Dogs and humans have a unique relationship as a result of our shared history. Humans are an unavoidable component of the laboratory environment, responsible for cleaning and feeding, regulated procedures as well as designated periods of interaction. Humans and domestic dogs have been closely associated for 12-15,000 years , and during this time, dogs and humans have developed a close and often mutually-beneficial relationship, for example cooperation in hunting or herding of other animals . This has resulted in significant adaptation, with selective breeding for traits promoting close cooperation with humans.
Dogs also demonstrate patterns of behaviour similar to those of attachment in human infants in tests of Ainsworth's 'Strange Situation' paradigm [3,4], using a human caregiver as a secure base to explore a novel environment and showing distress at separation. This close relationship is potentially a critical welfare problem in the laboratory environment where dogs are subject to adverse interactions with staff. While laboratory-housed dogs do not have the same relationship with a human carer as a pet dog might, interactions with humans still have the potential to be a tool for improving welfare.
However it also provides an opportunity to increase welfare in a manner not possible with species which view humans as a threat; an improved relationship between dog and handler leads to the use of the handler as a 'secure base' during adverse events.
The table below summarises the findings of studies investigating the effects of human interaction programmes on behaviour and physiological measures in pet, shelter, working and laboratory-housed dogs. Studies differed in the frequency, duration and content of human interaction periods, and in the results found. Studies investigating the effect of a single human interaction period have found an immediate positive change in physiological and behavioural signs of stress, however no lasting effects were found. This conclusion was also drawn by Taylor and Mills  in a review of human interaction programmes. In contrast, Normando et al. (2009) found that the changes in behaviour exhibited by dogs which underwent the human interaction programme were maintained for two weeks following termination of the programme.
Importantly, Normando et al.  found that interruption of a regular walking programme resulted in an increase in cortisol levels in laboratory-housed beagles, so once started, it is important that a programme is maintained. However, all of these studies found a positive impact from brief interventions, suggesting that such interventions should be beneficial in the laboratory environment.
The quality and quantity of an intervention clearly has an influence on its efficacy. Hennessy, Voith, Young et al.  found only two behavioural differences following intensive positive-reinforcement training when compared to a control group which didn't receive training (fewer non-directed licks and fewer escape attempts), while Normando et al.  found differences in behaviour (increased tail wagging and increased inactivity) and proximity (less time spent out of sight) following fewer interaction periods, but which involved positive social contact. Differences in the behaviour of a human handler (when measured in dimensions of enthusiasm and discipline) can have an effect on whether a play session increases or decreases a physiological measure of stress, and training alone may not be sufficiently positive to improve welfare .
Wells  suggested that play especially may be a particularly helpful tool in interaction with dogs in shelters, as it establishes "appropriate dog-human relationships" and prepares a dog for rehoming. In the laboratory environment, a positive social bond between dogs and handlers may facilitate learning in training for procedures, in completing husbandry protocols without introducing an element of stress through fear responses, and in providing a tool to increase resilience when welfare may be compromised by aversive events such as conspecific isolation. Dogs housed in animal units may be provided with little opportunity to interact with humans (from between 0.3-2.5% of time observed, Hubrecht, Serpell & Poole ). Human contact clearly has the ability to mitigate the effects of some adverse events upon welfare, particularly through the use of a human carer as a 'secure base' and to provide positive experiences in the laboratory environment. The ability of human contact to cause positive changes in haemodynamics and endocrine responses to stress suggests that the structured implementation of a Refinement protocol may improve welfare and data quality in a feasible manner. However care must be taken when developing such a Refinement protocol so that it is of a suitable frequency and duration and that it is feasible, since the loss of predictable positive human interaction may be more detrimental to dog welfare than never having it at all.
The result of our history of cooperation in a shared environment has led to the dog developing abilities to communicate with and understand humans which in some ways exceed those of nonhuman primates. Dogs demonstrate an understanding of human eye gaze similar to that of human infants and exceeding that of the dog's ancestor, the wolf (Canis lupus) or the chimpanzee (Pan troglodytes, [11, 12]).
3. Gácsi, M., Győri, B., Miklósi, Á., Virányi, Z., Kubinyi, E., Topál, J., & Csányi, V. (2005). Species‐specific differences and similarities in the behavior of hand‐raised dog and wolf pups in social situations with humans. Developmental Psychobiology, 47(2), 111-122.↩
4. Palmer, R., & Custance, D. (2008). A counterbalanced version of Ainsworth's Strange Situation Procedure reveals secure-base effects in dog–human relationships. Applied animal behaviour science, 109(2), 306-319↩
6. Normando, S., Corain, L., Salvadoretti, M., Meers, L., & Valsecchi, P. (2009). Effects of an enhanced human interaction program on shelter dogs' behaviour analysed using a novel nonparametric test. Applied Animal Behaviour Science, 116(2), 211-219.↩
7. Hennessy, M. B., Voith, V. L., Young, T. L., Hawke, J. L., Centrone, J., McDowell, A. L., ... & Davenport, G. M. (2002). Exploring human interaction and diet effects on the behavior of dogs in a public animal shelter. Journal of Applied Animal Welfare Science, 5(4), 253-273.↩
8. Horváth, Z., Dóka, A., & Miklósi, Á. (2008). Affiliative and disciplinary behavior of human handlers during play with their dog affects cortisol concentrations in opposite directions. Hormones and behavior, 54(1), 107-114.↩