Animal welfare assessment at group level is a scientific discipline that is rapidly developing. The interest in welfare assessment systems is based on an ethical concern for the welfare of farm animals. The scientific community plays an important role in delivering an appropriate repeatable, valid and feasible framework for these assessments. Consideration of the potential applications of these techniques is important for deciding upon the requirements of specific assessment systems. This paper provides a brief overview of the different types of applications, which can be categorised broadly into research, legislative requirements (non-voluntary), certification systems (voluntary) and advisory/management tools. These applications may have various goals: quantification of welfare, provision of welfare assurance or welfare management. Assessment systems vary in many characteristics, such as whether they are animal- or resource-based, and whether they are based on single or integrated scores. Different applications will require different elements of these features.

The notion ‘quality of life’ (QoL) suggests that welfare in animals encompasses more than just an absence of suffering; it concerns the quality of an animal's entire relationship with its environment, of how it lives its life. Judgements of such quality are based on the integration of perceived details of how animals behave over time in different contexts. The scientific status of such judgements has long been ambiguous, but in recent decades has begun to be addressed by animal scientists. This paper starts with a brief review of qualitative approaches to the study of animal behaviour, which tend to address characteristics such as individuality, personality, and emotionality. The question then arises whether such characteristics involve a subjective, experiential aspect, and identify animals as sentient beings. The second half of this paper argues that taking the integrative nature of qualitative judgements seriously enables a ‘whole animal’ perspective, through which it becomes possible to view behaviour as a dynamic, expressive body language that provides a basis for assessing the quality of an animal's experience (eg contented, anxious). Judging this quality is a skill that requires knowledge of species-specific behaviour, experience in observing and interacting with animals in different contexts, and a willingness to communicate with animals as sentient beings. A substantial body of research indicates that this skill can function reliably in a scientific context, and can be applied usefully as a practical welfare assessment tool. Thus qualitative approaches to the study of animal behaviour should make an important contribution to the growing interest in animal QoL.

There is a trend toward increasing intensification in dairy farming in the United Kingdom. In particular, there is concern over systems in which cows are housed throughout the year, as the behavioural restriction implicit in these systems is associated with poor welfare in other species. The aim of this on-going project is to determine how this affects the behaviour and welfare of dairy cattle. A range of behavioural, physical and health measures are being used to assess cow welfare on about 40 commercial British dairy farms. Initially, five farm types were identified from analysis of returns from a farmer questionnaire. Milk production level and housing type were the principal factors explaining variation in farm type. The sample groups are: high-, mid- and low-production cubicle-housing units, mid-production straw-court units, and cubicle-housing high-production zero-grazing units. Observations will take place over three winter housing periods (2000/01 to 2002/03), with recording on each unit taking five days. Our main hypothesis is that the behavioural and physical responses of ‘at-risk’ younger cows provide a sensitive indication of farm-level ‘stress’. Cows are marked according to age, and the feed-face videotaped continuously to record feeding time and social interactions. Temporal organisation of behaviour will be analysed using fractal mathematics and qualitative assessment approaches used in a human interaction test. Cow cleanliness, condition score, response to a novel object and the incidences of lameness and leg injury are recorded, and building quality is assessed. Ultimately, multivariate methods will be used to test our underlying hypothesis and to assess effects of housing and production type on behaviour and welfare. This analysis may identify key objective measures of welfare for use in farm assurance schemes.

Given the absence of a ‘Golden Standard’ for the objective determination of welfare, the collection and interpretation of data involving different parameters is essential for assessing the well-being of farm animals. The choice of parameters and the relative weights assigned to each of them are crucial for the outcome of the assessment. Both elements involve a certain degree of subjectivity. In this paper we discuss the basics of different methods used to integrate welfare parameters, focussing on the issue of scientific objectivity. We begin by addressing parameter selection, the assignment of parameter weightings or rankings and the qualifications necessary for ‘experts’ designing and applying the methodology. Five different approaches to integrating parameters are then discussed. The paper does not state a preference for any method, but aims to encourage discussion of key elements involved with the on-farm assessment of welfare.

