Chickens have panoramic vision of about 300°, and possible binocular vision of 26°. Sight is an acute sense and colour vision is important. Preference is probably for colours that are easiest to see against a green background, and this ability to discriminate colour is unlearned.

One-day-old chicks prefer to peck at round rather than angular objects (Goodwin and Hess, 1969). They also prefer to peck small (0.3 cm), solid objects, but would rather approach fairly large ones, especially if moving or making a rhythmical noise.

Selection of food is based on visual cues and immediate taste cues. This is important to know when managing poultry and their food. If the composition of the food changes due to availability of grains, the hens may not change easily to eating seeds of a different shape or colour.

Hearing is an acute sense in chickens, and communication within and among flocks of birds takes place mainly via signals provided by postures, displays and vocalisations (Mench and Keeling, 2001).

Postures and displays are used to signal threat and submission (Kruijt, 1964).

The varieties of vocalisations are in the categories of warning and predator alarm calls; contact calls; territorial calls; laying and nesting calls; mating calls; threat calls; submissive calls; distress, alarm or fear calls; contentment calls; and food calls (Mench and Keeling, 2001).

Morphological features associated with the head and neck are important for both communication and social recognition (Mench and Keeling, 2001). Comb size and colour in males and females are influenced by the levels of sex hormone and are indicators of social status (Guhl and Ortman, 1953).

Little is known about the sense of smell.

There are three common types of husbandry systems used for intensively housed chickens:

Cages. Chickens are kept in groups of 3–10 birds in cages with space allowances of 350-600 sq cm per bird (Mench and Keeling, 2001). Stocking densities vary around the world, 350 sq cm on average in the United States, to as high as 700-800 sq cm in Norway and Switzerland (Savage, 2000).

Meat chicken sheds. These hold from 10,000–70,000 meat birds, housed on litter in either semi-enclosed or environmentally closed houses. Stocking densities vary from 30–50 kg live weight per square metre (Mench and Keeling, 2001).

Breeder sheds. These house flocks of several thousands in semi-enclosed or enclosed housing on litter or wire. The male to female ratio is about 1 to 8–15, with the space allowance of 0.2–0.3 square metre per bird (Mench and Keeling, 2001).

The social organisation differs in these systems but peck orders emerge in cages and breeder sheds. This has not been shown in meat chickens. In cages, there is a definite hierarchy established by pecking and threatening when the hens are placed in the cage, usually a few weeks before laying commences at six months.

The social order in broiler flocks is relatively unimportant as they are generally processed at an age when the establishment of social stratification is just beginning (Siegel, 1984).

Laying hens have complex interrelationships involving social rank, aggression, feeding behaviour and egg production (Mench and Keeling, 2001).

In large groups kept together for some months, subgroups form and become restricted to an area. This means that birds can recognise their own group members and those of an overlapping territory. It was suggested that this territorial behaviour is important in large flocks as it reduces the numbers of conflicts when strangers meet (McBride and Foenander, 1962). It has also been shown that individuals are more dominant in the area where they spend most time. Thus in larger flocks, hens tend to live in neighbourhoods where they are well-acquainted (Craig and Guhl, 1969).

Laying hens choose to feed close to each other when given a choice of feeding locations, which demonstrates the importance of social attraction (Meunier- Salaun and Faure, 1984).

Hens that are in the same cage and in neighbouring cages synchronise their feeding.

Chickens show socially facilitated feeding, in particular, they peck more at feed when they have company than when alone (Keeling and Hurink, 1996).

It has also been shown (McBride et al.,1963) that hens do not move randomly in normal intensive housing conditions—they maintain their heads at regular patterns of spacing and orientate them to avoid the frontal aspects of other birds. However, they turn, probably in  defence, to face approaching birds.

In cages that are too low for the chickens to raise their heads in a threat, aggression is provoked by an approaching bird rather than by a bird that is in continuous close proximity (Hughes and Wood-Gush, 1977).

Recognition of each other is based on features of the head, the comb being the most important cue (Guhl, 1953).

Hens can distinguish between breeds that are dissimilar but are unable to distinguish between individuals of such breeds.

The ability of flock mates to recognise and remember one another becomes very difficult under commercial poultry husbandry conditions where group sizes are very large (Mauldin, 1992).

Dim or coloured lighting can affect a chicken’s ability to discriminate between other birds (Mench and Keeling, 2001).

Mortality, production and behavioural problems are all worse in large groups of hens, which implies the formation of unstable social groups (Mench and Keeling, 2001), so this is particularly a problem in barn/aviary egg-production systems.

Peck orders are regarded as highly stable once established, and in mixed groups, males and females have their own peck order (Guhl, 1958).

