Possible improvement of measures within the principle ‘appropriate behaviour’ of the Welfare Quality® boiler assessment protocol

The aim of the current study was two-fold. First, we tested whether or not a scan sampling technique can be used as a relatively quick, simple scoring of the behaviour in commercial-size broiler chicken flocks, and could thus be included as a measure of appropriate behaviour in the Welfare Quality® broiler assessment protocol. Four existing datasets were analysed, these were collected for different purposes but were based on the same sampling technique, i.e. counting the number of broilers engaged in different behaviours using direct observations, at different time intervals and at different locations in the house. Analysis showed that various factors affected the outcome. Whereas those such as observer, time of the day and location could be taken into account when designing an observation protocol (e.g. by training, by selection of different observation locations and times), it must first be determined how to take into account the effect of subsequent scans, the apparent difference between flocks and genetic strains, and the effects of different behaviours. This is important when e.g. thresholds need to be set or scores reliably assigned regarding the prevalence of birds showing certain behaviours during a farm visit. Second, it was tested whether alternative measures (rather than distance to observer or a novel object) could be used as indicators of fear in broiler chickens. Previous research suggested a possible relationship between walking ability and the number of birds within arm’s reach in the touch test, thus implying that the touch test may be confounded by impaired walking ability and might thus be a suboptimal method of assessing fear of humans in broilers. The present study was carried out at an experimental farm and reported in detail in a student thesis. An extended summary is included in this report. The results showed that both distance to observer or novel object and behaviour of the bird (alert body posture, neck posture, body position towards object or observer) seemed to be valid indicators of fear, however, when a relatively short time is available for testing, the birds’ behaviour should always be measured as older broilers need more time to move away.