Dogs are kept in a wide variety of environments, including laboratories and kennel establishments housing large numbers of working dogs for much of their lives; short-term housing in rehoming centres; kennels within owners’ gardens; and within owners’ homes with access to roam freely between multiple rooms. Each environment presents its own challenges. Whilst it is often assumed that home-dwelling dogs experience near optimal conditions, and indeed Shore et al. found that people keeping dogs indoors, compared to in a yard, paid more attention to their social needs, it has also been asserted that modern companion animals experience ‘‘a relatively dull life’’, and it has been shown that owners lack the knowledge necessary to safeguard their dogs’ welfare (PDSA 2011).
Recent legislation and social pressure in many cultures has led to dogs being kept predominantly indoors, an environment very different from that in which they were originally domesticated. Below we describe a number of current welfare challenges faced by pet dogs and discuss the extent to which each of the three approaches to welfare discussed above would describe them to be a problem. Since this volume is primarily concerned with cognition and behaviour, and health has tended to dominate much of the literature on dog welfare, we will primarily discuss issues that arise from lack of understanding of, and provision for, dogs’ behavioural needs.
However, that is not to say that physical health is not a major issue worldwide. Many street dogs are chronically unhealthy, and Non-Government Organisations invest heavily in mass vaccination and neutering programmes aimed at safeguarding the health of current and future generations (although little attention is paid to affective and behavioural aspects of their welfare). In Western nations, many owners appear to protect the health of their dogs, most dogs are vaccinated against transmissible diseases (PDSA 2011) and many receive prophylactic treatment against parasites. Mass rearing establishments (puppy ‘farms’ or ‘mills’) are the obvious exception, where dogs are often bred and reared in large numbers with apparent disregard for disease prevention. However, even in the owned population severe health issues persist.
Most dogs in developed nations are apparently healthy and well-fed, but obesity is now a major issue. It has even been suggested as the most important welfare problem of dogs in the post-industrial developed world due in part to the sheer numbers of animals involved: for example, the (Pet Food Manufacturers’ Association 2009) state that up to one-third of companion animals in the UK are obese.
Obesity was originally highlighted as a welfare issue due to its effects upon health and longevity, but it is now also known to exacerbate a whole variety of clinical conditions some of which have also shown to be painful and hence to cause psychological distress. In addition, dietary restrictions placed upon animals under treatment for obesity are likely to result in negative feelings of hunger and frustration as the animals will be unaware of their treatment’s long term aim or benefit. It is possible that in the process of breeding dogs which are amenable and easily trained using food rewards, we have inadvertently selected individuals which are excessively food-motivated and which feel hungry even when they are physiologically well-nourished, a stereotype anecdotally apportioned to many Labrador Retrievers (among others). If so, it is possible these dogs may suffer psychologically, even when well fed, let alone when they are placed on ‘health-enhancing’ feed management plans. This situation provides an example of when psychological and health approaches to welfare may be contradictory.
Obesity can also reduce a dog’s ability to behave normally, and hence when extreme, would be an issue affecting ‘naturalness’. Obesity is a rare example of a welfare issue which all three approaches indicate as a welfare concern; however, the cut-off point at which it would be highlighted as an issue is likely to differ, dependent upon which criteria (health, feelings, or natural behaviour) are paramount.
Lack of Human Company
Dogs have been selectively bred to value and even crave human attention, and the beneficial effects of human contact on dogs have been well demonstrated experimentally. Obedience and other favourable behaviour improved in shelter dogs and in working dogs when they receive increased contact from prisoners or handlers. Positive interactions (stroking, scratching, talking, playing) result in increased beta-endorphin, oxytocin, prolactin, beta-phenylethylamine, and dopamine and reduced cortisol levels in new arrivals to rehoming kennels, and gentle stroking inhibits the increase in cortisol during venipuncture. Longer-term, enhanced human contact not only reduces physiological stress in dogs per se but has also been shown to ameliorate stress responses when dogs are subsequently presented with novel environments or unfamiliar people. Contact with people generally has greater success in inhibiting stress in novel environments than does contact with familiar dogs; hence, human contact has been suggested to be more important for the well-being of dogs than conspecific contact, although this undoubtedly varies with an individual dog’s temperament and past experience.
The detriments due to lack of contact with people are well proven for kennelled dogs, but these likely also apply to the majority of owned dogs. One of the two most common reported reasons for owners surrendering a dog to a rehoming centre in the UK is the perception that it requires more attention than the owner is able to provide. Increasingly, dogs are kept within households where all members of the family are absent for at least part of the day. Reliable data is lacking, but undoubtedly a large proportion of dogs are routinely left for periods in excess of 4 h per day (e.g. 73 % in Sweden). So paradoxically whilst we ‘‘value the social feedback they give us’’ and derive great benefits from it we also ‘‘leave them alone for lengthy periods’’, unsurprisingly, many dogs behave in ways that indicate that they find such separation stressful, and this presents one of the biggest challenges of the modern day companion-animal niche.
