All cats require environmental enrichment to keep them mentally and physically stimulated. This is particularly true for those housed exclusively or mainly indoors due to their reduced opportunities for exercise and increased risk of boredom. Environmental enrichment refers to the addition of factors into the environment that improve the physical and/or psychological welfare of an animal (for more information on environmental enrichment, see the AAFP and ISFM feline environmental needs guidelines: http://journals.sagepub.com/doi/pdf/10.1177/1098612X13477537). These factors can be as simple as added additional bedding or as complex as cognitively-challenging food puzzles. Enrichment has been shown through scientific studies to reduce stress and abnormal behaviour in a range of species such as dogs, rabbits, rats and baboons, as well as cats.
One type of environmental enrichment for cats is the provision of vertical space. Climbing and using vantage points off the ground to survey their environment are natural behaviours for the territorial domestic cat. Therefore, raised areas allow cats to express these natural behaviours, give cats increased opportunity for seclusion, and are considered to provide cats with a sense of environmental control (eg, enabling them to avoid stressors in their environment). For cats in multi-cat environments, vertical space may also help cats cope with and alleviate any social stress, by allowing cats to separate themselves from each other.
However, something introduced into an animal’s environment, even if it is hypothesised to be beneficial, can only be classed as enrichment if it provides positive consequences (for example allowing an animal to express natural behaviours) and if it improves the welfare of the animal. As such, enrichment items should be tested to assess whether they produce these effects.
A research group from the WALTHAM® Centre for Pet Nutrition in the UK set about to find out whether the provision of vertical space in the form of a ‘screen’ with shelving could be classed as enrichment for their indoor-housed cat colonies, by increasing the cats’ opportunities to climb and get up high, as well as decreasing agonistic behaviours by allowing cats to separate themselves.
Who participated in the study?
The study involved 29 neutered domestic short-haired cats: 16 males and 13 females, housed in four mixed-sex social groups. Each group occupied rooms of equal size (37 square metres).
How was the study carried out?
The cats’ behaviour and space utilisation of the area was recorded for two days before the enrichment object was added.
The enrichment object was then added, in the ‘test phase’. It consisted of a shelving unit with two vertical columns of eight horizontal shelves (hereafter referred to as the ‘screen’). Again, behaviour and space utilisation were measured, this time for four days.
Measurements were then taken for a further two days after the removal of the screen: the ‘removal phase’.
All observations were made from video footage collected from 8am to 4pm. There was only one enrichment object for the entire study, so each of the four rooms went through the eight-day long experiment in turn.
Space utilisation was measured by recording which area of the room the cat was in every 10 minutes, and whether the cat was on the ground or off the ground (on any piece of furniture except the screen), or on the screen (when it was present). Behaviour was measured using an ethogram (a catalogue of different kinds of behaviours including a description of each behaviour). All occurrences of agonistic (hostile) or affiliative (friendly) behaviours were recorded during 30 minute sessions immediately before and after the provision of food, which was provided in the morning and afternoon.
What were the results?
All three phases (‘baseline’, ‘test’ and ‘removal’) were compared to determine the effects of adding and removing the screen on the cats’ behaviour.
When the screen was introduced, the cats were approximately three times as likely to be off the ground compared to when there was no screen present – either before it was introduced or after it was taken away. A significantly larger proportion of cats were found to be spending time on the screen compared to existing shelving when they were off the ground.
Therefore, the results of this study suggest that the screen was a valuable resource to the cats and increased space utilisation by providing increased vertical space within a confined area. The screen was almost twice the height of the wall mounted shelves, suggesting that cats may have valued the screen more than the wall shelves because it was higher, providing a better vantage point. However, an alternative explanation for the increased time using the screen compared with existing shelving may relate to its novelty, which may have increased exploratory behaviour towards it. A further study providing the screen for longer periods of time would help identify the answer.
Affiliative and agonistic behaviours
There were significantly more agonistic interactions between the cats before the morning feed than there were after feeding, regardless of whether the screen was present or not. This suggests that the presence of the screen had no positive effect on reducing agonistic interactions between cats before the morning feed. As the researchers explain, this may have been because of the effective value of the food at morning feeding time, due to an overnight fast, leading to high levels of arousal and so potential tension and agonistic interactions between cats.
Significantly fewer agonistic behaviours were found to occur after the morning feed when the screen was present compared with the baseline and removal phases, suggesting that the screen did have a positive effect in reducing agonistic interactions once feeding had occurred. This could be because the screen provided cats with an increased opportunity to separate themselves from each other and to hide and retreat in response to a threatening situation, and thus avoid hostile interactions, whereas when anticipating food, cats are more likely to come together to the place they are expecting food to be delivered.