Agonistic pecking begins to occur within a few weeks after hatching, stable dominance and subordinate relationships usually do not become established until 6–8 weeks of age in cockerels and 8–10 weeks in pullets (Guhl, 1958).

A potential problem in the industry, depending on spacing and the strain of poultry, is the frequency and severity of agonistic acts. Al-Rawi and Craig (1975) did an interesting experiment, beginning with relatively generous space allowances per hen and then decreasing the space. They found that social interaction rate increased as space decreased then suddenly fell off as space decreased further. It has been shown that individuals behave less aggressively towards subordinates in the near presence of dominant flockmates. This ‘third-party-effect’ (Ylander and Craig, 1980) is associated with a reduction in agonistic behaviouror it may be due to the lack of space for threat displays.

The results of this experiment on spacing indicate that interpretation of the results is important. Another point to be aware of is that selection for productivity traits may cause behavioural changes. Increased aggressiveness and social dominance, prior to full maturity (Bhagwat and Craig, 1977) has accompanied the selection for early onset of egg production in several genetic stocks studied.

Higher-ranking hens may have better egg production than the lowest ranked bird in a cage, possibly because the higher ranked birds have greater access to feed (Cunningham and van Tienhoven, 1983).

Most aggression is seen at the feed trough, where there is some competition among the chickens (Mench and Keeling, 2001). Aggression in cages is relatively low, as the small group size in the cages allows the hens to establish a stable dominance hierarchy (Mench and Keeling, 2001). Once a social group becomes organised, the incidence of agonistic interactions decreases (Mauldin, 1992).

A series of displays occurs before mating, based on a stimulus-response sequence (Fischer, 1975) initiated by the male (see diagram).

Male courtship displays are generally elaborate, involving vocalisations and noises, postures, spreading of the feathers to increase apparent size and emphasise plumage characteristics (Kovach, 1975).

Sexual behaviour and dominance relationships are important in the management of mating. Because the female must crouch to elicit courting behaviour in the male and this is also a submissive behaviour, high-status females are often difficult to mate. Although it is never done commercially, research suggests that to overcome this, chickens may be sub-flocked and this reduces the number of individuals each may dominate or be submissive towards. When high-ranking hens are isolated from hens lower in the peck order, they crouch more often than when in the larger flock, and hens in the middle and lower thirds of the peck order crouched less often (Guhl, 1950).

Maternal behaviour or broodiness has been selected out of commercial laying strains so it is not important in intensive poultry husbandry systems.

In a broody hen with chicks, a bond is formed and the chicks learn to respond to the maternal feeding call, distress call and to the hen’s ‘purring’ sound as she settles down. Repeated exposure to her, accompanied by food, guidance and protection, strengthen the filial bond. Exposure to maternal calls during embryonic development may be important for the development of post-hatch species-specific maternal call recognition (Gottleib, 1976).

Being precocial, birds are self-sufficient after hatching, but parents serve an important protective function while also teaching the chicks about edible and inedible foods (Nicol and Pope, 1996).

Precocial chicks imprint on their parents in the first few days of life (Rodgers, 1995). Imprinted chicks remain close to the imprinted object, which is normally a parent, but under laboratory conditions may be a variety of different objects (Mench and Keeling, 2001).

1. Sometimes males will hound other males, which can be a problem.

2. Caged birds may exhibit some abnormal behaviour such as head flicks and feather pecking, i.e., pecking and pulling the feathers of other birds (Mench and Keeling, 2001). Feather-pecking may be a form of redirected ground pecking (Blokhuis, 1989). Experience in early life with ground pecking may influence pecking behaviour in later life (Blokhuis, 1991). The motivation for the redirection of ground-pecking happens when the incentive value of the ground is low, compared with the incentive value of pecking substrates (Bindara, 1969). In high-density situations, the birds and feathers make up a higher proportion of stimuli relative to the litter area. It is possible that the birds may perceive the feathers  as dust and that may cause a redirection of groundpecking to feather-pecking (Hansen and Braastad, 1994).

3. In some housing systems, cannibalism can be a problem.

4. Pseudo-mating occurs most frequently between highranking males and low-ranking males, who are pursued and trodden (Guhl, 1949) and indicates that dominance relationships are important. The same situation may occur in flocks of hens.

There is ongoing research into alternative husbandry systems, which may in the future replace the cage. Commercial egg layers are housed mostly in cages in groups of 3–10, with much restriction on the bird’s movement. Meat chickens are housed on litter in either semi-enclosed or environmentally closed houses, often in groups up to 70,000 birds.

Very few birds are maintained in free-range conditions as it is uneconomical in terms of labour, food requirements and wastage in egg handling, problems of predators and disease control, and lack of egg hygiene (Sainsbury, 1980). Free-range eggs have a 15–20% rate of dirty eggs that are classed as second-rate eggs, whereas the rate in conventional caging is 2% (Slack- Smith, 2000).

