Hartpury Student Research Journal

Home » Issue 1 (Summer 2015) » Project Articles » The relationship between domestic cat (Felis catus) personality and attachment to owner

The relationship between domestic cat (Felis catus) personality and attachment to owner

Author Names:  Rachel Costa (BSc (Hons) Animal Behaviour and Welfare) and Dr Tamara Montrose

 

Abstract

The domestic cat is one of the most commonly kept companion animals. Their individual personalities have become a recent study topic of high interest. The most widely studied and accepted cat personality traits are related to activity, sociability, neuroticism, and aggression. In the past cats were not viewed as social animals, although it has now been accepted that cats do display social behaviours, such as their wide range of vocalisations, facial expressions, body postures and chemical signalling. However, the strength of the attachments that they are able to form is thus far unknown. This study aims to determine whether specific personality types affect the level of attachment that cats have with their owner. The personalities examined in this study are bold, playful, friendly and aggressive. The effects that specific cat personality traits have on attachment were determined via two different measures; a direct behavioural measurement and an owner assessed attachment questionnaire. Firstly the animal’s behavioural response to a novel object test and pick-up test was utilised to separate cats into distinct personality categories of bold, playful, friendly, or aggressive. An adaption of the Ainsworth Strange Situation test was then implemented within the animal’s home, enabling the comparison of 20 cats’ behaviours when in the presence of their owner, when alone and when with a stranger. A cat attachment questionnaire based on previous Canine attachment surveys was also completed by the animal’s owner. The owners scored their cats’ attachment behaviours based on how they interact with them.  A Friedman ANOVA was performed to compare each category of personality to the behaviours displayed by cats when with their owner, alone and with a stranger. With the additional use of a post-hoc Wilcoxon Signed-Rank test, significant differences between personality types and attachment were found. It was determined that bold, friendly and non-playful cats displayed a higher level of attachment behaviours towards their owners. The aggressive personality type was unable to be analysed due to an insufficient proportion of animals possessing aggressive traits. A Spearman’s rank correlation coefficient test revealed that there was a positive correlation between the playful personality and the cat’s attachment as scored by the owner. However playful cats were not observed to have a higher level of attachment to their owners during the behavioural measurements. This may suggest that the pet owners have an unrealistic perception of their animals’ attachment and may score them to display a higher level of attachment than they actually possess. During the study, attachment behaviours similar to those displayed by infants toward their mothers were observed. This may suggest that cats can also experience separation anxiety; a disorder commonly related to attachment. The results showed bold, friendly and non-playful cats to have the highest levels of attachment toward their owners, which may suggest that they are most susceptible to separation anxiety. However, this study conflicts with previous research undertaken on the effects of personality, which found shyness to have a significant impact on the formation of social ties. Therefore it is essential to differentiate between the effects of personality on social behaviours and attachment behaviours.

 

1.0 Introduction

Personality in the domestic cat has become a study topic of interest as a result of the increasing number of cats kept as companion animals (Raihani et al., 2014; Gartner and Weiss, 2013). Personality is defined as consistent individual differences in behaviour across time and context (Bergmiller, 2010; Briffa and Weiss, 2010). The five factor model of behavioural traits observed in humans include: Extraversion, Neuroticism, Conscientiousness, Agreeableness and Openness (Costa and McCrae, 1992). This five factor model can also define personality traits in non-human animals (Briffa and Weiss, 2010). The most commonly studied and accepted cat personality traits are related to activity, sociability, neuroticism, and aggression (Gartner and Weiss, 2013).

Animal personality studies are important as they can allow for the improvement of an animal’s welfare conditions (Gartner and Weiss, 2013; Gartner and Powell, 2012; Iki et al., 2011). Previous personality research has found that reduction of stress can be achieved by catering for an animal’s individual differences; for example, by providing a fearful animal with more access to hiding places in its environment (Gartner and Powell, 2012). Personality experiments can reveal whether a cat’s personality affects their suitability to be homed in a multi-cat household (Ramos et al., 2013). They also help to determine the correct sex ratio of the social groupings, such as preventing conflict occurring between two dominant males within the same household (Iki et al., 2011). The level of enrichment needed within an animal’s environment may also be dependent on their personality (Rochlitz, 2005). Understanding these differences may permit animals to be paired with potential owners, increasing their adoption success (Casey and Bradshaw, 2008). It also allows for the identification of personality-related behavioural problems. For instance, the reason for a cat’s inappropriate elimination may be due to its boldness/fearfulness, or could be related to its level of dominance (Iki et al., 2011).

Attachment is a common behaviour amongst social animals, and is displayed by infants towards their mothers and in social and romantic relationships (Edwards et al., 2007; Noftle and Shaver, 2006). The theory of attachment was first hypothesised by Bowlby (1958) who stated that attachment is an innate behaviour. Its two main functions are to ensure that the young stays close to another animal for warmth and protection, and for mothers to stay close to their young to ensure the survival of their offspring. Health benefits of attachment are also displayed in social relationships, such as an increase in reproductive success and allogrooming behaviours in primates (Mitani, 2009; Silk, Altmann and Allberts, 2006).

In the past, cats were not viewed as social animals (Schwartz, 2002) although it has now been accepted that cats do display social behaviours, including  their wide range of vocalisations, facial expressions, body postures and chemical signalling (Turner and Bateson, 2014; Schwartz, 2002). Edwards et al. (2007) discovered that cats are able to form social attachments, but the strength of these connections is yet to be determined. Separation anxiety is a disorder related to attachment and is often displayed by dogs when separated from their attachment figure (Storengen et al., 2014). It can be characterised by excessive vocalisation, inappropriate elimination, trembling, and an increased heart rate (Storengen et al., 2014; Sherman and Mills, 2008). The disorder can have a detrimental impact on an animal’s welfare. (Parthasarathy and Crowell-Davis, 2006). It is therefore essential to determine if the disorder is present in cats to allow for its identification and management.

