Article Text

Download PDFPDF

Efficacy of dog-appeasing pheromone in reducing behaviours associated with fear of unfamiliar people and new surroundings in newly adopted puppies
  1. E. Gaultier, DVM, MSc, DipECVBM-CA1,
  2. L. Bonnafous, DVM1,
  3. D. Vienet-Lagué, DVM2,
  4. C. Falewee, DVM1,
  5. L. Bougrat, MSc1,
  6. C. Lafont-Lecuelle, MSc1 and
  7. P. Pageat, PhD, DVM, DipECVBM-CA1
  1. 1 Phérosynthèse Research Institute, Le Rieu Neuf, 84090 St Saturnin-les-Apt, France
  2. 2 Clinique Vétérinaire du Chêne à Leu, 544 Avenue de Buchholz, 76380 Canteleu, France
  1. Clinique Vétérinaire des Ocres, 174 Route de Gordes, Coustellet, 84220 Cabrières d’Avignon, France, vet-comportementaliste{at}


This study was designed to determine the potential value of dog-appeasing pheromone (DAP) in reducing behaviours associated with fear of unfamiliar people and new surroundings in puppies newly adopted from a pet shop. The study was triple-blinded, randomised and placebo-controlled. It used 66 puppies (32 fitted with a DAP collar and 34 control) and the adoptive owners were contacted by phone three days and 15 days after they had adopted the puppy to question them about its reactions to specific situations eliciting fear. Fifteen days after the treatments significantly fewer of the puppies with the DAP collars showed signs of fear when facing unfamiliar people at home and/ or during outings. This difference was irrespective of breed size.

View Full Text

Statistics from

SINCE experimental work carried out in the 1970s, novelty has become one of the main fear-eliciting stimuli used in the field of experimental ethology (Gray 1987, Morméde and Dantzer 1988, Boissy 1995). The exposure of animals to novelty was used in a long series of open-field test experiments that enabled behaviourists to define the processes of habituation and sensitisation (Groves and Thompson 1970). Although habituation and sensitisation are concepts that are applied practically in clinical situations, little has been done to assess dogs’ reactions to novelty under normal conditions. The effects of novelty have been explored, but mainly in experimental situations, for example, in open-field studies (Goddard and Beilharz 1984, Beerda and others 1997, 1998, 1999, King and others 2003) or semi-experimental situations (Netto and Planta 1997, Kroll and others 2004), in order to explore its association with fear aggression. Appleby and others (2002) suggested that dogs that have spent the sensitive period of their development in an environment with little complexity may perceive more novelty in everyday experience, and that it may contribute to the development of avoidance and/or aggressive behaviour in adulthood. However, little has been done to explore the reactions of dogs to novelty in normal clinical or domestic settings, for example, when puppies are newly adopted.

Adoption involves a series of traumatic events, as the puppy’s maternal bond is abruptly severed and it is transferred from a familiar world into an entirely new one, with new places and surroundings, unfamiliar people and new rules (Elliot and Scott 1961, Pettijohn and others 1977, Taylor and Mills 2007). Moreover, inadequate efforts may be made to ease the transition or to minimise any adverse effects (Plujimakers and others 2006), and this is particularly true when the puppy is obtained from a pet shop. Such puppies experience conditions that may affect their behavioural development adversely; they have a shorter period of maternal contact, followed by two periods of transport and ‘re-homing’. Previous studies have indicated that there is a relationship between a puppy’s origin in a pet shop and its tendency to show fear-based behavioural problems, particularly social fear, in adulthood (Serpell and Jagoe 1995).

From the owner’s perspective, adopting a puppy is successful only if its arrival enriches family life, but difficulties in coping with life in the new family setting may severely affect the development of the puppy’s human-animal bond. These difficulties may also induce signs of fear that may interfere with the puppy’s ability to urinate or defecate when it goes outside, and lead to house-soiling. Unexpected fear reactions towards new people and dogs will impair the puppy’s social skills and may often compel owners to further restrict its social life. It therefore appears to be important to control any signs of fear displayed by puppies after their adoption. A more subtle and more common problem may be that owners live with a problem that may limit their bond with the puppy and cause problems for society, for example, aggression caused by fear. Any method for preventing fear reactions in dogs may therefore be considered valuable for improving their welfare.

