In Canada and the US there are an estimated 70 million pet cats (1, 2), with many cat-owning households consisting of two cats [average of 1.6 cats/ Canadian household, average of 1.8 cats/ US household; (1, 2)]. Although current house cats descend from a wildcat thought to be largely solitary (Felis silvestris lybica), partial domestication has adapted cats to group living associated with food availability near human settlements (3). As cats made their way into our homes, owners often choose to house cats with conspecifics, without understanding the impact on the cats themselves. Thus, multi-cat households vary in composition, with cats housed together that are unrelated, related, acquired together, and/or acquired separately. The welfare of pet cats in multi-cat households is dependent on many factors including their relationship with conspecifics (4, 5), the physical home environment (6–8), and caretaker interactions (4, 9). Cat-cat interactions include positively valenced interactions such as affiliative behaviors (example: allo-grooming, playing), as well as negatively valenced interactions such as agonism (example: staring, resource guarding), with negative interactions garnering more research and attention.

Inter-cat conflict is one of the most frequent owner-reported problems in multi-cat households (4, 10–12) and one of the main reasons cats are presented to behavior clinics (13–16). A UK-based survey suggests 62% (N = 616 cats) of multi-cat owners see signs of inter-cat conflict (hissing, spitting, or blocking) in their household (17). Similarly, 50% of cat-owner participants from a Canadian survey assessing (N = 1,146) fostered kittens adopted into multi-cat households, reported seeing behavioral signs of aggression between cats in their home (18). When inter-cat conflict is not mitigated, the welfare of both cats may be compromised due to prolonged stress (5, 19) leading to health and behavioral problems such as house soiling (20–22), as well as increased risk of owner-directed bites and scratches (23). In addition, behavior problems can diminish the human-animal bond (9, 24) and may lead to relinquishment (25–27).

Many factors impact the cat-cat relationship, including resource provision, outdoor access, and cat characteristics (28–30). Research shows that inadequate resource availability and distribution may increase inter-cat conflict in the home (30). Thus, it is recommended to provide multiple, well-distributed resources (example: food bowls, litter boxes, and environmental enrichment items) throughout the home to reduce individuals from monopolizing resources (30). To further support this recommendation, observational studies of group-housed indoor cats have found that cats time-share resources, meaning they choose to access valuable resources at different times (29, 31). A large survey of US and Canadian cat owners by Tan and colleagues shows that owners providing cats with uncontrolled outdoor access are less likely to provide important in-home resources such as those necessary for cats to perform naturally motivated behaviors such as perching and playing (32). In addition, allowing cats outdoors has been associated with continued fighting (12) and an increased likelihood of aggression between household cats (18). However, in contrast, other research has not found associations between outdoor access and agonistic interactions in multi-cat households (33, 34). Although no consensus exists, the study by Tan and colleagues suggests that uncontrolled outdoor access is more likely to be provided by owners of cats showing aggression toward other cats or people in the home (32). Thus, there is some evidence to support a relationship exists, however the temporal directionality of this relationship is unknown.

Cat characteristics such as sex, age, weaning age, neuter status, as well as familiarity and relatedness have been suggested to impact the cat relationship in multi-cat households. Neuter status is known to reduce conflict behavior in both neutered male and female cats compared to intact cats of the same sex (35). Barry and Crowell-Davis (36) found that indoor, neutered males choose to spend more time in close proximity than other sex combinations. As well, age may impact cat-cat behavior; for example, Ramos and colleagues (37) found adult cats have significantly higher fecal glucocorticoid levels than young cats in multi-cat households (3-4 cats). As well, a UK-based survey of cat owners (N = 833) suggests cats 11 years and older tend to be less sociable toward other animals (including dogs, cats, etc.) in the home as they get older (38). In contrast, one survey (N = 2492) reports that young and adult cats display more active conflict-related behaviors (such as chasing and stalking) than households with mature or senior cats (34). However, chasing and stalking are also indicative of play behavior in cats, and thus may not be negative for all cats involved. As cats age, they are at an increased risk of developing diseases which may cause pain or discomfort (39), and thus their tolerance of interactions with other cats may change. The age at which cats are weaned from their mother also impacts the inter-cat relationship, with early weaned cats (< 8 weeks of age) more likely to show aggression toward other cats (40). Furthermore, kittens that are reared without their littermates during the early socialization period (approximately between 2 and 7 weeks old) display more agonistic interactions with conspecifics (41). Generally, social deprivation from the mother and littermates during the socialization period may negatively influence cats' subsequent social interactions with conspecifics (4). In addition, related and unrelated but familiar kittens and cats show more affiliative interactions compared to unfamiliar and unrelated cats (42–44).

