I.N.R.A., Recherches Avicoles,
Centre de Tours, Nouzilly,
France
Increased complexity is believed to improve welfare since it has been shown to increase behavioural repertoire and activity. It is also believed to improve the use of space and to reduce fearfulness. The complexity of the rearing environment can be enhanced for broilers by giving access to outdoor areas or to extra-furniture, or by manipulating the light or the diet. However the ways the different systems used are evaluated must be taken into account before determining their real value for improvement of animal welfare. Many studies are still required to test new devices and new ways of making the environment more complex, especially in commercial conditions.
Introduction
The importance of increasing environmental complexity to improve rearing conditions for farm animals has been stressed for several years, not only for mammals but also for birds (Newberry 1995, Mench 1998). However, most enrichment devices for birds have been studied in laying hens in order to reduce feather pecking, whereas there have been fewer studies involving broiler chickens.
Increased complexity is believed to improve welfare since it has been shown to modify behavioural repertoire. Increasing the complexity of the areas that are little used by broilers improves the use of these areas. This reduces local crowding that may indeed have deleterious effects on animal health and comfort. It is also believed to increase physical activity in meat-type birds reared indoors and thus to reduce leg problems (see Leterrier et al. 2001a, for review). Increasing complexity in the rearing room also reduces fearfulness (Broom 1969, Jones 1982, Gvarayahu et al. 1989) but this effect is not consistent and the duration of tonic immobility was not shortened in chickens reared in enriched pens in some experiments (Bizeray et al. 2002). It also modifies the consequences of transportation: broilers reared with objects and intermittent background music have shorter duration of tonic immobility after transportation than chickens reared in non-enriched room (Nicol 1992).
Giving chickens access to extra furniture is the most widespread method of increasing the complexity of the environment. In slow-growing chickens the simplest way of increasing complexity is to "open the doors" of the rearing room and give broilers access to outdoor areas. Chickens in free-range systems spend a long time pecking at the ground (Lubac and Mirabito 2000). They mostly use the part of the pasture near the rearing building (Weeks et al. 1994; Mirabito and Lubac 2000). The use of outside areas depends on the vegetation and is considerably enhanced by trees compared to areas with only grass since chickens seem to prefer areas with shadow for rest in sunny regions (Mirabito et al. 2002).
In indoor systems the use of visual cover also enhances the use of space. Chickens use more areas with rigid vertical panels than areas free of such equipment and have increased resting and preening behaviours in these areas.
In a study achieved by Cornetto and Estevez (2001) the chickens stayed near the vertical structures even when they only consisted of frames without solid panels. The use of space can also be improved by enriching pens with sand trays (Arnould et al. 2001) or by providing extra areas with sand (Knierim 2001). However, the provision of sand trays did not enhance the time spent standing and did not reduce the occurrence of tarsal angulations but it increased exploratory behaviour during the first weeks of life (Leterrier et al. 2001b).
Perches increase the space available at floor level and allow birds to have a "natural" posture (Hughes and Elson 1977). However, perching rates remained quite low in most studies (Fiscus le Van et al. 2000, Martrenchar et al. 2000, Pettit-Riley and Estevez 2001). Gait score was not improved in flocks bred with perches (Koene et al. 1999; Su et al. 2000) or with barriers allowing perching (Bizeray et al. 2002). The reasons why perches were not effective in preventing leg problems may partly be that they are often poorly used by birds. It seems that perches with easy access such as barriers on the floor are better used (17% at 5 weeks of age, Bizeray et al. 2002) since perching behaviour is impaired by high body weight (Koene et al. 1999) and lameness (Davies and Weeks 1995). The ergonomics and the location of the perches still need to be studied, in view of the fact that the high body weight of broilers may limit their physical capacities.
Many objects have been used to enrich pens: toys, ramps, wooden platforms, boxes filled with peat moss, mirrors, bales of straw (Balog et al. 1997, Newberry 1999, Kells et al. 2001), and so on. These items were expected to stimulate pecking, scratching, climbing, flying or jumping. Most of them have no negative effects on growth and introduce changes in behaviour, and improvements in growth have sometimes been reported (Jones et al. 1980).
