Home     Conference     Discussion     Online resources     Consortium     Credits    
 You are here  Conference Section  You are here  Workshop Series  You are here  Animal Welfare
 You are here  Natural Behaviour and Behavioural Needs of Farm Animals

Workshop Series
Virtual Conference
Current Discussion
World Food Supply
Production Siting
Quality and Safety
The Environment
Animal Welfare
Animal Health
Genetic Resources
Animal Nutrition
Global Trade
Basic Paper
Stress and Welfare
Pain in Farm Animals
Immunobiological Explanation...
Communication of the Welfare Status...
Natural Behaviour...
The Welfare of the High Producing Animal
An Evaluation of ‘Indexing’ Welfare...
Animal Welfare during Transport...
Biotechnology of Reproduction...
Does Present Legislation Help...
Farm Animal Welfare in an Economic...
Natural Behaviour and Behavioural Needs of Farm Animals



Per Jensen
Swedish University of Agricultural Sciences,
Department of Animal Environment and Health,
Section of Ethology,
POB 234, SE-532 23 Skara, Sweden
Evolution of natural behaviour

Evolution has selected animals for increased fitness in their natural environments over thousands of generations. Fitness in this context can roughly be defined as reproductive success (often measured at allelic level), and is not necessarily related to welfare; an animal with high fitness may have a poorer state of welfare than one with low fitness. One of the major achievements of behavioural research over the last decades is the realisation that animals are equipped with flexible behavioural programs, usually referred to as strategies (Krebs and Davies, 1991). A strategy involves a number of different possible behavioural outputs and a decision mechanism that helps the animal in selecting the best behaviour under different conditions. The best strategy in this context is the strategy that gives the largest possible fitness for the animal. For example, one specific social situation may require that the animal avoids conflict and assumes a subdominant role, while in a different situation, the same animal may be better off seeking conflict and attempting to become dominant. The best strategy will be the one that allows the animal to make the choice which will render the largest fitness in any group. Within a population of animals, different individuals may therefore show different behaviour in a similar situation, even though they use the same strategy. The trait that is selected by evolution is net benefit (in fitness units, but often approximated by, for example, energy), so animals are expected to assess the costs and benefits of each possible behaviour and then select the one that provides the largest net benefit under prevailing conditions.

Effects of domestication

During domestication, animals have been genetically changed in relation to their ancestors, which has affected various aspects of their adaptive traits, including behaviour (Price, 1997). However, as is obvious from many comparisons between domestic animals and their ancestors, behaviour is usually affected in a very limited manner. Generally, no new behaviours have developed and none have completely disappeared from the gene pool. The changes are usually referred to as altered release thresholds (Price, 1997), but a better description may be to say that the strategies have been modified. Since domestication provides animals with food and protection, and involves selection for specific traits, the optimal behaviour is likely to be different compared to when food is scarce and predation pressure high. Other strategies than in the wild may therefore give the highest fitness. We have found that domestic pigs and poultry use slightly different strategies than their wild counterparts, wild boars and jungle fowl, during food search, and the differences appear to reflect an adaptation of strategies to the conditions offered during domestication (Schütz and Jensen, 1999; Gustafsson et al., 1999a; Andersson et al., 2000). However, when comparing maternal behaviour betwen domestic pigs and wild boars, almost no differences were detected, which indicates that some fundamental behaviour patterns are virtually unaffected by domestication (Gustafsson et al., 1999b).

It is therefore clear that domestic animals still possess natural behaviour and motivational systems that are inherited from the ancestors, and have been little changed by domestication. Any understanding of how these systems function needs to be based on the assumption that they were developed to work in a completely different environment, the environment of evolutionary adaptation (EEA). Animals will therefore perceive stimuli and attempt to react on those in accordance the best strategy in the EEA, or sometimes with a slightly modified strategy.


Whereas motivation is a common concept in ethology and psychology, there is no universally accepted definition. A recent text-book describes the concept as follows: "The internal state of the animal, which is the net result of stimuli arising from both inside and outside its body, constitutes its 'motivation'" (Manning and Dawkins, 1998). As pointed out by Jensen and Toates (1997), the interesting aspects of motivation go on in the brain, and what we really seek to understand in motivational research are the processes by which an animal integrates information from different stimulus sources, both from inside and outside the body, and translates that into behaviour. So motivation is concerned with the state of the nervous system.

Different models of motivation have different merits. Elements of homeostatic motivational models and "drive"-based models both have to be taken into account in any analysis of a specific behaviour. Based on that, Jensen and Toates (1997) offered a simple model which outlines the possible pathways whereby motivational states (tendencies to engage in particular behaviour) could be affected in the brain, and which therefore have to be considered in any analysis of a specific behaviour pattern (Fig 1). Unlike many other models, this one does not claim that every motivational system works in this way; rather, it offers an account of the different pathways according to which any motivational system could work. It is therefore a conceptual model, offering a tool for specific analyses, and not a predictive model.