Although no research has been undertaken into the relationship between personality and attachment in cats, it has previously been explored in humans and other species (Asendorpf and Wilpers, 1998). Asendorpf and Wilpers (1998) discovered that shy individuals are likely to make more permanent attachments than those formed by bolder individuals. Croft et al. (2009) also found that shy fish were able to form a higher quantity of strong relationships than bolder fish.

The majority of studies conducted on attachment focus on the level of attachment that the owner feels toward their pets (Winefield, Black and Chur-Hansen, 2008; Crawford, Worsham and Swinehart, 2006; Serpel, 1996).  More research needs to be undertaken to establish the attachment that animals have with their owners and to discover whether specific personalities affect this bond (Edwards et al., 2007). This research will help increase re-homing rates, depending on the specific traits and level of attachment the owner requires with their pet (Casey and Bradshaw, 2008). It will also help to identify attachment behaviours and enable the management of the problems associated with them, such as separation anxiety (Schwartz, 2003). A study conducted by Serpell (1996) explored the effects that cat personalities have on the owner’s attachment levels, but thus far no research has occurred into individual personality and cat attachment.

The main aim of the project was to determine if there is a relationship between a cat’s personality and the attachment it has with its owner. The first objective of this study was to identify whether the boldness, playfulness, friendliness, or aggressiveness of an animal affects the level of attachment it has to its owner. The second objective was to investigate the effects cat personality traits have on attachment via two different measures; the first being an owner-assessed questionnaire and the second a direct behavioural measurement.

 

2.0 Methodology

2.1 Site and subjects

During this experiment, 20 cats were monitored. This figure was based on similar personality studies of domestic cats and should enable any effects to become apparent (Siegford et al., 2003; Lowe and Bradshaw, 2001). Adult cats between the ages of 1-10 years old were utilized; older animals would not have been suitable due to their increased likelihood of health issues or cognitive dysfunction syndrome (Landsberg, Denenberg and Araujo, 2010). Younger cats are also not eligible, as animal personality is affected greatly by the socialisation they receive during their critical period between approximately 2-9 weeks of age (Casey and Bradshaw, 2008).

The experiments were conducted within the animals’ homes to reduce the stress that they would experience as a result of transportation (Dickens, Delehanty and Romero, 2010). Similar sized rooms ranging from 25m² to 32m² were selected for the study. Cats are crepuscular and are most likely to hunt during dusk and dawn (Turner and Bateson, 2014). This anticipation for the hunt could have an effect on the cat’s levels of activity and aggression (Rodan, 2010), therefore the tests took place well before dusk, between 1-3pm (Aslimoski et al., 2014).

 

2.2 Personality Test

The cat’s level of boldness, playfulness, friendliness, and aggressiveness was tested using a novel object test and its reaction to contact with a stranger. Firstly, the observer placed the novel object (a stuffed owl toy) in the study room approximately 5ft away from the animal.  An ethogram based on Wedl et al.’s (2011) ethogram to measure behavioural processes was utilised.  To measure the cat’s boldness, the duration it took for the animal to initially approach the novel item was recorded. The ethogram included whether the cat ignored the object, looked at the object, hesitated at the introduction of the object, retreated from the object, sniffed the object, or marked the object; all of which measured its boldness. Its playfulness was determined by the frequency and duration of time it spent playing with the object, and its aggressiveness was calculated by the frequency of the cat’s attempts to threaten or attack the object (Wedl et al., 2011).

For the second test, the experimenter approached the cat and stopped roughly 5ft away from it, crouching down with their hand extended and beckoning the cat. If it did not approach after 30 seconds the examiner moved closer to the cat, stopping around 3ft in front of it (Iki et al., 2011). The cat was then petted and picked up if it was possible to do so. Iki et al.’s (2011) feline temperament ethogram and Wedl et al.’s (2007) behavioural process ethogram were used in conjunction to compose the ethogram implemented in this section of the study. Table 1 shows the behaviours used to determine the cat’s levels of friendliness and aggressiveness.

 

Table 1: List of observed test behaviours.

Behaviour Friendly Aggressive
Body Posture Relaxed Defensive, Threatening
Vocalisations Meow Hiss
Contact with examiner Sniff, rub lick Strike with paw, bite

 

The cat’s friendliness was also measured by its willingness to approach the examiner and the ease with which the cat was picked up. An ordinal scale was used to record the approach and handling test, with a quick approach and acceptance of handling being a friendliness score of 2, slow approach and hesitance of handling being 1, and avoidance of approach and handling being a score of 0 (Wedl et al , 2011).

 

2.3 Ainsworth Adapted Strange Situation Test

The original Ainsworth Strange Situation Test was used to determine an infant’s level of attachment to its mother (Ainsworth et al, 1978). Unlike previous adaptions (Mariti et al., 2013; Edwards et al., 2007) this study would not allow for all experiments to be conducted within the same environment, due to the wide spread locations of subjects. The experiments therefore took place in the owners’ homes.

The test was divided into 7 different sections with a duration of 3 minutes each, and was used to determine the differences in the cat’s behaviour whilst with its owner, with a stranger, and whilst alone. The specific times stated in the experimental procedure are based on Edwards et al.’s (2007) Ainsworth Adapted Strange Situation Test using cats. Previous strange situation tests using dogs also follow similar time frames during the experimental period (Mariti et al 2013; Parthasarathy and Crowell-Davis, 2006).

Introduction- 30 seconds: Owner and cat are introduced to their experimental area.

Section 1- Cat and owner: Owner does not engage with cat until 1.5 minutes, when they then initiate play.

