The Pros And Cons Of Animal Welfare

Animal welfare assessment has been based on the five freedoms1 concept. This considered negative aspects of mental state (fear and stress) and the compromise of physical domains (nutrition, environment, health, behavior), but did not take into account any positive states [1]. Positive welfare has been gaining importance in science [2,3,4] and among public opinion in the past years [5]. As a result, the five freedoms definition started to change and included positive experiences or emotions such as satiety, vitality, reward, contentment, curiosity and playfulness as part of animal welfare [1].

As public concern about animal welfare has increased in the last few years [5,6,7,8], NGOs in USA have started to increase their interest in the inclusion

of positive welfare indicators in extensive pig production systems. The main reason why they might want to introduce positive indicators in animal welfare schemes is because it complements the five freedoms1 approach, hence, it might help to improve animal welfare since positive welfare indicators will mostly be observed and measured when basic needs are satisfied, namely, the five freedoms are met [1].
One possible way to assess positive welfare is through play behavior, which is defined by Burghardt [9] as a list of five criteria: ’it is not fully functional in terms of current survival, it is considered to be autotelic (self-rewarding), voluntary and spontaneous; it differs from serious performance of adults either structurally or temporally, is performed ‘repeatedly’, but not stereotypically and is initiated when the animal is in a good welfare state (e.g.

well fed, safe and healthy)’.

The last criterion might be the reason why NGOs are interested in promoting play behavior as an indicator of positive welfare. Play behavior is more likely to be expressed when animals are healthy and well-fed, and less likely when they are subjected to stressful conditions [9,10]. For instance, play behavior is observed less in malnourished piglets [11,12] and also in bad weather conditions [13]. Play behavior may indicate a state of good welfare, because once the primary needs of the animals are satisfied, the motivation for playing might be higher.

In order to state that play behavior is a good indicator of positive animal welfare the relationship between both has to be robust, behavior assessment has to be practically feasible and wider societal reasons for using play behavior to indicate and promote positive welfare have to be compelling. These criteria are discussed in the following paragraphs and scored in Table

1.
It is clear that play behavior, in order to be a robust indicator of animal welfare, should logically be expressed always and only when piglets have good welfare or never when there is poor welfare, and should never be increased because of other factors different than good welfare. These statements and score given on Table 1 are discussed briefly in relation to pig behavior in the following paragraphs.
Play behavior is expressed when pigs are in a good state of welfare [15,16] and its frequency decreases when pig are under poor welfare [15]. Lower frequency is observed in pigs under social stress, for instance, there is a reduction of locomotor play behavior when piglets are mixed with unfamiliar individuals [17], isolated from conspecifics [18,19] and from their mother at early stages [16]. Play behavior is also reduced when space allowance is small and barren environment is provided. Piglets housed in smaller crates play less than those reared in bigger ones [15,20,21,22]. In fact, they increase locomotor and social play once more space is offered [20]. Frequency of play behavior in pigs also decreases when environmental enrichment is poor [21,23]. Pigs on unbalanced diets show less play behavior [11,12] and once balanced diet is provided they increase play behavior again [11].
To consider play behavior as a good indicator of animal welfare, no other factors except good welfare states should increase or alter it. Scientific evidence on factors altering play behavior in pigs different than welfare state is poor. Dudink et al. [24] showed that food expectation generates an increase in play behavior in pigs with high levels of play markers (hopping, scampering, pivoting, pawing, flopping, and head tossing), when food was going to be provided. Also Blackshaw et al. [25] demonstrated that the shape not the size of the pen can alter play behavior characteristics.
In conclusion, even though higher levels of play behavior is more frequent in pigs with good welfare, we can still observe it among pigs on poor welfare state but in lower rates. However it is possible to compare between different groups of pigs as well as use this indicator along with others such as health state, diet, environmental enrichment. We also have to be careful about assessment results when using play behavior as an indicator since there is poor scientific evidence on other factors that might affect the expression of it. Although more research can be carried out, in this moment, play behavior and animal welfare have not a completely robust relationship. Hence, score for this criteria is seven (Table 1).

2.