Separation-related behaviour (usually expressed as destruction, vocalisation and/or toileting) is a common and often overlooked issue. Dog-walkers in southern England reported 24 % of dogs as currently exhibiting, or previously exhibiting, separation-related behaviour, and up to 50 % of Border Collies and Labradors filmed in a longitudinal study had shown this behaviour by 18 months of age. This problem is not restricted to the UK: 56 % of dogs living in urban Rio de Janeiro were reported to show clinical signs of separation anxiety.
It may seem contradictory that loving owners accept such signs of apparent distress, but a large proportion seem unaware of their dog’s behaviour in their absence, and amongst those that are aware, anthropomorphic attitudes lead many to believe that destructive or eliminative behaviours are motivated by higher-level emotions such as ‘revenge’. Dogs often show no obvious health decrements as a result of their distress, and perhaps for this reason, the severity of separation as a welfare issue has been widely overlooked. Surprisingly, there appear to have been no published studies examining physiological indicators of stress during separation from the owner. However, recently developed cognitive measures (discussed below) have shown that animals which exhibit separation related behaviour are more likely to be experiencing a negative emotional state, not just when they are actually performing the behaviour, but also at other times.
Protocols to treat separation-related behaviour can be very effective, producing improvements in 81 % of dogs and methods to prevent the behaviour developing also show promise. Both can potentially improve a dog’s quality of life, as well as reducing the chances of it being surrendered. More general adoption of such techniques, combined with a change in owner expectations such that they no longer routinely leave susceptible dogs alone for long periods of time, has the potential to improve the welfare of large numbers of dogs. However, research is also needed to fully understand the respective roles of past experience, expectation, and owner behaviour in determining why some individuals develop this problem whilst others appear unaffected.
Training methods can be classified according to the emphasis they place on the four types of reinforcement recognised under learning theory: positive reinforcement (e.g. food reward), positive punishment (ranging from harsh vocalisation to the deliberate infliction of acute pain), negative punishment (usually the withholding of an anticipated reward), and negative reinforcement (the cessation of pain or other aversive stimulus). In practice, no training method relies exclusively on just one of these: even ‘‘reward-based training’’ incorporates negative punishment (e.g. withholding human attention) to reduce the performance of undesired behaviour. Positive punishment, especially the infliction of pain through hitting, twisting of the ears, choking the windpipe and electric shocks, is intrinsically aversive and hence, according to the ‘affective approach’, welfare-reducing. It is often justified by its supporters on the grounds that it is more than compensated for by greater benefit to the individual dog’s future welfare (in terms of health), because the dog becomes more obedient and hence less likely cause damage to itself, other dogs or livestock, and less likely to be abandoned. There is, however, an increasing body of evidence pointing to the contrary, that dogs trained using methods that incorporate positive punishment tend to be less obedient and/or more aggressive than those trained using ‘‘reward’’, and moreover many such dogs also show signs which may be indicative of fear and anxiety. Studies that have focused specifically on dogs trained with remotely-activated electric shock collars have found associated evidence of chronic fear to an extent not shown even by dogs trained with other types of positive punishment yet such devices remain available to dog owners in many countries, including the USA and parts of the UK. Reward-based training has been criticised on ostensible welfare grounds, as producing ‘‘soft’’ dogs that are disobedient, likely to stray or be injured, and hence at risk of abandonment (For example, UK trainer Charlie Clarricoates, who stated ‘‘We are seeing dogs now who are spoiled rotten, and never have any discipline, mainly because owners are force-fed incorrect impractical information…This moralistic attitude that you can only train dogs by loving them and being kind is ridiculous. There are some dogs you can’t do this with because it doesn’t work, even if you have a year with them’’) although there does not appear to be any systematically collected data to support this conclusion, and increasing evidence to the contrary that it is associated with increased obedience and learning ability.
It is traditional to make a distinction between problems that result from seemingly abnormal behaviours — whose relevance to welfare is best understood through the ‘affect’ approach — and problems that are essentially ‘natural behaviour’ and arise due to conflict with owners’ expectations. The latter (e.g. barking, chasing, and possibly even coprophagy) cause a problem when performed within the domestic environment and hence potentially threaten the animal’s future welfare through a secondary route by jeopardising the dog-owner relationship. However, as our knowledge base increases, this distinction has become increasingly arbitrary, as almost all ‘problems’, possibly with the exception of neurological disorders, can be explained by ‘natural’ reactions to unnatural or suboptimal environments. None more so than canine aggression, a highly contentious topic that is too complex to be dealt with here.
Undesired or problem behaviour is generally important to dog welfare as it may not only be associated with poor welfare in its own right, it is also often the primary reason given when dogs are surrendered to animal centres, or euthanased. There are few longitudinal studies examining dogs’ welfare before and after relinquishment, but since abandonment increases the likelihood of euthanasia, and there are also numerous factors associated with life in a rehoming kennel which can lead to psychological distress, it is likely that for many dogs, welfare would be better were the dog to remain in its original home. Therefore one of the most effective ways to improve companion dog welfare is to prevent problematic behaviours from developing. This could be aided both by increased owner education and matching of appropriate dogs to owners, and also by mitigating those problems that do develop through professionalized clinical behaviour advice.