Section 2- Stranger, owner and cat: Stranger enters the room, takes a seat, and does not begin conversation with owner for 30 seconds. At 2 minutes, stranger initiates play with cat, and owner leaves without any further interactions.

Section 3- Stranger and cat: Stranger initiates play with cat; if cat does not play then stranger attempts to pet cat. After 2 minutes stranger ceases play, but may continue petting if initiated by cat.

Section 4- Owner and cat: Owner calls cat from outside the door and enters the room, waiting for cat’s response. Owner then pets cat whilst stranger leaves the room. After 2 minutes, owner leaves the room again.

Section 5- Cat: Cat is left without owner or stranger for this period.

Section 6- Cat and stranger: Stranger enters the room and initiates play with cat; if cat does not play then stranger attempts to pet cat. After 2 minutes stranger ceases playing, but may continue petting if initiated by cat.

Section 7- Cat and owner: Stranger focuses their attention on cat. Owner enters the room and pauses, allowing cat to respond before greeting them. Owner does not play with/pet cat, but allows it if cat initiates it (Ainsworth et al, 1978).

Continuous sampling was employed during the behavioural observations, and an ethogram based on Edwards et al’s (2007) cat attachment study was used to record behaviours displayed by the cats. The behaviours observed during the experiments were as follows: locomotion, alertness, inactivity, approach to the door, physical contact, vocalisations, marking, and playing. The frequency and duration of these behaviours were recorded (Edwards et al., 2007).

To determine the cat’s level of attachment, the frequencies and durations of the cat’s behaviours upon the return of its owner (section 4 and 7) were compared to the behaviours displayed when the cat was alone (section 5) and when the stranger returned to the cat (section 6) (Edwards et al., 2007; Ainsworth et al., 1978).

 

2.4 Questionnaire

In conjunction with the Ainsworth Adapted Strange Situation Test, a questionnaire was administered to the owners to gauge their perception of their cat’s attachment. The questionnaire was based on an adaption of the Canine Behavioural Assessment & Research Questionnaire (Serpell, 2014). The questionnaire consisted of 6 questions:

Does your cat:

  1. Show a strong bond with one person in your house?
  2. Follow you around the house?
  3. Sit close to you or in contact with you when sitting down?
  4. Initiate contact when you are sitting down?
  5. Become agitated when you display affection to another person?
  6. Become agitated when you display affection for another cat or animal?

The owner was asked to rate their animal’s behavioural traits on a Likert scale of 1-5, 1 being not at all and 5 being a great deal. The results from the questionnaire were used alongside the results obtained from the Ainsworth Adapted Strange Situation Test to determine the animal’s level of attachment.

 

2.5 Statistical Analysis

To test the difference between a cat’s personality and their attachment to its owner, it would seem that the best method would be to use a robust two-way mixed ANOVA, which is a form of generalized linear model (GLM). This is due to the fact that there are two independent variables; one being the personality test and another being the Ainsworth Strange Situation Test, with a mixture of between-group and repeated-measure variables (Hawkins, 2009). The Strange Situation Test requires a between-group design, because different participants are assigned to different groups; the cat is in all situations, but the owner and stranger are only introduced to selected situations. However, the personality test is a repeated measure as the same subjects will be used during the experiments. Although this does seem to be the best method of analysis, it is not applicable to non-parametric data used with SPSS and could only be generated using R. Given the non-parametric state of the data, it would not be possible to use this form of analysis (Field, 2013).

Consequently, the most appropriate form of statistical analysis to use was a Friedman ANOVA test of difference. This is due to the fact that the data is related and the repeated measures were conducted on the same set of participants (Field, 2013; Hawkins, 2009). Additionally, the test allows for essential comparisons to be made between groups of dependant variables (behaviours) and independent variables (when with the owner, alone, or stranger) (Martin and Bateson, 2007). Firstly, the means of personality traits were calculated to separate the cats into distinct sections. To determine the cat’s boldness, the mean time it took each cat to approach the novel object was calculated. All cats below this mean number were classed as bold. Only 5 cats engaged in play with the novel object, and therefore these cats were deemed playful. However, if a larger proportion of the participants had played with the object, then the mean time the cats spent playing with the item for would have been calculated, and everything above the mean number would have been categorised as playful. Again, for friendliness and aggression the means were calculated using the frequency that the cats expressed friendly and aggressive behaviours during the novel object and pick up test.

The Friedman ANOVA test was then used to compare the different categories of personalities whilst the cat was with the owner, with a stranger, and alone. If a difference was found then a post hoc Wilcoxon Signed-Rank Test was used to determine exactly where the difference lay. The post hoc Wilcoxon test was applied as it is suitable for analysing non-parametric related data (Hawkins, 2009). A Bonferroni correction was also applied to the post hoc test as it reduces the risk of false positive results from occurring. This is a common problem when conducting multiple comparisons on the same set of data (Perry, 2004). Considering the standard significance level is 0.05 and there were 3 categories of comparison (owner-stranger, alone-owner, and stranger-alone), the new significance level was 0.05/3=0.017. Therefore if the P value was below 0.017 then it was classed as significant (Perry, 2004).

Finally, a test of relationship was carried out to analyse the results of the cat attachment questionnaire to determine if there was a relationship between personality and attachment. A Spearman’s rank correlation coefficient test was used due to the non-parametric state of the data (Hawkins, 2009). This test allows two related variables to be compared, one being personality and the other attachment, and determines the strength of relation between these variables (Hawkins, 2009). The totals from the Likert scale were combined, providing a calculation of the individual animal’s level of attachment out of a possible score of 30. These results were then compared to the total score of each individual’s boldness, playfulness, friendliness, and aggressiveness that were previously obtained from the novel object and pick up test.