Figure 1. A conceptual motivational model showing the different causal and feedback pathways that need to be considered in a motivational analysis. The evolutionary processes are not necessarily accessible to the animals and the functional significance of a behaviour therefore has no immediate effect on the causal events. However, the whole causal system is shaped by evolution.

Behavioural needs

The model offers some insights into the needs of animals with respect to their behaviour. In particular, the following two aspects of the model may be of interest: (1) There is no qualitative difference to an animal whether a particular motivational state is evoked by external or internal stimuli. In fact, both will usually be the case (Toates and Jensen, 1991), and the state of the nervous system is likely to be the same in both cases. (2) Sometimes, the animal may be programmed to perform a specific motor program before it can detect any feedback effect on the stimuli which have aroused the motivation. An animal feeling hunger may be programmed to forage, ingest, chew and swallow before it can detect that the gut is filling up; attempts to shortcut such pathways by tubular feeding has sometimes proven unsuccessful in reducing the feeding motivation (see Toates and Jensen, 1991).

The concept of behavioural needs is often used to describe a behaviour which is normally released predominantly by the action of internal stimuli (see Jensen and Toates, 1993 for an overview). However, since the motivational state of an animal is not affected whether it has been aroused by internal or external factors, the needs of the animal are consequently not related to the source of stimuli. It is therefore difficult to split the behavioural repertoire of an animal into some that constitute needs, and others that do not. However, a close understanding of the motivational system governing a particular behaviour will be helpful in specifying the needs of the animal with respect to that behaviour.

Nest building in sows

As an example, consider a case which has been in the focus of the debate regarding the needs of farrowing sows. Under free-range conditions, sows build farrowing nests during the last day before giving birth (Jensen et al., 1993). This is partly triggered by a preparturient rise in prolactin (Lawrence et al., 1994), and sows in all housing conditions attempt to carry out nest building activities during this period (Jensen, 1993). It has been argued that sows will not be motivated to nest build if they are not stimulated by external stimuli, and that they may be unmotivated if they have a sufficiently comfortable lying place or a ready-made nest (i e are exposed to stimuli associated with the goal). This has been strongly refuted by experimental evidence (Haskell and Hutson, 1994; Hutson and Haskell, 1990; Jensen, 1993). It has also been suggested that it may be sufficient for the sows to carry out the motor patterns of nest building, even if this does not result in a nest. This would be possible for sows that are tethered or confined in stalls, and they may therefore not experience any stress in this situation. However, sows in tethered conditions have considerably higher cortisol levels prior to farrowing than loose sows with straw (Lawrence et al., 1994), which strongly suggests that carrying out the motions is in itself not sufficient to reduce the nest building motivation.

In this particular case, the behavioural needs of sows during the last day before farrowing appear to consist of a combination of being able to carry out the motion patterns involved in nest building, but also to receive feedback from the activities. The importance of feedback is also evident from the fact that under free-ranging conditions, sows will adjust their behaviour to the prevailing condition, for example collect more nest material in harsher climate (Jensen, 1989).

Why do animals perform "unnecessary" behaviour?

The example of the nestbuilding sow strikes some people as unlogical. If a sow is provided with a protected, thermally optimal farrowing site, such as a farrowing pen, the evolutionary appropriate action would appear to be to save energy and refrain from activity. However, in the EEA, sows were unlikely to ever encounter such conditions, and the motivational system is designed to work in that context. In addition, pre-farrowing nest building is an essential behaviour which is only performed a few times in the total life-time of a sow, which may explain why the behaviour is largely hard-wired. Whereas behaviour like this may appear "unnecessary" under farming conditions, it certainly is easy to understand from the perspective of the EEA.

A second source of apparently "unnecessary" behaviour is ignorance of the function of a particular behaviour. This may be the case for a type of behaviour termed "contrafreeloading". It was discovered already in the 1960:s that animals offered a choice between eating free food and working for food by pressing levers would often consume the highest proportion of its food from the source which required effort (Jensen, 1963). Again, this was surprising, since it would appear most evolutionary logical to save energy when free food is available. It was therefore interpreted as a sign of a behavioural need in the sense that foraging and working for food was hard-wired behaviour which an animal would always attempt to express (Gardner and Gardner, 1988). However, studies of free living wild starlings feeding from food sources requiring different amount of work, led to another, more plausible explanation: Animals work for food in order to obtain information about possible alternative food sources (Inglis and Ferguson, 1986). This has rendered experimental support in studies of gerbils (Forkman, 1993).