Previous studies have shown the value of using a synthetic analogue of the dog-appeasing pheromone (DAP) (CEVA Santé Animale) in a variety of fear-inducing circumstances, such as transport (Gaultier and Pageat 2003, Gandia Estellés and Mills 2006), separation from people (Gaultier and others 2005, 2008), firework displays (Sheppard and Mills 2003), attendance at veterinary clinics (Mills and others 2006) and living in a shelter for dogs (Tod and others 2005). In two placebo-controlled studies, in which DAP was used with puppies during an eight-week training programme, Denenberg and others (2005) and Denenberg and Landsberg (2008) observed that puppies without DAP were more distracted and expressed more fear-related signs, such as aggression, avoidance, submissive urination, and fearful postures when facing new circumstances. DAP therefore produced positive effects during the socialisation phase in a typical population of puppies.

The purpose of this study was to assess the effect of DAP on the behaviour of puppies under the more extreme conditions of being adopted from a pet shop. Its aims were to investigate the various expressions of stress shown by puppies newly adopted from a pet shop, and to assess the effects of DAP on them. Puppies adopted directly from a pet shop are particularly vulnerable, and they were studied during the period immediately after they were adopted. Descriptive analyses of the data (Gaultier and others 2007) showed that the puppies’ stress after adoption could be classified into three main categories according to the triggering situation: nuisance activities when they were isolated, stressrelated behaviours in novel situations, and signs of non-specific general anxiety (excessive arousal, defecation). The impact of DAP in reducing the puppies’ stress associated with isolation has been described by Gaultier and others (2008). This study focused on the effect of DAP on the puppies’ behaviour in coping with unfamiliar social and physical circumstances.

Materials and methods

The materials, methods and protocols were the same as those described by Gaultier and others (2008); a summary is presented here.

Treatment and study design

The study was a triple-blinded, placebo-controlled, single-centre, parallel- group trial. A 35 cm collar containing 2·5 per cent DAP was fitted to the neck of 32 puppies, and 34 puppies were fitted with an identical placebo collar. The owners were asked to remove the collar whenever they washed their puppy because detergent products may strip the pheromone from the collar.

Recruitment of puppies and procedure

The puppies were healthy puppies bought from a pet shop in Rouen, whose owners agreed to participate in the study. The puppies received a DAP or control collar at random when they arrived in the pet shop, so that the treatment could be tested during the process of adoption.

The puppies were evaluated during the period immediately after their adoption, a procedure that involved their transport from the pet shop to the new owner’s home, and thus a change of location and social group, and novel circumstances in all aspects of their life.

The collars were renewed on the day they were adopted, to ensure that the release of DAP was consistent in the treated group during the evaluation period. Each owner provided written consent and allowed the puppy to continue to wear the collar for two weeks. They also agreed to be contacted by the investigator twice by telephone to answer a short questionnaire about their puppy’s behaviour. These interviews were carried out between two and four days and between 13 and 17 days after their adoption.

Data collection

Epidemiological criteria Information about the characteristics of each puppy at the time of its adoption, the duration of its stay in the pet shop, its level of activity in the new environment, and the new owner’s previous experience with a dog, provided information about the comparability of the two treatment groups. Some of these criteria might be related to the dependent variables as confounding factors.

Treatment tolerability The tolerability of the collars was evaluated in two ways: first, by the pet shop staff, who regularly examined each puppy’s neck, and secondly by the new owners, who reported their experience through the questionnaire.

Criteria assessing the behavioural reactions of puppies A wide range of potentially unfamiliar circumstances and experiences were investigated through the questionnaire, for example, the puppy’s reactions to the exploration of a new object, its interactions with the owners and with unfamiliar people, and its reactions to outdoor and indoor situations. The results of the two assessments of these reactions are shown in Table 1.

Structure of the questionnaire As fear may be expressed in a number of ways, the questions were semi-open and the owners could choose between several pre-defined responses. These responses, which indicate the ‘presence’ or the ‘absence’ of fear, are shown in Tables 1 and 2, with those relating to the presence of fear shown in italics.