 

3.0 Results

3.1 Personality means

A combination of the frequency and duration data obtained from the experimental personality tests were used to determine the means of each personality. The mean boldness of the cats was 0.69 (Standard deviation (S.D)=1.11), hence all cats below this score were deemed as bold (Dyer et al., 2008). In total there were 12 bold cats and 8 non-bold cats. The mean score for friendliness was 3.8 (S.D=2.63). There were 12 cats above this mean number and were thus labelled as friendly, and there were 8 cats below the mean that were classed as not friendly. Only 5 cats performed play behaviours during the experiments, therefore these cats were characterised as playful and the 15 remaining cats were classed as not playful. Only 2 cats displayed aggressive behaviours throughout the personality experiments. Therefore it was not possible to analyse this category of personality as the small sample size would not allow for accurate or reliable results (Grissom and Kim, 2012).

 

3.2 Behavioural comparisons of the bold personality

3.2.1 Bold

Rubbing:

Statistical analysis revealed a significant difference in the cats’ rubbing behaviours whilst with the owners, alone and with strangers (X²(2)16.294, P=0.000). To determine where the difference occurred, a post-hoc Wilcoxon Signed-Rank Test with a Bonferroni correction was applied, resulting in a significance of P<0.017. The test displayed a significant difference between the rubbing behaviours of the cat towards the owner (mean rank=6.72) and stranger (mean rank=2.75) (z=-2.446, P=0.014). It also revealed differences between the owner (mean rank=5.50) and when the cat was alone (mean rank=0.00) (Z=-2.805, P=0.005). Figure 1 illustrates the differences observed between the conditions.

CostaFigure1

Figure 1: Mean rank of rubbing during conditions.

 

Purring:

Statistical analysis displayed a significant difference in the cats’ purring behaviours throughout the three conditions (X²(2)=9.905, P=0.007). A Wilcoxon Signed-Rank test with a Bonferroni correction with a significance level of P<0.017 was then applied. A significant difference was determined between the cats when alone (mean rank=0.00) and when with a stranger (mean rank=4.50) (Z=2.527, P=0.012), as displayed in Figure 2.

CostaFigure2

Figure 2: Mean ranks of purring during conditions.

 

Meowing:

A significant difference in the cats’ meowing behaviours between the conditions was found upon statistical analysis (X²(2)=2.000, P=0.368). A Wilcoxon Signed-Rank test with a Bonferroni correction with a significance level set at P<0.017 was then applied. This displayed a significant difference between the meowing behaviour of cats when with their owner (mean rank=4.50) compared to a stranger (mean rank=0.00) (Z=-2.524, P=0.012), as displayed in Figure 3.

CostaFigure3

Figure 3: Mean rank of meowing during conditions.

 

3.3 Behavioural comparisons of the friendly personality

3.3.1 Friendly

Rubbing:

A significant difference in the rubbing behaviours of cats displayed with the owner, when alone and when with a stranger was found upon statistical analysis (X²(2)=13.029, P=0.001). A Wilcoxon Signed-Rank Test with a Bonferroni correction with a significance level set at P<0.017 was then applied. There was a significant difference in the rubbing behaviour between the cat when it was alone (mean rank=0.00) and with its owner (mean rank=5.50) (Z=-2.805, P=0.005), as displayed in Figure 4.

CostaFigure4

Figure 4: Mean rank of rubbing during conditions.

 

Facial Relaxation:

Statistical analysis displayed a significant difference in the cats’ facial relaxation behaviours between the conditions (X²(2)=6.350. P=0.042). A Wilcoxon Signed-rank Test with the Bonferroni correction with a significance level set at P<0.017 was then implemented. It revealed a significant difference between the facial relaxation behaviour displayed between the owner (mean rank=6.50) and the stranger (mean rank=1.50) (Z=-2.499, P=0.012), as displayed in Figure 5.

CostaFigure5

Figure 5: Mean rank of facial relaxation during conditions.

 

3.4 Behavioural comparisons of the playful personality

3.4.1 Not playful

Body Relaxed:

Statistical analysis revealed a significant difference in the cats’ relaxed behaviours when with their owners, alone and with strangers (X²(2)=10.863, P=0.004). A Wilcoxon Signed-Rank Test and a Bonferroni correction with a significance level of P<0.017 was applied. A significant difference was then found between the stranger (mean rank=0.00) and owner (mean rank=5.00) (Z=-2.668, P=0.008), as displayed in Figure 6.

CostaFigure6

Figure 6: Mean rank of relaxation during conditions.

 

Facial Relaxation:

Statistical analysis revealed a significant difference in the cats’ facial relaxation between the conditions (X²(2)=10.863, P=0.004). A Wilcoxon Signed-Rank Test with a Bonferroni correction with a significance level of P<0.017 was then implemented. There was a significant difference in the cats’ level of facial relaxation when in the presence of their owners (mean rank=7.00) compared to a stranger (mean rank=1.00) (Z=-2.981, S.D=0.003), as displayed in Figure 7.

CostaFigure7

Figure 7: Mean rank of facial relaxation during conditions.

 

Facial Alertness:

A significant difference in facial alertness when with the owner, alone and with a stranger was found (X²(2)=6.250, P=0.044). A post hoc Wilcoxon Signed-Rank Test with a Bonferroni correction with a significance level of P<0.017 was applied. This revealed a significant difference between a cat’s level of facial alertness when with its owner (mean rank=4.67) and with a stranger (mean rank=8.27) (Z=-2.417, P=0.016) as seen in Figure 8.

CostaFigure8

Figure 8: Mean rank of facial alertness during conditions.

 

Rubbing:

Statistical analysis revealed a significant difference in the cats’ rubbing behaviours between the conditions (X²(2)=12.684, P=0.001). A post hoc Wilcoxon Signed-Rank Test with a Bonferroni correction with a significance level set at P<0.017 was then applied. This revealed a significant difference in the cats’ rubbing behaviours between the owners (mean rank=6.78) and strangers (mean rank=2.50) (Z=-2.492, P=0.013) and between the cats alone (mean rank=0.00) and the cats with their owners (mean rank=6.00) (Z=-2.941, P=0.003). Figure 9 illustrates the differences observed between the conditions.