We studied the behaviour of semi-naturally kept jungle fowl and White Leghorn laying hens in enclosures where they could feed freely available food or food that was mixed with sawdust, and therefore required that the hens worked for it. Whereas jungle fowl followed the contrafreeloading pattern and obtained most of its food from the source requiring work, the laying hens did the opposite (Schütz and Jensen, 1999). We interpreted this as a domestication induced change in the foraging strategy of the birds. Whereas it may pay off to a wild omnivore like a jungle fowl to use energy to gather information about alternative food sources, laying hens have been selected to use their energy mainly for growth and egg production, and this selection has occurred under conditions where food has been abundant. This may have caused the strategy of the laying hens to be modified towards relying more on safe food supply.


The two examples above show that we need to be careful in interpreting the behavioural needs of domestic animals. Whereas the behaviour may sometimes be hard-wired and the animal therefore motivated to perform the motion patterns regardless of environment, it could also be that the animals are searching for specific goals which we are not aware of. Any conclusions about the needs of animals with respect to their behaviour therefore has to be based on a thorough functional and motivational analysis.

The major source of information about the needs of animals is therefore their natural behaviour. Although it is not certain that animals have a need to perform every behaviour they show in a natural surrounding, their behaviour under such conditions will always reflect some sort of need. By careful studies, we can examine the relevant stimuli and feedback pathways associated with each behaviour and thereby assess the needs of the animals.


Andersson M, Jensen P, Nordin E (2000) Domestication effects on foraging strategies in fowl (Gallus gallus). In press: Applied Animal Behaviour Science.
Forkman B (1993) The gathering and use of information in foraging. Doctoral thesis: Department of Zoology . Stockholm: University of Stockholm.
Gardner RA, Gardner BT (1988) Feedforward versus feedbackward: An ethological alternative to the law of effect. Behaviour and Brain Sciences, 11, 429-493.
Gustafsson M, Jensen P, de Jonge F, Schuurman T (1999a) Domestication effects on foraging strategies in pigs (Sus scrofa). Applied Animal Behaviour Science, 62(4), 305-317.
Gustafsson M, Jensen P, de Jonge FH, Illmann G, Spinka M (1999b) Maternal behaviour of domestic pigs and crosses between domestic pig and wild boar. Applied Animal Behaviour Science, 65, 29-42.
Haskell MJ, Hutson GD (1994) Pre-farrowing behaviours of sows and gilts with access to space for locomotion. Australian Journal of Experimental Agriculture, 34, 1099-1105.
Hutson GD, Haskell MJ (1990) The behavior of farrowing sows with free and operant access to an earth floor. Applied Animal Behaviour Science, 26, 363-372.
Inglis IR, Ferguson NJK (1986) Starlings search for food rather than eat freely available food. Animal Behaviour, 34, 614-616.
Jensen ED (1963) Preference for bar pressing over "free-loading" as a function of unrewarded presses. Journal of Experimental Psychology, 65, 451-454.
Jensen P (1989) Nest site choice and nest building of free-ranging domestic pigs due to farrow. Applied Animal Behaviour Science, 22, 13-21.
Jensen P (1993) Nest building in domestic sows: the role of external stimuli. Animal Behaviour, 45, 351-358.
Jensen P, Toates FM (1993) Who needs 'behavioural needs'? Motivational aspects of the needs of animals. Applied Animal Behaviour Science, 37, 161-181.
Jensen P, Toates FM (1997) Stress as a state of motivational systems. Applied Animal Behaviour Science, 54, 235-243.
Jensen P, Vestergaard K, Algers B (1993) Nestbuilding in free-ranging domestic sows. Applied Animal Behaviour Science, 38, 245-255.
Krebs JR, Davies DB (1991) Behavioural ecology: an evolutionary approach. Oxford: Blackwell Scientific Publications.
Lawrence AB, Petherick JC, McLean KA, Deans LA, Chirnside J, Vaughan A, Clutton E, Terlouw EMC (1994) The effect of environment on behaviour, plasma cortisol and prolactin in parturient sows. Applied Animal Behaviour Science, 39, 313-330.
Manning A, Dawkins MS (1998) An Introduction to Animal Behaviour. Cambridge: Cambridge University Press.
Price EO (1997) Behavioural genetics and the process of animal domestication. In: Genetics and the behaviour of domestic animals (Ed. by Grandin, T.), pp. 31-65: Academic Press.
Schütz K, Jensen P (1999) Foraging behaviour and activity in red junglefowl (Gallus gallus) and in domesticated breeds. In: 33rd International Congress of the International Society for Applied Ethology (Ed. by Böe, K. E., Bakken, M. and O, B. B.), pp. 92. Lillehammer, Norway: NLH, Agricultural University of Norway.
Toates F, Jensen P (1991) Ethological and psychological models of motivation - towards a synthesis. In: International conference on simulation of adaptive behaviour (Ed. by Meyer, J. A. and Wilson, S. W.): MIT Press: Cambridge.

Top Print version Sitemap Copyright, Legal Disclaimer Contact