As this type of questionnaire does not allow the investigator any direct control over the circumstances in which the owner has observed the puppy’s reaction, some specifications were made in order to improve standardisation. For example, to ensure that the reactions of all the puppies were assessed in the same kind of fear-eliciting situations, only typical everyday events were chosen for the owner’s observation. These included times when the owners brought bags back from the supermarket (novel object test), when friends visited the home (exposure to an unfamiliar person) and when the puppy went on an outing (reaction to a new environment). Apart from the last, these events could be considered to be quite standardised for all the puppies; however, during outings, the level of complexity and stimulation would have varied greatly from one place to another. In order to assess the variability of this test situation between puppies, a question was therefore added to assess the type of environment and the level of the puppy’s stimulation and activity.

Data processing and statistical analysis

To assess the puppy’s behaviour in its new environment, the efficacy of the DAP collar was evaluated in terms of a composite parameter. The main criterion combined the occurrence of signs of fear in the presence of unfamiliar people at home and the presence of signs of fear when walking outside on day 15. These two signs provide a good indication of the reactivity of the puppies because they were not linked to a specific fear-eliciting situation, and they were independent because they were assessed in entirely different situations. This combination of two independent signs generated four combinations for the main criterion: fear + fear, fear + no fear, no fear + fear and no fear + fear. In order to minimise the risk of there being too few puppies in any of these combinations, the signs were further combined to produce three final outcomes: no sign of fear observed in either situation, at least one sign of fear observed in one of the two situations, and signs of fear observed in both situations. All the other parameters were considered as secondary (Table 1).

Missing data Consistent with the intention-to-treat principle, when there was a partial lack of data, for example, when data were collected from only the first questionnaire, rather than exclude the case, the results collected on day 3 were also applied to day 15, in accordance with the carrying forward imputation technique (ICH 1999). In this way, the analysis would not advantage the treatment. In the case of a total lack of data for one parameter, the imputation technique should take into account the reason for the lack of data (Shih 2002)

Statistical analyses The data were analysed using commercially available computer software (StatView 4.5; Abacus Concepts). Values of P<0·05 were considered significant, and values between P>0·05 and P<0·10 were considered as showing a tendency. Quantitative values were analysed using a t test. Qualitative parameters were analysed using a chisquared test of independence. When theoretical expected values were less than 5, Fisher’s exact test was used.


During the study, the pet shop received 100 puppies but 21 were returned for medical reasons. Of the 79 puppies that entered the randomisation (40 in the DAP group and 39 in the control group), 76 (39 and 37) were sold. Of these, 68 owners could be contacted (34 and 34), but two puppies in the DAP group lost their collars very soon after being adopted. As a result, 66 puppies (32 and 34) fulfilled the criteria required to participate in the study. The owner of only one puppy in the DAP group could not be contacted on day 15.

Demographic characteristics

The demographic characteristics of the 66 puppies are shown in Table 3. There were 39 males and 27 females. There were 24 different breeds, 10 of which were common to both treatment groups. Small breeds accounted for 68 per cent of the puppies, with medium and large breeds accounting for 17 per cent and 15 per cent respectively. The puppies remained in the pet shop for an average of 19 days and were nearly three months old when they were adopted.

For one third of the owners, the puppy was their first experience of dog ownership. Two thirds of the puppies went into a house that had a garden. There were no significant differences between the two treatment groups in terms of any of the puppies’ characteristics, the family composition or the owner’s previous experience of dog ownership (Table 3). However, this was not the case for the ‘level of activity/ noise in the environment where the puppy is usually taken for a walk’. The size of the sample was smaller for this parameter because two puppies (one in each treatment group) never went outside during the two weeks after they were adopted. This parameter could have influenced the puppy’s behaviour because it might have shown more fearful signs when walking in a loud and overcrowded area than in an isolated and calm one. The general level of activity and stimulation in the environments where the DAP puppies were walked appeared to be greater. This difference had an influence on the comparability of the two groups for the dependent variable ‘behavioural reactions during a walk outside’ and the main criterion therefore marginally favoured the control over the DAP group. However, as the analysis did not advantage the DAP treatment, the interpretation of the results of these dependent variables could proceed.

Assessment of efficacy

The main criterion was a composite variable based on two secondary parameters, and these parameters are analysed first.

Fear-related behaviours when facing unfamiliar people at home Regardless of date, the control puppies were more fearful (Fig 1). On day 3, significantly more of the DAP puppies engaged immediately in play with an unfamiliar person (P=0·027). This difference became more significant (P=0·017) on day 15 when 28 (87 per cent) of the DAP puppies and 21 (62 per cent) of the control puppies spontaneously engaged in playful contact with a neutral unfamiliar person (Table 2).