CostaFigure9

Figure 9: Mean rank of rubbing during conditions.

 

3.5 Personality comparisons of the owner assessed attachment questionnaire

A Spearman’s rank correlation coefficient test revealed that there was a positive correlation between the playful personality and the cat’s attachment as scored by the owner, as displayed in Figure 10. The correlation was statistically significant (rs(18)=0.503, P=0.024). There was no statistical significance between bold cats and attachment scores (rs(18)=-0.330, P=0.156) or friendly cats and attachment scores (rs(18)=0.163, P=0.20).

CostaFigure10

Figure 10: Correlation of the playful personality and the owners attachment score.

 

4.0 Discussion

The main aim of this project was to determine if there is a relationship between a cat’s personality and its attachment to its owner. Bold, friendly, and non-playful cats displayed a variety of significant attachment behaviours towards their owners compared to when alone or with a stranger. A positive correlation was also present between the playful personality and the cat’s attachment strength as scored by their owners. These findings suggest that a cat’s personality does in fact impact upon the level of attachment a cat has with its owner.

 

4.1 Behavioural traits and attachment

Significantly higher levels of rubbing were present in bold, friendly, and non-playful cats when with their owners compared to when alone or with a stranger. Rubbing is an affiliative behaviour that enables an exchange of pheromones, allowing for the recognition of familiar objects and colony members (Ellis et al., 2013). Case (2003) declared that rubbing is a form of friendly communication allowing cats to feel a sense of ease whilst with their owners. Edwards et al. (2007) also stated that marking via rubbing may be an indicator of the high level of comfort that a cat experiences whilst with its owner. The rubbing behaviours displayed by the cats within this study demonstrated their ability to distinguish between their owners and unfamiliar people (Saito and Shinozuka, 2013). It may also suggest the greater sense of ease and comfort they experience when with their owners compared to when alone or with a stranger. Within colonies, cats have been observed to have preferred associates with which they frequently engage in allorubbing (Cromwell-Davis, Curtis and Knowles, 2004).  Therefore, rubbing is likely to be an indication of attachment as it is consistently displayed towards specific individuals (Ellis et al., 2013). These findings correspond with Edwards et al.’s (2007) study who observed an increase in the rubbing of cats whilst with their owner compared to when alone or with a stranger.

Purring is predominantly a behaviour that displays a level of contentment in the domestic cat (Eldredge et al., 2008). However, purring may also be exhibited by a cat that is stressed or in pain (Schotz and Eklund, 2011). It has also been hypothesised that purring may serve as a social indicator that the cat does not pose a threat to others (Eldredge et al., 2008). During this study, bold cats most frequently purred when with a stranger rather than when alone. As no other behavioural indicators of alertness or anxiousness were observed, it is fair to assume that this was unlikely to be caused by stress (Rodan et al., 2011). This may therefore comply with the previous theory of purring to show a lack of threat, as the only significant level of purring was performed when in the presence of a stranger.

Bold cats were observed to meow significantly more frequently when with their owner compared to with a stranger. Meowing has been classed as a friendly vocalisation (Rodan, 2010) commonly occurring in human–cat interactions (Yeon et al., 2011). During a section of the Strange Situation Test, owners were instructed to call their cats from outside the room before entering, which prompted the cats to meow. Saito and Shinozuka (2013) previously discovered that the domestic cat is able to differentiate between the vocalisation of their owner and a stranger. The ability to distinguish owners from unfamiliar people reinforces the theory that cats are able to form stronger attachments with certain individuals. The results from this study also correspond with that of Edwards et al. (2007) who discovered a higher level of vocalisation in cats when with their owners.

Additionally, friendly and non-playful cats displayed significantly higher levels of facial relaxation when with their owner compared to with a stranger. Conversely, a higher level of alertness was displayed by non-playful cats when with a stranger rather than their owner. Facial alertness has previously been identified as an indicator of anxiety (Rodan et al., 2011). This may have been expressed due to fear of the unfamiliar person (Rodan et al., 2011) or an anxiety-related reaction due to the absence of their owner (Mariti, 2013). Similar expressions of anxiety are present in dogs that are parted from their attachment figures (Parthasarathy and Crowell-Davis, 2006) and can often be classed as an indicator of separation anxiety (Mariti, 2013).

 

4.2 Non-significant behaviours

These findings conflict with previous research on the connections between personality and social ties. Studies conducted on humans, fish, and birds concluded that less bold individuals were more likely to form strong social bonds (Aplin et al., 2013; Pike et al., 2008; Neyer and Lehnart, 2007). However, this study found no connections between shy individuals and their attachment levels. The previous studies focused on the relationship between personality and social connections rather than attachment (Aplin et al., 2013). Therefore further research must be conducted to differentiate between personality effects on social behaviours and attachment behaviours.

Locomotion and exploration in the presence of an owner has previously been identified as a measure of attachment (Edwards et al., 2007). Human infants were first observed to explore their environment more when in the presence of their mother compared to when alone or with a stranger, using their mothers as a reference point whilst exploring (Ainsworth, 1978). Similar attachment behaviours have previously been seen during the Strange Situation Test when adapted to the use of cats, dogs and primates (Topal, 1998; Harlow and Harlow, 1965). The cats in this study, however, did not display this form of attachment behaviour and may therefore conflict with the theory that personality affects attachment. However, as the experiments were conducted within the animals’ own homes, it may render the need for exploration to be unnecessary as the animals have already acclimatised to their environments (Quinn et al., 2009).

