often cited as a reason for enrichment. In dogs, however, the relationship between environment and repetitive behavior is not straightforward. Functional analyses have shown that what maintains a repetitive behavior differs between dogs: in some, the owner's response maintained the behavior, in others the stimulus itself (Hall, Protopopova & Wynne, 2015). Dogs with stereotypic behavior were also more persistent during extinction than matched controls in one experiment (Protopopova, Hall & Wynne, 2014).

For enrichment, this means two things. First, adding activities may reduce the time available for a repetitive behavior, but it does not necessarily address what maintains it. Second, enrichment that resembles the repetitive behavior, such as light or shadow games in a dog that chases lights, may strengthen the pattern rather than replace it. Repetitive behavior needs an individual analysis, not a general increase in stimulation (what the research on compulsive behavior shows).

9.4 Predictability

Enrichment research often emphasizes variety. But variety and predictability pull in different directions. A review of predictability in captive animals found that predictable conditions are generally associated with better welfare than unpredictable ones, and that unpredictability is one of the conditions under which stereotypies develop (Bassett & Buchanan-Smith, 2007).

The balance lies in what is varied. A dog may benefit from new things to explore within a predictable daily structure, where it knows when it will be fed, walked and left to rest. Constant change in routine is not enrichment, especially for anxious dogs.

10. How to Tell Whether It Works

10.1 Engagement Is Not the Same as Benefit

The most common measure of enrichment is engagement: whether the dog uses the toy or device. Engagement is a necessary condition for benefit, but it is not sufficient. A dog can engage intensely with something that frustrates it, and a dog that ignores a toy may simply have no need for it.

The definition at the start of this article requires an improvement in functioning (Newberry, 1995). For a particular dog, the relevant questions are therefore not only whether the dog uses the item, but whether something in the dog's broader behavior changes in a desirable direction.

10.2 Indicators That Can Be Observed

Researchers use several kinds of outcome measures: resting and activity patterns, vocalization, repetitive behavior, stress-related behaviors, physiological markers such as cortisol and heart rate, and indirect affective measures such as cognitive bias tasks (how stress is measured in dogs). Each has limits. Cortisol rises with positive arousal as well as with stress, and resting can reflect calm or withdrawal (what cortisol does and does not indicate).

For a dog at home, physiological measures are rarely available, but the behavioral indicators can be observed. Useful questions include whether the dog shows broadly unproblematic behavior during and after the activity, whether stress-related or repetitive behavior decreases over days and weeks when that is the goal, whether the dog seeks the activity out when given a choice, and whether the dog returns to it after novelty has worn off. Calmer behavior afterward may be desirable for some goals, but it is not a universal criterion for successful enrichment.

10.3 Choice as a Measure

Choice tests are one of the few ways to ask the dog directly. The snuffle-mat study used choice to show that many dogs in that sample were willing to work for food, while the majority preferred freely available food (Rothkoff et al., 2024). The same logic can be applied informally: offering two options and observing which the dog chooses over several days gives more information than observing whether the dog uses a single item that is presented without alternative.

Choice has its own limits. A dog may prefer what is immediately rewarding over what benefits it in the long term, and a choice in one situation may not reflect a stable preference. But as a complement to observation, it moves the question from what we think the dog likes to what the dog actually selects.

11. What Follows for Practice

11.1 Start with What Is Missing

The studies reviewed here suggest a different starting point from the usual product list. In restricted environments, social contact was described as the most important form of enrichment (Wells, 2004). For a companion dog, the first question is whether the basic needs for social contact, movement, exploration, rest and predictability are met. If one is missing, adding a toy is unlikely to compensate.

11.2 Principles Supported by Research

Several principles are supported by the studies reviewed here, even if their magnitude for companion dogs is not established. Variation can renew interest after dogs habituate to unchanged objects (Pullen et al., 2012). Tasks in which the dog's behavior repeatedly produces a consequence can maintain engagement, but there is no general direct evidence that active tasks as a class hold attention longer than static objects. Human interaction can reduce physiological stress responses in shelter dogs (Shiverdecker et al., 2013). Temporary foster stays were also associated with lower cortisol, but they changed the whole environment rather than social contact alone (Gunter et al., 2019). Achievement may be rewarding in itself (McGowan et al., 2014). And dogs differ: in the snuffle-mat study, the majority of this sample preferred free food, while many were still willing to use the mat (Rothkoff et al., 2024).