Numbers of dog-appeasing pheromone (DAP)-treated and control puppies reacting to the main criterion and to the two feareliciting situations that composed it, on days 3 and 15 after adoption. * 0·05>P>0·01, † 0·1>P>0·05

Fear-related behaviours when walking outside Three days after the puppies were adopted there was a slight difference (P=0·067) in the occurrence of fear-related behaviours when walking outside, with DAP puppies showing fewer signs of fear (Table 2, Fig 1). However, although fewer of the DAP puppies showed signs of fear, a higher proportion of them showed freezing behaviour on day 3. On day 15 the difference in fear-related behaviours became significant (P=0·026), with six (20 per cent) of the DAP puppies and 15 (45 per cent) of the control puppies showing fear-related signs.

Main criterion The data were incomplete for the two puppies that were not taken outside for a walk on day 3 or day 15 because their owners thought that puppies should not go outside before their vaccinations were completed. Because this lack of data was not due to the treatment, it was considered appropriate to implement an imputation technique. The main criterion was a composite of two other parameters, of which only one was known for the two puppies that were not taken outside (fear reaction when facing unfamiliar people at home). Each of the two parameters has two potential values (fear or no-fear) yielding four potential combinations of fear reactions for the main criterion. The imputation technique involved combining the single parameter that had been observed with the most common pairing that had been observed in the appropriate treatment group. The most frequent pairings observed were fear/fear and no-fear/no-fear, whatever the treatment. So, if the result for the first parameter was ‘fear’, the result for the missing parameter was imputed to be ‘fear’ and if the result of the first parameter was ‘no fear’, the result of the missing parameter was imputed to be ‘no fear’.

Two weeks after adoption, significantly more of the puppies in the control group had shown fear reactions since adoption (P=0·015) (Table 2, Fig 1). Twelve of the puppies (35 per cent) in the control group showed signs of fear in both situations, and two (6 per cent) of the DAP puppies did the same, and 17 puppies (50 per cent) of the control group and eight (25 per cent) of the DAP puppies showed at least one sign of fear.

Fear reactions when facing a new object Three days after adoption, only seven (21 per cent) of the control puppies and nine (28 per cent) of the DAP puppies reacted fearfully to the presence of a new object at home. Two weeks after adoption, five puppies in each group showed signs of fear in this situation. There was no significant difference between the two groups at any time. However, 10 of the puppies failed to make contact with the new object on day 3 compared with only four on day 15, and slightly more of the puppies in the control group failed to make contact on day 15 (Table 2, Fig 2).


Numbers of dog-appeasing pheromone (DAP)-treated and control puppies that showed different types of behaviours in response to fear-eliciting situations, on days 3 and 15 after they were adopted. * 0·05>P>0·01, † 0·1>P>0·05

Fear reactions after an order from the owner It was expected that by three days after adoption the puppies would consider their owner to be a familiar person, but 12 (35 percent) of the puppies in the control group showed fear-related reactions after an order from their owner, compared with four (12 per cent) of the DAP puppies. The difference between the two groups was significant (P=0·031) on day 3, but not on day 15. Three of the control puppies lost their fear of training commands by day 15, but there was no change in the behaviour of the DAP puppies (Table 2, Fig 2).

Fear-related behaviours when unfamiliar people tried to make contact at home Three days after adoption 14 (41 per cent) of the control puppies showed fear-related behaviours when unfamiliar people tried to make contact with them, compared with seven (22 per cent) of the DAP puppies (P=0·093); on day 15 the numbers were 14 and six respectively (P=0·048) (Table 2, Fig 2).

Results of fear-eliciting tests The different tests can be ranked according to the number of control puppies showing signs of fear. On day 3, fewest control puppies showed signs of fear when facing a new object (20 per cent); the next fewest were fearful when receiving an order from their owner (35 per cent); the two tests involving unfamiliar people caused signs of fear in 41 and 44 per cent of the puppies, and 52 per cent showed signs of fear when taken out. The same order was observed on day 15 for both the control and DAP puppies, but the proportion of DAP puppies still showing signs of fear was less than 20 per cent in all the tests. The proportion of control puppies showing signs of fear was always greater than the proportion of DAP puppies, except when the puppies were facing a new object.