 

4.3 Owner-assessed attachment questionnaire

The results of the owner-assessed attachment questionnaire revealed that cats with a playful personality had stronger attachments to their owners, whereas boldness and friendliness had no effect on attachment. However, these results contradict those that were found during the direct behavioural observations, where in fact only the non-playful, friendly, and bold personalities were observed to have an impact on attachment. This may suggest that pet owners are unable to correctly identify behavioural indicators in their animals (Wemelsfelder, 2007). This could be due to that fact that anthropomorphism regularly occurs in pet owners, therefore they may have been scoring for human behavioural indicators rather that the more complex feline signals (Rollin, 2007). These findings conflict with those of Konok, Doka and Miklosi (2011) who reported that owners had a realistic view of their dog’s separation behaviours. This may imply that an owner’s attachment awareness is dependent on the type of pet they own (Rollin, 2007). Further research should be conducted to determine whether species and breed has an impact on a pet owner’s awareness of attachment.

 

4.4 Limitations

One of the largest limitations of this study was due to the fact that the Ainsworth Adapted Strange Situation Tests were conducted within the animals’ homes, and were therefore not in an unfamiliar environment. A strange situation was originally implemented as it allows for home behaviours and strange-situation separation behaviours to be distinguished (Ainsworth, 1978). It is possible that alternative results may have been obtained if the tests were conducted in an unfamiliar study site. However, the tests were conducted within certain rooms of the house that are less frequently explored by the individual animals. The study situation which the cat was subjected to was also unfamiliar. Therefore a strange situation, in some respects, was simulated.

The small sample size of the study, consisting only of 20 cats, was another limitation (Arnaud et al., 2007). A larger population would likely have revealed a wider range of personalities (Dingemanse and Reale, 2005). This would have enabled the analysis of the aggressive personality and presumably would have provided a more proportional number of participants in each category of personality (Dingemanse and Reale, 2005). However, previous personality studies conducted on cats utilised a similar range of participants and still produced valid and reliable results (Siegford et al., 2003; Lowe and Bradshaw, 2001).

 

4.5 Further research

As an insignificant number of aggressive animals were identified in this study, analysis of this personality type was not conducted. Aggression towards humans and other animals is the main reason for euthanasia in domestic cats (Dantas-Divers et al., 2011). If aggression had an impact on attachment, then it is possible that the cat may be displaying fear-related aggression directed towards owners or strangers due to separation anxiety (Banberger, 2006). This has previously been displayed by dogs in an apparent attempt to prohibit their owner’s departure (Sherman and Mills, 2008). Separation-related aggression can be displayed upon the owner’s return, as well as before their departure (Sherman et al., 2007). Further research is essential for the treatment and prevention of aggression-related behaviour problems, and to reduce the amount of aggression-related feline euthanasia.

The results of this study support the theory that cats may be likely to suffer from separation anxiety. Similarly to Edwards et al.’s (2007) experiment, cats that displayed a high level of attachment were observed to greet their owner, follow them, initiate physical contact, and jump onto their laps when they were seated. These were also the attachment indicators that were observed in infants during the original Ainsworth Strange Situation Test (1978). The results of this study revealed that bold, friendly, and non-playful personality types expressed a higher level of attachment behaviours towards their owners compared to when alone or with a stranger. This suggests that these personality types may experience stronger attachments and may therefore be more susceptible to separation anxiety. Further research is essential to fully validate these findings, and to further explore the symptoms and risk factors of separation anxiety in the domestic cat.

 

5.0 Conclusion and Applications

This study suggests that specific personality traits of the cat affect their level of attachment to their owner. Cats with bold, friendly, and non-playful personality types displayed more attachment behaviours towards their owners compared to when alone or with a stranger. The aggressive personality type was unable to be analysed due to an insufficient proportion of animals possessing aggressive traits. Therefore it is still undetermined whether aggression has an impact on attachment.

Playful cats were determined to have a higher level of attachment to their owners as a result of the owner-assessed attachment questionnaires. However, playful cats were not observed to have attachments with their owners during the behavioural measurements. This suggests pet owners may have an unrealistic perception of their pet’s attachment due to an inadequate ability to distinguish feline behaviours.

This study conflicts with previous research undertaken on the effects of personality on social ties, which found shyness to have a significant impact. It is therefore essential to differentiate between the personality effects on social behaviours and attachment behaviours. The results of this study further validate the theory that the domestic cat is able to experience separation anxiety when not in the presence of their owner. As bold, friendly, and non-playful cats were observed to have a higher level of attachment to their owners, this may suggest that they are also more susceptible to separation anxiety. A comprehensive understanding of the relationship between personality and attachment will also enable cats to be suitably paired to potential owners based on their personality type and the strength of attachment they desire.

 

References

Ainsworth, M., Blehar, M., Waters, E., Wall, S. (1978) Patterns of Attachment: Psychological Study of the Strange Situation.  Erlbaum:Hillsdale

Aplin, L., Farine, D., Morand-Ferron, J., Cole, E. (2013) Individual personalities predict social behaviour in wild networks of great tits (Parus major) Ecology letters [online] 16(11) p. 1356-1372 [Accessed 15th February 2015]

Arnaud. S., Duarte, M., Alberto, F., Serrao, E. (2007) Standardizing methods to address clonality in population studies. Molecular Ecology [online] 16 p. 5515-5139 [Accessed 27th February 2015]

Asendorpf, J., Wilpers, S. (1998) Personality effects on social relationships. Journal of Personality and Social Psychology [online] 74(6) p. 1531-1544 [Accessed 15th February 2015]

Aslimoski, P., Atanasoski, S., Krstanoski, N., Tuteska, J., Hristobska, O. (2014) Horizons. International Scientific Journal [online] 1 p. 7-17 [Accessed 5th March 2015]