However, some devices such as barriers between feeders and drinkers (Bizeray et al. 2002), sand area (Knierim 2001), provision of toys, dustbaths and perches (Tejeda and Galindo 2001) have sometimes been shown to be ineffective in increasing activity. In contrast, quite simple objects such as bales of straw are effective in stimulating exploration, not only in experimental conditions (Newberry 1999) but also in commercial flocks (Kells et al. 2001).
More complex systems combining ramps, dustbaths and platforms seem to be very attractive, even for lame birds (Mench et al. 2001). The effectiveness of the "enrichment" depends a great deal on the novelty of the objects that are offered and thus on their attractiveness (see Newberry 1995 for review). According to Jones et al. (1998), the use of video images may therefore be useful to provide novelty without increasing labour for people in charge of rearing. The effectiveness of the devices also depends on the age at which the animals are observed since bodyweight appears to limit locomotion and general activity in fast-growing chickens when they reach 4-5 weeks (Reiter and Bessei 2001). Even when they are kept under free-range conditions with foster hens, time spent sitting is not reduced at 5 weeks of age compared to birds reared in standard rooms (Knierim 2000).
Modifying light with specific light schedules has significant effects on activity, feed consumption, growth and health problems (Newberry et al. 1988, Sørensen et al. 1999). However few attempts have been made to increase complexity with changes in light intensity. Bizeray et al. (2002) tried to enhance broiler activity with moving spot lights but neither locomotor behaviour was modified nor gait score. Chickens have preferences for different light intensities depending of their age and strain, and resting and perching are associated with dim light (6 lux) (Davis et al. 1999). The effects of light intensity should therefore be further studied since complexity could be increased by using various light intensities during different parts of the light phase.
Diet manipulation may also increase environmental complexity, for example by modifying feeding rhythms or by introducing variety in diet composition. Meal feeding is a simple way to vary access to nutriments. In a study where birds were offered 2 or 3 meals per day, they had fewer lesions of tibial dyschondroplasia and better walking ability (Su et al. 1999). The authors suggested that this beneficial effect may be related to changes in the organization of behaviour. Different diets might be offered during different parts of the day. Sequential feeding - in which chicks are fed - consisting in low lysine diet during half of the day and a normal lysine diet during the other half of the day leads to a significant increase in locomotor activity and improved gait score (Bizeray et al. 2001). However, bodyweight was reduced at slaughter age in this experiment, and other schedules must be tested in order to achieve change in behaviour without impairing growth parameters. This seems possible since substantial changes in behaviour were observed with only slight reduction in growth. Moreover, a sequential schedule combining standard complete diet and whole wheat has been tested in commercial conditions and was shown to increase overall activity in broilers (Noirot et al. 1998).
Conclusion
Fundamental information is still required to establish the relationships between increased environmental complexity and welfare and methodological improvements are needed for this. The parameters used to evaluate the various systems are often different between studies (growth, fearfulness, use of space, exploration or general activity, various health parameters). It would be very useful to standardise the parameters to be taken into account in such studies. It would help to measure the consequences of complexity on chickens' lives and to make it possible to interpret them for animal welfare, which is much more difficult (Kells et al. 2001). The parameters that are mostly used are behaviour patterns. Some special patterns such as exploration are sometimes used as markers. Chronic stress is seldom measured and the quantification of fluctuating asymmetry (Møeller et al. 1995) or of endocrine changes could help in the evaluation. The consequences on health should be better measured because beneficial effects can probably be achieved (general immunity) but the consequences of introducing new objects on the occurrence of parasites and infectious diseases are still unknown.
Furthermore the conditions in which "enrichment" is tested are very important. Most studies are run in experimental rooms and very few have been performed in commercial conditions. More studies should be undertaken in commercial farms to obtain good validation of the devices being tested in order to investigate whether techniques used in experimental conditions will prove to be practical and effective in commercial flocks.
New devices still need to be developed. It is therefore necessary to study the ergonomic features of the objects and furniture that are used to make the rearing environment more complex and to test them with a complete evaluation grid.
References are available on request
From Proceedings of 11th European Poultry Conference, Bremen, Germany.