Agonistic events occurred significantly more before feeding in the removal phase than in either of the other phases. This suggests that an increase in agonistic behaviours could be related to the loss of a valued resource (the screen). The researchers suggest that this provides rational for providing permanent environmental enrichment, or rolling replacement of enrichment of objects with similar value to the cats.
Cats were found to engage in fewer affiliative (friendly) behaviours after the afternoon feed when the screen was present. This could be because the screen provided the cats with more choice for private resting and post-feeding behaviours such as self-grooming, which the cats may have chosen over interacting with others. However, the overall mean number of affiliative interactions was not significantly different between all three phases of the test, suggesting that affiliation still occurred but at different times when the screen was present.
Are the results reliable?
There are some limitations to the study. First, observations were only made during the day. It would have been useful to record video footage to observe the cats overnight, as well as at dusk and dawn, when cats may be more active. This would give us a more accurate idea of their space utilisation and social interactions. However, it is good that video footage was collected to analyse rather than having people present to observe the cats directly, as the presence of people may have influenced the cats’ behaviour.
It would also have been more useful to observe the cats over a longer period rather than a four-day test period, in order to explore the effects of neophilia (interest in a new object). This would enable us to explore whether the preferential use of the screen over the existing shelves was purely because this was a new object, or because the cats truly preferred this object to the shelves, for example due to its height.
The ethogram used to measure behaviours in the study was not sufficiently detailed, meaning that one person observing the cats may interpret a cat’s behaviour differently to another observer, using the same ethogram. As an example, the behaviour ‘walk with cat’ was defined as ‘a cat travels side by side with another cat often with tails entwined’. The distance between two cats needed in order to categorise them as performing ‘walk with cat’ was not specified, which could lead to different interpretations of a cat’s behaviour (in fact, cats should be almost touching to be classified as performing this kind of social behaviour). There was further ambiguity in the ethogram in terms of some behaviours categorised as agonistic, which could also be observed in social play (in the form of play fighting), an affiliative interaction. These included ‘fight’ – ‘a cat engages in physical contact with another cat, attacking and rolling over’, as well as ‘pounce’, ‘raise paw’ and ‘chase’. The ambiguity in the ethogram reduces the ability for other researchers to reliably repeat this experiment using the same methodology.
Furthermore, the behaviours in the ethogram are treated as if they are equally agonistic or affiliative to each other; however, this is not the case. For example, fighting could be considered more agonistic than blocking a cat. Therefore, we are unable to gauge how agonistic the interactions between cats are in the different test phases, which would give us useful information on the cats’ welfare: a more agonistic interaction is likely to have a greater negative impact on a cat’s welfare than a milder agonistic interaction.
For more detailed, clear and accurate descriptions of cat behaviours, including photos and videos, see iCatCare’s free online resource, created in collaboration with the University of Lincoln and written by iCatCare’s Feline Behaviour Specialist Dr Sarah Ellis, ‘Cat Behaviour Described’: icatcare.org:8080/behaviour-described.
The ethogram also refers to ‘submissive’ behaviour (‘submissive crouch’), and submissive and dominant behaviour are discussed elsewhere in the study. However, cats are not considered to form dominance hierarchies when living in groups. This is because they have evolved from a solitary ancestor, the North African wildcat (see icatcare.org/advice/cat-care/origins-cats for more information on the origins of cats). As such, they have not evolved the complex visual communication system necessary to signal dominance and submission, as is seen in highly social species such as the domestic dog, and their ancestor the wolf.
The choice of the enrichment item used for the study is questionable. The layout of the shelving in the shelving unit means that the high shelves are hard to access, and cats have to climb past other cats to access them, which may be stressful for cats that do not get on with each other. This type of enrichment would also not be suitable for young and old cats, because of its height and difficulty to climb. Adding a ladder may enable older and young cats to access the unit. The shelving unit could also be improved by adding a small barrier (‘lip’) around the top of the unit. Cats may be more likely to sit at the top if it is slightly enclosed, enabling them to look out over their surroundings whilst feeling safe and partially hidden (cats should be able to duck their heads behind the lip). Using different substrates such as bedding, carpet and cardboard on the different shelves may also make the shelving more attractive.