It is practically feasible to introduce the scoring of play behavior into welfare assessment schemes.

It is clear that the introduction of play behavior into animal welfare assessment schemes is practically feasible only if play markers are known and are easy to be differentiated from other type of behavior, if type and frequency of play behavior is equal among different production stages, and if play behavior assessment is easy to be learnt by inexperienced observers and is a reliable method. This statements and the given scores (Table 1) are discussed briefly in terms of pig behavior in the following paragraphs.

The main play markers in pigs are hop, scamper, pivot, toss head, shake object and carry object as described by Newberry et al. [13], and it extends across all play behaviour categories, namely locomotor, social and object behavior [12,13]. However play behavior and markers are relatively easy to be recognized [14], sometimes it might be difficult to tell the difference between to play behavior, such as playful and mild fight [13].

Play behavior is more frequent and more likely to be observed in piglets between two and six weeks old and then it starts to decline [12,13]. Hence, play might be less present in

adults.
Even though evidence on reliability and easy learning do not exist, we can say that animal welfare assessment through recognition of play markers may be easy to learn and recognize by inexperienced observers and this method might be reliable, because even inexperienced observers are able to recognize and agree on emotional states in farm animals [26] as well as on the level of pain a rat might be feeling through behavioral markers [27]. Taking these into consideration and the fact that play is intuitively easy to recognize since it is an obviously pleasurable behavior [14] and markers are defined, people should be able to assess positive welfare using play as an indicator.
According to the exposed evidence, play behavior is likely to be a good welfare indicator only in piglets, as the frequency is higher here than in other production stages. However play markers are relatively well-defined, some difficulties can exists to differentiate playful fighting from a mild fight, which could affect welfare assessment. There is also no study on the reliability in the use of play behavior as an indicator and how easy it is to be learnt and applied in real life. Because of these reasons, in this moment the feasibility of the introduction of play into animal welfare scheme is low (Table 1), mostly because of the application is narrow and there is no study that shows reliability.

3. The wider societal reasons for using play behaviour to indicate and promote positive welfare are compelling.
On an animal rights point of view [28], we, as human beings, should take care of animals under our responsibility and provide them with a good quality of life as they are considered to be sentient beings, that are able to feel emotions and learn [29,30,31,32]. This point of view might be earning more and more followers in the last years as the amount of Americans following a vegan diet has increased [6,7,8] and people are willing to pay for some improvements in pig production systems such as space allowance, environmental enrichment and research into improved pig housing design [5]. This is an indirect measure of an increasing public demand that animals have a good quality of life.
So, the introduction of play behavior into animal welfare scheme might have a beneficial impact on society as they care about animals [5,6] and, so, they will have the feeling that animals are having a good life, and people in charge of animals might have positive experiences as it might be more rewarding for them than being penalized because of poor welfare conditions [3].
In conclusion, from the perspective of society, the inclusion of play behavior as an indicator of positive welfare in pigs might have a positive impact as public concern on animal welfare has increased and might make possible the development of this methodology. Because of that, this criteria has been scored with nine (Table 1).
CONCLUSION
Animal behavior can be a good indicator of good welfare in piglets only because it is more frequent at this stage and relationship between good welfare and play is strong enough as its frequency is higher when primary needs are satisfied than in poor welfare states. In addition, it may have a positive impact on society as public concern on animal welfare has increased and the inclusion of play behavior in animal welfare scheme might generate the feeling that animals are being well cared for.
In order to have accurate results we have to take into consideration that play is also expressed in animals with poorer welfare states and it might be difficult to differentiate between some types of behavior. However this can be solved in short term by using this method along with others such as health state, diet, environmental enrichment, and we can compare play behavior occurrence between different groups of pigs.
Even though some issues can be solved in a short period of time, there are still some that might be need a longer time period to be solved. More research must be done on factors other than good welfare that might affect the expression of play behavior, on the reliability of the use of play behavior as an indicator and on how easy it is to be learnt and applied in real life.
In conclusion, play behavior is a good indicator of positive welfare in pigs as there is a relationship between them. However, I would recommend that it should be developed before implementation in order to reduce errors.