The problem of deciding what is best for a hen is difficult and some researchers have devised experiments in which they provided birds with a choice of flooring (Hughes and Black, 1973; Hughes, 1976) and a choice of environments (Dawkins, 1980; Wegner, 1980).

The hens chose fine hexagonal mesh over coarse rectangular mesh and over perforated steel sheet. It seemed that the hexagonal mesh supported the bird’s foot at more points than the other two floors. In comparing wire and litter floors, it was found that previous experience with either wire or litter floors affected the choice: birds reared on litter spent more time on litter than those raised on wire.

In the environmental preference studies (Dawkins, 1980), hens were given a choice of cage or an outside run. Hens used to living outside in the garden all chose the run. Hens previously used to living in cages tended to choose the cage on first trial, although subsequently they came to choose the run. So choice is strongly influenced by previous experience. The fact that the hens prefer an outside run to a cage is not indicative of suffering in a cage. Preference in itself is no indication of suffering.

Before egg laying, hens will work to gain access to nest sites. The demand for this resource is inelastic. Hens in cages without nests often show abnormal activities during prelaying, such as increased pacing, reduced sitting and displacement behaviours (Sherwin and Nicol, 1993). So, if cages contain a nesting box, there is an opportunity for more normal behaviour. Different systems have been tested for performance and behaviour traits of laying hens since 1977 in Celle, Germany (Wegner, 1980).

Free-range system with deep litter, deep litter without range and battery cages. Egg production has been found to be more economical in cages (i.e., less feed per egg).

Comparison of different cage types with different numbers of hens per cage. There is reason to suppose that some differences exist in behaviour between battery and deep-litter systems, so cage types other than the conventional battery cages are being compared.

There are two types of ‘get-away’ cages developed in England and The Netherlands.

  (i) 80 cm x 100 cm x 65 cm (l x w x d) cage with three perches and two feed troughs at equal levels on both sides of the cage, with 16 and 20 hens, respectively, with four nests per cage. The cages are either filled with wood shavings or the sloping floor that forms the roof of the cage below allows the eggs roll into a collecting tray. Fibre mats are put into the nests to make them more attractive. It is a two-tiered system. This design had problems as the hens used the nests to scratch, sand bath and rest in, so 25–30 per cent of eggs were dirty, destroyed or eaten. Too much labour was required to supply litter daily and collect the eggs.

  (ii) 1 m x 1 m x 55–80 cm (w x d x h) Celle cage with four nests, each with a sloping fibre floor, two feed troughs on both sides and one sand bath. The whole cage floor has a 5° slope from the side of the sand bath to the side of the nests. It is also a two-tiered system.

Egg production in these get-away cages with 15–30 hens per cage of 1 sq m floor area is similar to that in a conventional cage with three or four hens per cage. Some disadvantages of this system were cracked eggs at the floor, which was not elastic enough; and the sand bath, which was opened eight weeks after the start of laying, was immediately used for laying and was forced to be shut. Monitoring of hens was also difficult in this system.

Enriched cages (e.g., the Edinburgh modified cage). These cages are similar to battery cages but provide more features for the chickens, which allows a wider variety of behavioural characteristics. The cage houses four hens and provides them with a perch, a nesting box and a dust bath (Wathes and Charles, 1994).

The aviary and perchery systems. The main idea is to make better use of the space between floor and ceiling in the poultry house by installing several horizontal levels and so increase the number of hens per square metre of floor area. The aviary system used was made of plastic foil and measured 8.50 m wide x 12 m long (10 hens per square metre of floor area).

Better use of height is achieved by having two rows of perches along the house. One feed trough is on the floor and the other is on the perches. There is a ‘farmer automatic nest’ in two parallel rows and in two tiers, which can be moved out of the aviary into a special room where the eggs are automatically removed from the litter. Nipple drinkers provide water. Good results for laying performance and number of losses compared with hens in cages, have been achieved.

The final analysis of results was completed in April 1981, and a report referred to the West German government. Murphy (1982) gives an outline of the final conclusions in her comprehensive report. She states that if all aspects studied are considered, then none of the systems may be described as totally adequate for the overall welfare of the birds.

Cages showed overall advantages in economy and hygiene; there was significantly less social conflict among the birds with a lesser number and intensity of threats, agonistic pecks and associated vocalisations, and fewer deaths from cannibalism than in the floor systems. However, when compared to floor-housed birds, those in cages showed more behaviour indicative of conflict and frustration during nest selection and egglaying behaviour. It has been recommended that further research into completely new husbandry systems or modifications of existing systems be carried out to provide optimal conditions in all aspects of egg production.