Although there is an increasing body of research examining factors impacting the inter-cat relationship in multi-cat households, one large problem is the lack of research focusing on one group size. This gap in knowledge reduces generalizability of the research findings to specific compositions of cats in multi-cat households (10). This is important, given that it is generally recognized that varying the number of animals in a group impacts social structure and social complexity (45). Studies assessing multi-cat households often include large cohorts of cats (17, 29, 34, 46) which may have an increased risk of conflict compared to smaller cohorts of cats (34). To the authors' knowledge, one study has examined indoor-only cat dyads in the home, with results suggesting that neutered male dyads spend more time in close proximity than female dyads or male-female dyads (36). However, more research is needed to assess these findings in a larger population and achieve a better understanding of the factors impacting cat dyad behavior in the home. Overall, research focused on improved understanding of the social relationship between cats of a defined number in a household is needed. Since many households in Canada and the US contain two cats (1, 2), research focused on two-cat households may be more impactful. Given the limited research focusing on two-cat households, we aimed to explore and describe associations between cat owners' perceptions of their cats' relationship, management factors, and cat-specific characteristics. A cross-sectional questionnaire was used to survey US and Canadian owners of two adult cats (≥1 year old) and included questions on owner demographics, owner's self-perceived knowledge of cat behavior, owner's subjective perception of their cats overall relationship, household and management factors, and cat dyad characteristics. We predicted that cat owners rating their cats as having a negative relationship would be associated with provision of a single resource area (example: one litter box, one feeding area) compared to households with multiple resource areas (example: two or more litter boxes and feeding areas), and reduced odds of a negative relationship in households with related cats compared to unrelated cats.

This study was approved by the University of California Davis Institutional Review Board (IRB #1786341-1) to recruit human participants for research. Participation was anonymous and respondents provided consent before being able to participate in the questionnaire.

Our survey results suggest many factors impact cat owner ratings of their cats' relationship. Interestingly, spayed female dyads were more likely to have an owner perceived negative relationship compared to neutered male dyads, or mixed sex dyads. This is in line with other literature on the influence of sex on the cat-cat relationship in the home. One study by Barry and Crowell-Davis (36) shows that indoor neutered male dyads choose to spend more time in closer proximity than females or mixed sex combinations, suggesting male dyads may get along better than other sex combinations. However, there is conflicting results from behavioral clinic data, with one study from Australia suggesting that female cats display more inter-cat aggression than male cats (13), while another study in the US by Lindell and colleagues (4) suggests that male cats are more likely to act as aggressors toward other male or female cats. Albeit, behavioral clinic data stems from a limited sample and likely represents more severe cases of inter-cat conflict that may not be generalizable to the average two-cat household. Given that our results, as well as much of the existing scientific evidence, suggests spayed female dyads show more negative interactions, this could be something for cat adopters to consider when they already have one female cat at home. It should be noted that other factors such as age of weaning, socialization experiences, and how cats are introduced, likely impact the complex relationship between cat dyads in the home. As well, animal shelters typically place cats into a home with another cat if the shelter cat has a history of living with other cats, and if they display more social behaviors in the shelter or foster home such as playing, compared to fearful or avoidant behaviors (51).