Assessment of tolerability and compliance None of the owners indicated that they had deliberatively removed the collar for any reason.


These results show the potential value of DAP in preventing puppies’ fearful responses to novelty and suggest that it may help them to cope with the new situations commonly encountered when they are rehomed. Puppies coming from a pet shop that wore a DAP collar during the first two weeks after being adopted showed fewer signs of avoidance and uneasiness when facing unfamiliar people and new environments. There was also a significant effect on the puppies’ reaction to instructions from the new owner. The results varied, depending on the kind of fear-eliciting situation and the date of the evaluation, and the variations were in accordance with the literature (Boissy 1995), and the authors’ clinical experience that some fear-eliciting situations lead to more fearful reactions and are more difficult to cope with than others (Goddard and Beilharz 1984).

No difference was observed between the groups when the puppies encountered a novel object, a clinical situation used in open field tests (Boissy 1995, Beerda and others 1998). However, objects chosen in such studies are often presented in a way that is known to increase fear, for example, the opening of an umbrella (Goddard and Beilharz 1984, King and others 2003). In the present study, the novel objects were shopping bags brought home by the owner, and all the puppies were frequently exposed to them. The small number of puppies showing fear and the low intensity of their fear responses suggest that this situation was only mildly fear-eliciting. The majority of the puppies in both groups were easily able to cope with this situation as early as three days after adoption and there was no significant difference between them. This lack of response may have been because they had already had a number of neutral, or positive, experiences of the sound and appearance of shopping bags before they left the pet shop, and had been partially habituated to them. In addition, the puppies were being reunited with their owner, possibly after a period of isolation, at the same time as they encountered the shopping bags.

The first parameter used to assess the puppies’ ability to cope with social situations was their ‘reaction after an order from the owner’. Most of the puppies would have had experience of familiar people interacting with and speaking to them at the pet shop, but the experience of being given an instruction is likely to have been new, in terms of the style of communication, the requirement for a decision on the part of the puppy, and the expectation of an outcome. The puppies’ reactions to early experiences of commands and instructions is likely to have a significant bearing on their future responses to communications from their owner. On day 3 a small number of the DAP puppies showed fear when given an instruction compared with the control puppies, and this difference remained on day 15 although fewer of the control puppies showed signs of fear. Several of the puppies in each group remained fearful of instructions on day 15 and it is possible that this was due to a difference in the style of interaction by owners when giving commands (increased threat or forcefulness). However, twice as many of the control puppies as the DAP puppies showed persistent signs of fear in this situation on day 15, suggesting that the DAP enabled more of the puppies to adapt more quickly. This suggests first, that DAP enhanced the overall process of habituation, so that for 31 (94 per cent) of the DAP puppies it was already largely complete by day 3, and secondly, that DAP may have enabled the puppies to become habituated to a wider range of types of interactions, including the more fearful ones. The inference is that the control puppies became habituated more slowly and were less able to habituate to the more intense situations. The DAP treatment could therefore be considered to have helped the habituation process. This hypothesis is consistent with the finding of Sheppard and Mills (2003) that treatment with DAP does not modify a dog’s awareness or recognition of stimuli, but may allow it to react to stimuli as if they were less threatening. Thus, by day 3 the process of habituation to being given instructions had already ended for most of the DAP puppies but was still proceeding in the control puppies. By day 15 there was no significant difference between the two groups, but a few puppies in both groups still reacted negatively, possibly owing to individual differences, a behavioural condition such as kennel syndrome (Pageat 1986, Vastrade 1987), or differences between the assertiveness of tone in the instructions from the owners.

When the reaction of the puppies to unfamiliar people was assessed the contribution of other forms of stress and novelty was limited by confining the assessment to the home, which could be considered to be a ‘familiar place’ by day 3, and certainly by day 15. The presence of an unfamiliar person could then reasonably be considered to be the only fear-triggering stimulus. Two situations were explored: the reaction of the puppies to a stranger who made no direct attempt to interact with them, and their reaction to a stranger who did try to interact with them, similar to the ‘neutral stranger’ and ‘friendly stranger’ in the study by Tod and others (2005). On day 15, both these situations caused signs of fear in a higher proportion of the control puppies than any of the other tests. However, as observed by Denenberg and Landsberg (2008), the DAP puppies showed fewer signs of avoidance, fear or aggressive behaviours when facing unknown people. Only one puppy showed fearrelated aggression (growling) in this situation. On day 15 nearly 40 per cent of the control puppies still reacted negatively towards unfamiliar people, whereas only about 15 per cent of the DAP puppies did so. This underlines the difficulty that owners often meet when trying to socialise their dog, since inviting friends to the home two weeks after the puppies had been adopted still resulted in unpleasant or potentially dangerous behaviours.