Bergmiller, R. (2010) Animal personality and behavioural syndromes. Animal Behaviour: Evolution and Mechanisms [online] 10 p. 587-621 [Accessed 24th February 2015]

Briffa, M., Weiss, A. (2010) Animal personality. BioScience [online] 20(21) p. 912-914 [Accessed 17th February 2015]

Bowlby, J. (1958) The nature of the child’s tie to his mother. International Journal of Psycho-analysis. [online] 39 p. 350-373 [Accessed 25th February 2015]

Casey, R., Bradshaw, J. (2008) The effects of additional socialisation for kittens in a rescue centre on their behaviour and suitability as a pet. Applied Animal Behavioural Science [online] 144(1-2) p. 196-205 [Accessed 18th November 2014]

Croft, D., Krause, J., Darden, S., Ramnarine, I., Faria, J., James, R. (2009) Behavioural trait assortment in a social network: patterns and implications. Behavioural Ecological Socio-biology [online] 63 p. 1495-1503 [Accessed 22nd February 2015]

Costa, P., McCrae, R. (1992) Four ways five factors are basic.  Personality and Individual  Differences [online] 13, p. 653–665 [Accessed 21st February 2015]

Crawford, E., Worsham, N., Swinehart, E. (2006) Benefits derived from companion animals, and the use of the term ‘attachment’. Anthrozooz [online] 19(2) p. 98-112 [Accessed 07th February 2015]

Dickens, M., Delehanty, D., Romero, M. (2010) Stress: An inevitable component of animal translocation. Biological Conservation [Online] 143(6) p. 1329-1341 [Accessed 25th February 2015]

Dingemanse, N., Reale, D, (2005) Natural selection and animal personality. Behaviour [online] 142 p. 1165-1190 [Accessed 10th March 2015]

Dyer, J., Croft, D., Morrell, L., Krause, J. (2008) Shoal composition determines foraging success in the guppy. Behavioural Ecology [online] 20(1) p. 165-171 [Accessed 07th March 2015]

Edwards, C., Heiblum, M., Tejeda, A., Galindo, F. (2007) Experimental evaluation of attachment behaviours in owned cats. Journal of Veterinary Behaviour [online] 2(4) p. 119-125 [Accessed 10th December 2014]

Eldredge, D., Carlson, D., Carlson, L., Griffin, J. (2008). Cat Owner’s Home Veterinary Handbook. (3rd Edition) Wiley Publishing: New Jersey

Ellis, S., Rodan, I., Carney, H., Heath, S., Rochlitz, I., Shearburn, L., Sundahl, E., Westropp, J. (2013) AAFP and ISFM Feline Environmental Needs Guidelines. Journal of Feline Medicine and Surgery [online] 15 p. 219-230 [Accessed 08th March 2015]

Field, A. (2013) Discovering statistics using IMB SPSS statistics. (4th edition) Sage: London

Gartner, M., Powell, D. (2012) Personality assessment in snow leopards (uncia uncia). Zoo Biology [online] 31 p. 151-165 [Accessed 25th February 2015]

Gartner, M., Weiss, A. (2013) personality in fields: A review. Applied Animal Behaviour Science [online] 144 p. 1 – 13 [Accessed 27th November 2014]

Grissom, R., Kim, J. (2012) Effect Sizes for Research. Univariate and Multivariate Applications (2nd Edition) Routledge: New York

Harlow, H. (1958) The nature of love. American Psychologist. [online] 13(12) p.673-685 [Accessed 26thJanuary 2015]

Hawkins, D. (2009) Biomeasurements: A students guide to biological statistics (2nd ed) Oxford University Press: Oxford

Iki, T., Ahrens, F., Pasche, K., Bartels, A., Erhard, M. (2011) Relationships between scores if the feline temperament profile and behavioural and adrenocortical responses to a mild stressor in cats. Applied Animal Behavioural Science [online] 132(1-2) p. 71-80 [Accessed 21st February 2015]

Konok, V., Doka, A., Miklosi, A. (2011) The behavior of the domestic dog (Canis familiaris) during separation from and reunion with the owner: A questionnaire and an experimental study. Applied Animal Behaviour Science [online] 135(4) p. 3000-308 [Accessed 10th March 2015]

Landsberg, G., Denenberg, S., Araujo, J. (2010) Cognitive dysfunction in cats: A syndrome we used to dismiss as ‘old age’. Journal of Feline Medicine & Surgery [online] 12(11) p. 837-848 [Accessed 11st February 2015]

Lowe, S., Bradshaw, J. (2001) Ontogeny of individuality in the domestic cat in the home environment. Animal Behaviour [online] 61, p. 231–237 [Accessed 17th January 2015]

Mariti, C., Ricci, E., Carlone, B., Moore, J., Sighieri, C., Gazzano, A. (2013) Dog attachment to man: a comparison between pet and working dogs. Journal of Veterinary Behaviour [online] 8 p. 135-145 [Accessed 29th February 2015]

Martin, P., Bateson, P. (2007) Measuring behaviour. Cambridge: Cambridge University Press

McCrae, R., Costa, P. (2004) A contemplated revision of the NEO Five-Factor Inventory. Personality and Individual Differences [online] 36(3) pp.587-596 [Accessed 25 February 2015]

Mitani, J (2009) Male chimpanzees form enduring and equitable social bonds. Animal Behaviour [online] 77 p. 633-640 [Accessed 27th February 2015]

Noftle, E., Shaver, P. (2006) Attachment dimensions and the big Five personality traits: Associations and comparative ability to predict relationship quality. Journal of Research in Personality [online] 40 p. 179-208 [Accessed 27th February 2015]

Parthasarathy, V., Crowell-Davis, S. (2006) Relationship between attachment to owners and separation anxiety in pet dogs. Journal of Veterinary Behaviour [online] 1 p. 109-120 [Accessed 29st December 2014]