Age also impacted owner ratings of their cats' relationship, with younger (1–3 years old) cat dyads less likely to be rated negatively, compared to all other age group combinations. In addition, dyads consisting of a young and adult cat (4–6 years old) were less likely to have a negative relationship compared to mature-cat combinations. As well, pairs of mature cats (7+ years old) were more likely to have a negatively perceived relationship compared to pairs of adult cats. This suggests that cat owners perceive younger cats as getting along better with other young or adult cats, compared to mature cat combinations. Thus, when pairing cats, such as during the adoption process, it may be beneficial to pair younger cats together and avoid mature-cat combinations. Other research examining age-related impacts on the multi-cat relationship suggests that younger cats display chasing and fleeing behaviors, which the authors categorized as conflict-related, more frequently than older cats (34). However, chasing and fleeing behaviors are also seen during play (52), and thus may indicate increased play and not conflict, in younger vs. older cats. Another study examining the impact of conflict behaviors in households following the introduction of a new cat did not find an age effect on the cat-cat relationship (12). Overall, more research is needed to establish stronger links between cat dyad age combinations and the cat-cat relationship. For example, a prospective cohort study with direct behavioral observations of cat dyads of various age combinations in two-cat households would be beneficial.

Cats' relatedness was a factor that impacted participant ratings of their cats' relationship. Cats that were not related were more likely to be rated as having a negative relationship compared to cats that were siblings or placed into the “other” category. Participants that selected “other” had cats that were parent and child or that were bonded before adoption. This finding is not unexpected given the natural history of cats. For example, in free ranging cat colonies, individuals choose to socialize with preferred conspecifics and related females typically interact and may even form small colonies with other females from their lineage and their offspring (3, 11). Similarly in a private colony of neutered cats, Curtis and colleagues (42) found related cats were significantly more likely to be within 1 m of each other and display allogrooming, an affiliative inter-cat behavior. In shelter environments, littermates show more physical contact and allogrooming behavior than unrelated cats from the same household (43). Thus, related cats may be more likely to get along due to the strong bonds formed early in life (11, 43). It is important to educate owners on the importance of relatedness and early social bonds when they are adopting. If there is an opportunity to adopt related cats, owners should be educated on the positive influence it may have on the cats' relationship and the possible consequences of introducing an unrelated cat later on. Future studies should investigate the motivations of cat owners seeking additional cats into their households. Further, research is needed to provide evidence-based recommendations for introducing an unknown cat into a household with existing cat(s).

Outdoor access (either supervised or unsupervised), was another factor that impacted participant ratings of the cat-cat relationship in 2-cat households; those with outdoor access were more likely to have a reported negative relationship than pairs kept indoors-only. Other cross-sectional cat-owner survey research examining multi-cat households have found similar results, with outdoor access associated with increased aggression toward other household cats (18), outdoor access associated with increased fighting during the period of time when cats are being introduced (12), and a negative correlation between outdoor access and inter-cat affiliative behaviors (34). It is possible that cats with outdoor access may bring new and unfamiliar odors into the home which may initiate cat-cat conflict, however no research supports this. Another possibility is that cat owners may be more likely to let their cats outdoors when they do not get along in the home. Given our survey had a cross-sectional design, we were not able to assess temporality of factors associated with a negative cat-cat relationship, which induces uncertainty about causation (53). Future research should use a study design that allows for temporal investigation such as a longitudinal study where cats in the home can be followed over time. This type of study design would also help reduce recall bias which may exist in cross-sectional research.

A high proportion of participants reported that at least one of their cats have had, or currently has, a health problem (59.9%). Although this is concerning, it is comparable to other studies. For example, O'Neill et al. (54) found 68% (2,449/3,584) of cats seen by veterinarians in England have at least one health disorder. Moreover, 48% of US and Canadian cat owners (580/1,208) reported a health disorder in their own cat(s) (55). The most commonly reported health problems in the current study were obesity (25.2%) and dental disease (23.4%, Table 5) and other studies have reported similar prevalence of these health issues. Roberts and colleagues (17) found 19.9% (150/755) of UK owners reported their cats as overweight or obese. In the US, the prevalence of obesity in cats seen by veterinarians during 1995 (N = 8,159) was 35% (56), while more recently, Dodd and colleagues (55) found 33% (405/1,233) of US and Canadian cat owners rated their cat's body condition as overweight. Veterinary practices in the US (N = 15,226) report cat patients are most commonly diagnosed for dental calculus (24.2%) and gingivitis [13.1%; (56)]. Similarly, O'Neill et al. (54) report that periodontal disease was the most prevalent disease (13.9%, N = 499/3584) in cats of UK veterinary clinics. While the current study did not find any health issues to significantly impact the cat-cat relationship as perceived by the owner, future studies should assess if symptoms of health issues (i.e. pain, fatigue) would affect the latter.