11.3 Principles from Practice

Several further principles come from practical experience and from learning theory rather than from controlled enrichment trials, and are marked as such here.

In practice: Enrichment works best when difficulty matches the dog's ability, when activities are rotated rather than permanently available, when the dog has a choice about whether to engage, and when there is enough recovery time between activities. A dog that seems restless despite many activities may need less stimulation rather than more.

In practice: The type of enrichment should match the dog. A dog with strong scenting motivation often responds well to search games, a social dog to shared activities, and a dog that is easily overwhelmed to calm, predictable routines with low-intensity options such as chewing or licking.

11.4 Observe and Adjust

Because the effect of enrichment cannot be assumed, it has to be observed. Watching how the dog behaves during and after an activity, and whether broader patterns change over weeks, gives more reliable information than the label on a product. If an activity repeatedly produces frustration, problematically high arousal or other unwanted changes, or fails to achieve the intended goal, it is reasonable to change its difficulty, duration or type. A calmer dog afterward is a useful outcome for some goals, not a universal test of whether enrichment works.

12. Summary at a Glance

Environmental enrichment is defined by its effect: a modification only counts as enrichment if it improves the animal's functioning (Newberry, 1995). Most controlled studies come from kennels, laboratories and shelters, where the baseline environment is poor and the room for improvement is large.

In those settings, social contact was described as a particularly important form of enrichment (Wells, 2004), and direct human interaction reduced cortisol in shelter dogs in a controlled study (Shiverdecker et al., 2013). A separate study found lower cortisol during short foster-home stays, but could not isolate that effect from the broader change in environment (Gunter et al., 2019). Toys changed behavior, but dogs habituated to them, and changing the toy renewed interest (Hubrecht, 1993; Pullen et al., 2012). Sound and scent were associated with changes in resting, activity and vocalization, with effects depending on the specific stimulus (Wells et al., 2002; Graham et al., 2005; Brayley & Montrose, 2016; Binks et al., 2018; Amaya et al., 2020).

In companion dogs, the evidence is thinner. In one choice study, the majority of the dogs tested preferred freely available food over a snuffle mat, while many were nevertheless willing to use the mat (Rothkoff et al., 2024). Short-term toy provision did not significantly change the measured welfare outcomes in a controlled study; assignment to toy-complexity levels was only pseudo-randomized (Chan et al., 2023). Scent work shifted a cognitive bias measure (Duranton & Horowitz, 2019), but its mechanisms remain incompletely understood (Fountain et al., 2025). Enrichment can turn into a burden through frustration, excessive arousal or a lack of predictability, and whether it works for a particular dog has to be observed rather than assumed.

13. Research Gaps and Critical Appraisal

13.1 Population and Setting

The largest gap is the mismatch between where the evidence comes from and where it is applied. Most studies were conducted in kennels, laboratories and shelters, while most enrichment products are sold for companion dogs. The few controlled studies in companion dogs show mixed or absent effects, and it is unclear whether this reflects a ceiling effect, insensitive outcome measures, interventions that were too short or the specific interventions tested.

13.2 Sample Size and Design

Many enrichment studies are small, some with fewer than twenty dogs, and several use fixed-order designs in which treatment effects cannot be separated from time effects (Binks et al., 2018). Pilot studies dominate parts of the field (Schipper et al., 2008; Hunt et al., 2022). Results from such studies are useful for generating hypotheses but cannot establish effect sizes.

13.3 Outcome Measures

The outcome measures are mostly behaviors that are easy to observe, such as resting, activity and vocalization. Their link to welfare is plausible but interpretive. Physiological measures are used less often and have their own ambiguities, and cognitive bias tasks are indirect. Few studies combine several types of measures in the same dogs, and few follow dogs over long periods.

13.4 Individual Differences

Enrichment studies usually report group averages. The choice study on snuffle mats showed large differences between individual dogs (Rothkoff et al., 2024), and practical experience suggests that responses to novelty, scent and social contact vary widely. Research that identifies which dogs benefit from which form of enrichment is largely missing.

13.5 Long-Term Effects

Most studies measure effects during or shortly after an intervention. The main exception is the two-year study on aging beagles (Milgram et al., 2005), which used a combined program in a laboratory setting. Whether everyday enrichment in a home has lasting effects on behavior, stress resilience or cognitive aging is not established.