Perry, H. (2004) Statistics explained. Susses: Routledge

Pike, T., Samantha, M., Lindstrom, J., Royle, N. (2008) Behavioural phenotype affects social interactions in an animal network. Proceedings B [online] 282(1804) p. 2515 – 2520 [Accessed 27th February 2015]

Quinn, J., Cole, E., Bates, J., Payne, R., Cresswell, W. (2009) Personality predicts individual responsiveness to the risks of starvation and predation. The Royal Society B [online] 78 p.  1203–1215 [Accessed 27th February 2015]

Raihani, G., Rodriguez, A., Saldana, A., Guarnernos, M., Hudson, R. (2014) A proposal for assessing individual differences in behaviour during early development in the domestic cat. Applied Animal Behaviour Science. [online] 145 p.48-56 [Accessed 2nd December 2015]

Ramos, D., Reche-Junior, A., Fragoso, P., Plame, R., Yanasse, N., Gouvea, V., Beck, A., Mills, D.(2013) Are cats (Felis catus) from multi-cat households more stressed? Evidence from assessment of fecal glucocorticoid metabolite analysis. Psychology & Behaviour [online] 122(2) pp. 71-75 [Accessed 21 February 2015]

Rochlitz, I. (2005) A review of the housing requirements of domestic cats. Applied Animal Behavioural Science [online] 93(1-2) p. 97-109 [Accessed 17th December 2014]

Rodan, I. (2010) Understanding feline behaviour and applications for appropriate handling and management. Topics in Companion Animal Medicine [online] 25(4) p.178-188 109 [Accessed 17th December 2014]

Rodan, I., Sundahl, E., Carney, H., Gagnon, A., Landsberg, G., Seksel, K., Yin, S. (2011) AAFP and ISFM Feline-Friendly Handling Guidlines. Journal of Feline Medicine and Surgery [online] 13 p. 364-375 [Accessed 10th March 2015]

Rollin, B. (2007) Ethical issues in geriatric feline medicine. Journal of Feline Medicine and Surgery. [online] 9 p. 326-334 [Accessed 11th March 2015]

Saito, A Shinozuka, K (2013) Vocal recognition of owners by domestic cats (felis catus). Animal Cognition. [online] 16 p. 685-690 [Accessed 08th March 2015]

Schotz, S., Eklund, R. (2011) A comparative acoustic analysis of purring in four cats. Speech, Music and Hearing [online] 51 p. 9-12 [Accessed 08th March 2015]

Schwartz, S. (2002) Separation anxiety syndrome in cats: 136 cases (1991-2000) Journal of the American Veterinary Medical Association. [online] 220(7) p.1028-1033 109 [Accessed 17th January 2015]

Schwartz, S. (2003) Separation anxiety syndrome in dogs and cats. Journal of the American Veterinary Medical Association. [online] 222(11) p. 1526-1532 [Accessed 17th January 2015]

Serpell, J. (1996) Evidence for an association between pet behaviour and owner attachment levels. Applied Animal Behavioural Science [online] 47(1-2) p. 49-60 [Accessed 27th January 2015]

Serpell, J. (2014) Canine Behavioural assessment and Research Questionnaire. Available from:  http://www.appliedanimalbehaviour.com/cms/ attachment/2003012089/2011582582/mmc1.pdf [Accessed 19th December 2014]

Sherman, B., Mills, D. (2008) Canine anxieties and phobias: an update on seperation anxiety and noise aversions. Veterinary Clinics of North America: Small Animal Practice [online] 38:5 p.1081-1106 [Accessed 03rd March 2015]

Sherman, B., Landsberg, G., Reisner, I., Ciribassi, J., Horwitz., Houpt, D., Kroll, T. (2007) Effects of Reconcile (Fluoxetine) Chewable Tablets Plus Behavior Management for Canine Separation Anxiety. Veterinary Therapeutics [online] 8(1) p. 18-31   [Accessed 12th March 2015]

Siegford, J., Walshaw, S., Brunner, P., Zanella, A., (2003) Validation of a temperament test for domestic cats. Anthrozoos [online] 16, p. 332–351 [Accessed 29th November 2014]

Silk, J., Altmann, J., Alberts, S. (2005) Social relationships among adult female baboons (papio cynocephalus) Variation in the strength of social bonds. Behavioural Ecology Sociobiology [online] 61 p. 183-195 [Accessed 27th February 2015]

Storengen, L., Boge, S., Strom, S., Lingaas, F. (2014) A descriptive study of 215 dogs diagnosed with separation anxiety. Applied Animal Behaviour Science [online] 159 p.82-89 [Accessed 28th February 2015]

Turner, D., Bateson, P. (2014) The domestic cat: The biology of its behaviour. 3rd ed. Cambridge university press: New York

Wedl, M., Bauer, B., Gracey, D., Grabmayer, C., Spielauer, E., Day, J., Kotrschal, K. (2011) Factors influencing the temporal patterns of dyadic behaviours and interactions between domestic cats and their owner. Behavioural Processes [online] 86(1) p. 58-67 [Accessed 26th November  2014]

Wemelsfelder, F (2007) How animals communicate quality of life: the qualitative assessment of behaviour. Animal Welfare [online] 16(5) p. 25-31 [Accessed 12th  March 2015]

Winefield, H., Black, A, Chur-Hansen, A. (2008) Health effects of ownership of and attachment to companion animal in an older population. International Journal of Behavioural Medicine [online] 15 p. 303-310 [Accessed 29th January 2015]

Yeon, S., Kim, Y., Park, S., Lee, S., Lee, S., Suh, E., Hout, K., Chang, H, Lee, H., Yang, B, Lee, H. (2011) Differences between vocalisation evoked by social timuli in feral cats and house cats. Behavioural Processes [online] 87(2) p. 183-189 [Accessed 08th March 2015]