A large portion of cat owners (78.5%) also reported that at least one of their cats has a current or previous behavior problem, with fears/phobias (45.7%), unwanted behaviors (45.2%) and destructive behaviors (40.6%) most commonly reported. However, we did not require these to be diagnosed by a veterinarian or animal behaviorist. One survey of US cat owners (N = 547), state that 47% of participants answered “yes” when asked if their cat(s) ever misbehave (9), and they found similar prevalence for anxiety/fear (59.4%) and destructive behaviors (49.7%) as the current study. Another survey of US and Canadian cat owners (N = 2465) found that 58% reported inappropriate scratching (57), which is similar in prevalence to destructive and unwanted behaviors in the current study. It is possible some participants selected a behavior problem because they have seen it in their cat(s) (i.e. excessive night time activity, unwanted behaviors, fear/phobias, etc.), but it may not be displayed to the intensity and frequency where it would constitute as a behavior problem (58).

Households with at least one cat that has shown animal or human-directed aggression were associated with owners rating their cats' relationship negatively. Aggression toward people and other animals can vary from subtle agonistic displays to more obvious displays that may lead to serious injuries. Inter-cat aggression is a major stressor for cats, and may lead to further behavioral problems such as house soiling (5, 6, 57), which may increase the risk of relinquishment (25–27). Aggression toward people and other animals may be affected by many factors such as socialization experiences, management of the home environment, and interactions with people and other animals in the home (58). It is important for the type of aggression to be identified (i.e., fear-related, territorial, play-related, petting-induced, redirected, social stress, pain-induced), as well as sources or triggers that may lead to an aggressive event (58). Research suggests cats reported to show human and/or animal-related aggression may involve redirected aggression, which is commonly enticed by inter-cat conflict and loud noises (59). However, the current study did not assess causes of aggression-related behavior problems given this is not possible with a cross-sectional survey design. Cat owners may benefit from addressing aggression through early management of the problem, to minimize the risk of stress, injuries, and further behavioral or health problems.

We also found that provision of resources in the home is associated with owner perceptions of a negative cat-cat relationship. Owners perceiving their cats' relationship as negative was associated with households with multiple litter boxes and feeding areas. These findings were not in line with our predictions; however, our survey design could not assess the temporality of these associations. Thus, it is possible that cat owners choose to provide multiple resources as a solution when conflict becomes present in the home rather than as a preventative measure. The American Association of Feline Practitioners recommends that multi-cat households should have multiple, easily accessible resources to meet cat behavior needs (60). Multi-cat households have been identified as a risk factor for behavior problems in the home (20, 22) such as inappropriate elimination (61–63), is a common reason for relinquishment (26). While providing multiple litter box areas may not solely prevent house soiling, it is an important consideration. In addition, providing multiple separate food areas in multi-cat households is recommended to help reduce agonistic interactions such as resource guarding, which may be present when one cat is more dominant and assertive over another more timid cat (60). Resource guarding around limited food areas may lead to rapid ingestion of food or inadequate nutritional intake, and may increase the risk of health issues overtime (64). Although the recommendation of multiple food areas is not based on scientific evidence, cats are naturally solitary hunters. Thus, feedings areas that are physically separate may reduce the potential of threat and may better mimic “solitary” eating (65, 66). Overall, there is little experimental evidence about the impact of resource distribution in the home on cat-cat interactions, and more research is needed.

Our survey results show that the cat-cat relationship in two-cat households is complex and impacted by many factors such as, cat sex, age, relatedness, outdoor access, resource provision in the home, and aggression directed toward other people and animals in the home. The complex interplay and directionality of these factors requires further investigation for a full understanding of how to improve the cat-cat relationship in the two-cat households. More research is also needed to provide cat owners with evidence-based recommendations for providing adequate resources in two-cat households. Cat owners may also benefit from information on factors to consider before acquiring a second cat to foster a stronger connection between their cats, as well as scientifically supported guidelines for introducing their cats.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by UC Davis Institutional Research Ethics Board. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

SK wrote the manuscript. CM edited the manuscript. All authors established the study question, devised the study methods, and questionnaire. All authors contributed to the article and approved the submitted version.