13.6 Commercial Influence

Enrichment is a commercial market, and the products that are most widely sold, such as food toys and puzzle feeders, are not necessarily those with the strongest evidence. The available companion-dog studies do not establish a clear advantage for these products over social or exploratory activities, and direct comparisons are scarce.

14. Conclusion

Environmental enrichment is a well-founded concept with a narrower evidence base than its popularity suggests. In restricted environments, social contact, objects, food tasks, sound and scent all change behavior, and social contact has the clearest physiological effects. Interest in objects fades through habituation and can be renewed by change. In companion dogs, evidence for added toys and food puzzles remains limited and mixed. In a single choice study, the majority preferred freely available food over a snuffle mat, although many dogs still used the mat.

The most useful lesson from the research is the definition itself: enrichment is not a product but an effect. Whether a toy, a game or an activity is enrichment for a particular dog depends on what that dog's environment lacks, how the dog responds and whether its behavior changes in a desirable direction over time. That makes enrichment less a shopping list than an ongoing observation.

Key Insights (Takeaways)

• Enrichment is defined by its effect: a change to the environment only counts if it improves the dog's functioning (Newberry, 1995).

• Most controlled evidence comes from kennels, laboratories and shelters, where the baseline is poor; effect sizes from these settings do not transfer directly to companion dogs.

• Social contact was described as the most important enrichment for kennelled dogs, and contact with people reduced cortisol in shelter dogs (Wells, 2004; Shiverdecker et al., 2013).

• Dogs habituate to toys as a whole, and changing the toy, not lengthening the interval, restored play (Pullen et al., 2012).

• Most pet dogs in one study preferred free food over a snuffle mat, but most were willing to use the mat (Rothkoff et al., 2024).

• Short-term toy provision did not significantly change the measured welfare outcomes in a controlled companion-dog study; assignment to toy-complexity levels was only pseudo-randomized (Chan et al., 2023).

• Enrichment can create frustration, excessive arousal or unpredictability, so whether it helps a particular dog has to be observed rather than assumed.

References

Amaya, V., Paterson, M. B. A., & Phillips, C. J. C. (2020). Effects of olfactory and auditory enrichment on the behaviour of shelter dogs. Animals, 10(4), 581. https://doi.org/10.3390/ani10040581

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

Binks, J., Taylor, S., Wills, A., & Montrose, V. T. (2018). The behavioural effects of olfactory stimulation on dogs at a rescue shelter. Applied Animal Behaviour Science, 202, 69–76. https://doi.org/10.1016/j.applanim.2018.01.009

Brayley, C., & Montrose, V. T. (2016). The effects of audiobooks on the behaviour of dogs at a rehoming kennels. Applied Animal Behaviour Science, 174, 111–115. https://doi.org/10.1016/j.applanim.2015.11.008

Chan, K., Arellano, C., & Horowitz, A. (2023). Here puppy, chew on this: Short-term provision of toys does not improve welfare in companion dogs. Animals, 13(21), 3340. https://doi.org/10.3390/ani13213340

Duranton, C., & Horowitz, A. (2019). Let me sniff! Nosework induces positive judgment bias in pet dogs. Applied Animal Behaviour Science, 211, 61–66. https://doi.org/10.1016/j.applanim.2018.12.009

Fountain, J., Fernandez, E. J., McWhorter, T. J., & Hazel, S. J. (2025). The value of sniffing: A scoping review of scent activities for canines. Applied Animal Behaviour Science, 282, 106485. https://doi.org/10.1016/j.applanim.2024.106485

Fountain, J., McWhorter, T. J., Seeley, M.-C., Bindoff, A. D., Handley, K., Hewings, R., & Hazel, S. J. (2026). Heart rate variability and behaviour in dogs during and after scent work. Applied Animal Behaviour Science, 299, 106986. https://doi.org/10.1016/j.applanim.2026.106986

Graham, L., Wells, D. L., & Hepper, P. G. (2005). The influence of olfactory stimulation on the behaviour of dogs housed in a rescue shelter. Applied Animal Behaviour Science, 91(1–2), 143–153. https://doi.org/10.1016/j.applanim.2004.08.024

Gunter, L. M., Feuerbacher, E. N., Gilchrist, R. J., & Wynne, C. D. L. (2019). Evaluating the effects of a temporary fostering program on shelter dog welfare. PeerJ, 7, e6620. https://doi.org/10.7717/peerj.6620