The positive subcategories used were not exactly the same for the two situations involving unfamiliar people: the positive subcategory was ‘immediate playful contact’ for the neutral person, and ‘playful contact’ for the person making contact. This difference was small and is unlikely to interfere with a comparison of the results of the two test situations. In agreement with the results of the study by Tod and others (2005), there was a difference between the reactions of the puppies to ‘neutral’ and ‘friendly’ unfamiliar people. Unpredictability enhances fear reactions (Gray 1987, Morméde and Dantzer 1988, Boissy 1995) and it may be hypothesised that people trying to interact would be more stressful than neutral people. Thus, puppies might need more exposures to become habituated to people who try to interact with them, and it would also depend upon the outcome of such interactions as perceived by the puppy. This may explain why the difference between the groups only became significant on day 15 for interactions with unfamiliar people who tried to make contact, but was significant on day 3 for neutral unfamiliar people.

The observed trend that more control puppies were walked in isolated places and more DAP puppies were walked in overcrowded areas may partly explain the lack of difference between the two treatment groups on day 3. The trend in this confounding factor may also explain why a higher proportion of the DAP puppies showed freezing behaviour on day 3. A larger proportion of the control puppies showed a startle response compared with the DAP puppies, a difference consistent with the protective effect of DAP towards intensity of threat (Sheppard and Mills 2003). The results of this outdoor fear-triggering situation were also in accordance with the results observed at home, because five of the control puppies showed active avoidance behaviour towards unfamiliar people during outings compared with none of the DAP puppies. Outings in new places were considered to be the most highly fear-generating situations, as nearly half the control puppies showed fear reactions on day 3 and on day 15, and one third of them were undesirable (freezing, reluctance to walk or active avoidance of unfamiliar people) that might impair outings and lead to time spent with the puppy being less enjoyable for the owner.

The results also showed that the more a given situation involved multiple intense fear reactions, for example, unfamiliar people and outings in new surroundings, the higher proportion of the control puppies that showed signs of fear that persisted until day 15. In contrast, the DAP puppies appeared more able to habituate to such situations, suggesting that the DAP collars reduced the number of exposures needed to habituate the puppies to a given situation. The fact that 15 days was enough to lead to a significant difference between the two groups for the types of situation assessed suggests that owners should continue to use the collars for longer, to maintain their positive effect over a wider range of novel situations.

Validity of the results

The main criticism of the study is that it depends upon information obtained from the owners; the investigator had no control over how the owners presented the fear-eliciting stimuli or their observations of it. Precautions were taken to reduce the effect of this lack of control on the reliability of the results. During the telephone questionnaire, the investigator assessed specific, simple, common situations, to be sure that each puppy experienced the same relatively standardised situations repeatedly. Furthermore, any potential bias due to the subjectivity of the owners’ evaluations and responses was controlled by the structure of the questionnaire. The questions were semi-open, so that they had to choose between pre-defined answers. The use of telephone follow-up helped to reduce the number of owners who dropped out of the study, and allowed the investigator, a veterinary surgeon trained in behaviour, to provide clarification and additional detail when needed. To help the investigator to spot inconsistencies in an owner’s responses, the questions were designed to include some overlap, to encourage the owners to provide greater precision in their answers and improve the accuracy of the questionnaire.