The actual difference in agonistic interactions, although statistically significant, was often very small, and the number of agonistic interactions themselves were also very small. For example, in the comparison between the mean number of agonistic events after the morning feed, the number of agonistic events per cat within a 30 minute time period varied from around 0.04 for the removal phase and 0.01 for the test phase. This highlights the difference between statistical significance – something that a statistical test deems to be a bigger difference than would be expected by chance – and a clinical difference – something that actually makes a difference to a cat’s welfare or health. A reduction from 0.04 to 0.01 agonistic interactions per cat per 30 minutes may have very little actual impact on the cats’ wellbeing.
Frequency measures (counting the number of times each behaviour occurred) were used to measure behaviour. However, for some of the behaviours it would be more useful to record the duration of that behaviour, which would provide more relevant information. For example, alloresting (resting alongside another cat) may only occur once during an observation period, but it may occur for a long period of time, which gives useful information on the relationship between the two cats.
All of these limitations call into question how confident we can be in the conclusions drawn from this study, and highlight why we need to take a critical look at such studies.
What can we take from this?
Despite the limitations we have highlighted, this study has produced some useful findings. It provides evidence that supplying environmental enrichment in the form of vertical space leads to enhanced space utilisation, enabling cats to spread themselves out more in a restricted environment, and that the object was a valued resource to the cats (although this could be due to its novelty). Vertical space is important to cats, allowing them to express their natural behaviour of climbing up to high vantage points. The results suggest that the provision of vertical space enabled cats to remove themselves from each other and increase their welfare by reducing hostile interactions. Cat owners, especially those of indoor-only cats and cats in multi-cat households, should therefore consider getting their cats objects to provide vertical space, such as cat trees, shelving units and wall-mounted shelves, to improve their cats’ wellbeing.
The study also provides support for the provision of permanent enrichment structures, as removal of the enrichment resulted in some hostility, and so potentially reduced welfare, perhaps due to frustration. The authors suggested that rolling replacement of enrichment could be used instead of permanent enrichment. Rolling replacement is good for enrichment items such as toys and puzzle feeders, because these are ‘hunted’ by cats. Prey would be variable and in different locations for each hunt. Therefore, changing toys and puzzle feeders and putting them in different locations would mimic this, and introduce novelty in order to keep a cat’s interest in the enrichment high, promote exploratory behaviour and reduce boredom. However, cats would not expect big changes in their territory and prefer a stable, predictable environment. Therefore, moving large structural enrichment such as a shelving unit may cause a cat distress, and so these should remain as permanent features of the environment.
The study has highlighted that arousal may be high around feeding time in cats fed twice a day, leading to increased agonistic interactions. This finding has especially important implications for multi-cat households, and highlights the importance of a cat’s feeding regimen on their behaviour and welfare. Cats are naturally solitary hunters, just like their wild ancestor. Since they hunt alone, their prey are generally small in size, such as small rodents. The average mouse only contains about 30 kilocalories, meaning cats must hunt, kill and eat around ten mice a day in order to meet their daily energy and nutrient requirements. Therefore, cats have adapted to eat multiple small meals over the course of a 24-hour period; this includes eating during the night, when their nocturnal prey are active. Thus feeding cats just twice a day, and out of a bowl which requires no expression of hunting behaviours to acquire their food, may be causing frustration and boredom. In order to mimic the cat’s natural feeding habit of eating little and often, owners should divide their cats’ daily food ration into several portions (iCatCare recommends a minimum of five), which should be fed throughout the 24-hour period. Feeding during the night can be achieved by using puzzle feeders. Puzzle feeders are objects that hold food and must be manipulated in different ways to release this food. These can be filled and left for cats overnight, as well as during the day if the owner is away at work. Puzzle feeders also encourage mental and physical stimulation of cats during feeding, and allow them to express some of their natural hunting behaviour. For information on how puzzle feeders are beneficial to cats, and how you can make your own puzzle feeders, see https://icatcare.org/sites/default/files/PDF/five-a-day_felix_report_final.pdf. A very welcome side effect of feeding a cat overnight is that you are much less likely to get woken up early in the morning by a hungry cat!
The work of International Cat Care
International Cat Care will continue to provide you with interpretations and summaries of the latest cat science studies – objectively critiqued and intelligently communicated to you by dedicated cat-loving feline scientists, professionals and veterinarians.
Reference for the study:
Desforges EJ, Moesta A and Farnworth MJ. Effect of a shelf-furnished screen on space utilisation and social behaviour of indoor group-housed cats (Felis silvestris catus). Appl Anim Behav Sci 2016; 178: 60–68.