Hall, N. J., Protopopova, A., & Wynne, C. D. L. (2015). The role of environmental and owner-provided consequences in canine stereotypy and compulsive behavior. Journal of Veterinary Behavior, 10(1), 24–35. https://doi.org/10.1016/j.jveb.2014.10.005

Hubrecht, R. C. (1993). A comparison of social and environmental enrichment methods for laboratory housed dogs. Applied Animal Behaviour Science, 37(4), 345–361. https://doi.org/10.1016/0168-1591(93)90123-7

Hunt, R. L., Whiteside, H., & Prankel, S. (2022). Effects of environmental enrichment on dog behaviour: Pilot study. Animals, 12(2), 141. https://doi.org/10.3390/ani12020141

Littlewood, K. E., Heslop, M. V., & Cobb, M. L. (2023). The agency domain and behavioral interactions: Assessing positive animal welfare using the Five Domains Model. Frontiers in Veterinary Science, 10, 1284869. https://doi.org/10.3389/fvets.2023.1284869

McGowan, R. T. S., Rehn, T., Norling, Y., & Keeling, L. J. (2014). Positive affect and learning: Exploring the "Eureka Effect" in dogs. Animal Cognition, 17(3), 577–587. https://doi.org/10.1007/s10071-013-0688-x

Mellor, D. J. (2017). Operational details of the Five Domains Model and its key applications to the assessment and management of animal welfare. Animals, 7(8), 60. https://doi.org/10.3390/ani7080060

Mellor, D. J., & Beausoleil, N. J. (2015). Extending the 'Five Domains' model for animal welfare assessment to incorporate positive welfare states. Animal Welfare, 24(3), 241–253. https://doi.org/10.7120/09627286.24.3.241

Milgram, N. W., Head, E., Zicker, S. C., Ikeda-Douglas, C. J., Murphey, H., Muggenburg, B., Siwak, C., Tapp, D., & Cotman, C. W. (2005). Learning ability in aged beagle dogs is preserved by behavioral enrichment and dietary fortification: A two-year longitudinal study. Neurobiology of Aging, 26(1), 77–90. https://doi.org/10.1016/j.neurobiolaging.2004.02.014

Newberry, R. C. (1995). Environmental enrichment: Increasing the biological relevance of captive environments. Applied Animal Behaviour Science, 44, 229–243. https://doi.org/10.1016/0168-1591(95)00616-Z

Protopopova, A., Hall, N. J., & Wynne, C. D. L. (2014). Association between increased behavioral persistence and stereotypy in the pet dog. Behavioural Processes, 106, 77–81. https://doi.org/10.1016/j.beproc.2014.04.009

Pullen, A. J., Merrill, R. J. N., & Bradshaw, J. W. S. (2012). Habituation and dishabituation during object play in kennel-housed dogs. Animal Cognition, 15(6), 1143–1150. https://doi.org/10.1007/s10071-012-0538-2

Rothkoff, L., Feng, L., & Byosiere, S.-E. (2024). Domestic pet dogs (Canis lupus familiaris) do not show a preference to contrafreeload, but are willing. Scientific Reports, 14, 1314. https://doi.org/10.1038/s41598-024-51663-x

Schipper, L. L., Vinke, C. M., Schilder, M. B. H., & Spruijt, B. M. (2008). The effect of feeding enrichment toys on the behaviour of kennelled dogs (Canis familiaris). Applied Animal Behaviour Science, 114, 182–195. https://doi.org/10.1016/j.applanim.2008.01.001

Shiverdecker, M. D., Schiml, P. A., & Hennessy, M. B. (2013). Human interaction moderates plasma cortisol and behavioral responses of dogs to shelter housing. Physiology & Behavior, 109, 75–79. https://doi.org/10.1016/j.physbeh.2012.12.002

Tiira, K., & Lohi, H. (2015). Early life experiences and exercise associate with canine anxieties. PLOS ONE, 10(11), e0141907. https://doi.org/10.1371/journal.pone.0141907

Wells, D. L. (2004). A review of environmental enrichment for kennelled dogs, Canis familiaris. Applied Animal Behaviour Science, 85(3–4), 307–317. https://doi.org/10.1016/j.applanim.2003.11.005

Wells, D. L., Graham, L., & Hepper, P. G. (2002). The influence of auditory stimulation on the behaviour of dogs housed in a rescue shelter. Animal Welfare, 11(4), 385–393. https://doi.org/10.1017/S0962728600025112