Creating an overlap between the questions helped to improve the consistency of the questionnaire, but had the disadvantage of creating inter-dependency between the variables that made it inappropriate to combine them mathematically into a single main criterion. The first decision, therefore, was to choose two parameters, one assessing the puppy’s fear outside and the other its fear inside the home, so that a wide range of possible situations could be covered while maintaining the independency of the variables. The parameter ‘signs of fear during outings’ was unique because it was the only one to assess the puppy’s fear-related behaviour outside the home, whereas there were four feareliciting situations assessed inside the home. It is important that the main criterion should assess when it is supposed to. Unpredictability is known to increase fear (Gray 1987, Morméde and Dantzer 1988, Boissy 1995 and King and others 2003) and situations involving people were therefore considered to be more likely to elicit signs of fear. The puppies would have regarded the new owner as unfamiliar during the first few hours after it was adopted, but not after two weeks. Situations involving unfamiliar people on day 15 might therefore be considered to be more likely to induce fear than those involving an ‘order from the owner’. Of the two situations involving unfamiliar people, ‘presence of a neutral unfamiliar person’ was a more standardised situation than ‘unfamiliar person trying to make contact,’ because in this case the type of interaction could be very variable. The situation involving a neutral encounter with an unfamiliar person was therefore chosen to make up the other part of the main criterion because it was less susceptible to criticism.

Experimental studies carried out between the 1950s and 1970s (Scott and Fuller 1965, Wright 1983) and more recent clinical studies (Goddard and Beilharz 1984, Appleby and others 2002), have demonstrated the deleterious effects of restricted experience on a developing puppy and on its behaviour as an adult. Puppies bred in an impoverished environment with limited exposure to human contact before the age of 12 to 14 weeks will show extreme fear responses to any change in their environment and their social skills will be severely impaired (Overall 1997, Lindsay 2001, Casey 2002 and Landsberg and others 2003). Every care was taken to control confounding factors in order to ensure the comparability of the two treatment groups, but precise data about the early development of the puppies could not be obtained, and this lack of information has to be considered in the evaluation of the study. The only information about the puppies was that they had been obtained from French breeders (none had been imported) that were approximately 15 to 20 hours travel away from the pet shop. However, the fact that experience with novel objects produced relatively little fear in the puppies suggests that they had been reared in circumstances that were not exceptional or unusually detrimental to their development; as a result, the results could be generalised. The possibility that the breed composition of the puppies could have affected the generalisation of the results has been discussed by Gaultier and others (2008).

The results of this study were consistent with those of other studies and make it possible to suggest some future applications of pheromone treatments in situations involving novelty. Adoption is a highly stressful period for puppies, during which they experience a wide range of potentially fear-inducing new circumstances in their environment and social interactions, without the support of a social group or maternal figure.

The puppy’s response to its new home environment has implications for its development. If its developmental conditions differ widely from the domestic setting into which it is adopted, the puppy will show frequent fear reactions in common every day situations. It will avoid, bark at or attack unfamiliar people, so that outings will become difficult, unrewarding or even impossible for the owner. Some puppies will quickly learn that aggressive displays can drive fear-eliciting people away, and these puppies can become more dangerous as they learn that aggression is a more effective form of self-defence than avoidance. Such fear reactions seriously interfere with the relationship between the puppy and its new owner and may result in the owner relinquishing it or having it euthanased (Neilson 2002, Marston and Benett 2003, Mondelli and Prato Previde 2004).

The results of this study and those by Sheppard and Mills (2003) and Tod and others (2005) suggest that pheromone treatment might help to neutralise the deleterious effects of threatening stimuli. Puppies remain aware of and recognise the stimuli as potentially threatening (Sheppard and Mills 2003), but their effects are neutralised or limited, helping it to become habituated rather than sensitised. Sheppard and Mills (2003) and Tod and others (2005) proposed that pheromones might provide a palliative treatment to ease desensitisation-based treatments. For puppies, pheromones could be an effective and safe way to help them to cope with new places and unfamiliar people.

By decreasing the puppy’s expression of undesirable fearful behaviour, treatment with pheromone could help owners to expose it to a wider variety of circumstances. It is proposed that DAP could help to prevent the aggressive and avoidance behaviours associated with fear. Appleby and others (2002) showed that a lack of exposure of puppies to urban environments between three and six months of age was significantly associated with fear-based behaviours in adulthood. Denenberg and Landsberg (2008) showed that puppies wearing a DAP collar for two months during puppy classes achieved a better level of socialisation three months later. If longer-term studies confirm these results, treatment with DAP should be considered as of general benefit for the welfare of dogs and the safety of the public.


The authors thank John Bowen for his help in critically revising the content and providing English language corrections of this paper.


View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.