The aim of this paper is to discuss the relationship between qualitative categories of sheep welfare and associated quantitative behavioural observations. Most scientific studies rely on quantitative measures, however to interpret those measures in terms of an animal's experience of welfare, the use of qualitative terminologies denoting various emotional states (eg fear, pain, distress) seems hard to avoid. This is especially so in cases where the same behavioural or physiological indicator could have different meanings for welfare; high levels of locomotion, for example, could indicate fear, aggression, or both. To resolve such dilemmas, scientists often resort to judging an animal's most likely state through direct observation; indeed qualitative characterisations are often intertwined with reported quantitative results. This paper reviews frequently used qualitative categories of sheep welfare and the behavioural context in which these categories are applied. Qualitative judgements of welfare tend to habitually be regarded as ‘subjective’, ie as anthropomorphic and unreliable; however from this review it appears that such judgements reflect the overall demeanour of an animal's behaviour, and as such may provide a useful empirical basis for the investigation of emotional expression in sheep. Further development of a qualitative behavioural approach may provide a more complete picture of the expressive repertoire of sheep, and enhance our understanding of ways in which these animals experience well-being or distress.

This paper reviews the debate that is currently taking place in the field of philosophy of mind on different conceptual models of consciousness. More and more philosophers argue that the explanation of subjective phenomena requires two complementary perspectives of understanding, known as the first- and third-person perspectives. The third-person perspective (ie conventional objectivity) accounts for the physical, functional aspects of consciousness, while the first-person perspective addresses the subjective, experiential aspects of consciousness. It is suggested that each of these conceptual perspectives may facilitate a different type of research in the study of animal emotion. Within the conventional, third-person perspective, a growing enthusiasm for issues of animal consciousness has led to sophisticated physical and cognitive models of animal emotion. The potential of the firstperson perspective, however, to provide a basis for models of animal subjective experience has remained largely unexplored. The paper concludes with a brief review of the author's recent experimental work on concepts of animal behavioural expression. The high reliability and repeatability of such concepts indicates that the first-person perspective may provide a valid research perspective in its own right.

Qualitative Behaviour Assessment (QBA) is a methodological approach to assess the whole animal using terms to describe and quantify the emotionally expressive qualities of behaviour and identifying larger patterns of expressivity through multi-variate statistical integration. A key condition for the success of QBA is achieving a common understanding of the meaning of descriptive terms by raters. Based on this, our study aimed to develop a list of terms in Brazilian Portuguese for the QBA of broiler chickens (Gallus gallus domesticus), and to test this list by studying its inter- and intra-rater reliability. Fourteen experts participated in a workshop and developed a list of 25 QBA terms, and 40 undergraduates tested this list by scoring 18 video clips using a 125-mm visual analogue scale. Principal Component Analysis was used to analyse observers’ scores. Principal Component (PC) 1 ranged from disturbed/frustrated to comfortable/lively, suggesting this PC may be interpreted in terms of emotional valence. PC2 ranged from calm/dull to agitated/active, suggesting this PC indicates the level of arousal/energy of the birds. Both PC1 and PC2 clip scores showed good inter- and intra-rater reliability. This study demonstrates the importance of producing QBA term lists bottom-up as opposed to merely translating pre-existing lists from the scientific literature. Results suggest the standardised Portuguese QBA term list developed in this study is reliable in assessing the expressive qualities of broiler behaviour; therefore, a next step is to test it on-farm with experienced raters and further refine it concerning terms related to poor welfare.

This study explores the use of Free Choice Profiling (FCP) methodology for the qualitative behaviour assessment of emotional expression in the domestic dog (Canis familiaris). Qualitative behaviour assessment is based upon the integration of many pieces of information that in conventional quantitative approaches are recorded separately or not at all. Observers are asked to focus on an animal's interaction with its surrounding environment, and to describe the animal's expressive demeanor, or ‘body language’. A specific characteristic of FCP methodology is that it allows observers the freedom to devise their own descriptive terms, and then to use these personal terms to quantitatively score observed subjects. Application of FCP to qualitative behaviour assessment in animals was originally tested for pigs, and more recently for dairy cows, horses, and ponies. The goal of this study was to apply FCP to the domestic dog, and to investigate the inter-observer reliability of assessments of emotional expression in 10 individual Beagles by a group of 18 untrained observers. The data was analysed using Generalised Procrustes Analysis (GPA), a multivariate statistical technique associated with FCP. The observers achieved highly-significant agreement in their assessments of the dogs’ expressions, thereby establishing the applicability of this methodology for the first time in the domestic dog.

Qualitative behavioural assessment (QBA) is based on observers’ ability to capture the dynamic complexity of an animal’s demeanour as it interacts with the environment, in terms such as tense, anxious or relaxed. Sensitivity to context is part of QBA’s integrative capacity and discriminatory power; however, when not properly managed it can also be a source of undesirable variability and bias. This study investigated the sensitivity of QBA to variations in the visual or verbal information provided to observers, using free-choice profiling (FCP) methodology. FCP allows observers to generate their own descriptive terms for animal demeanour, against which each animal’s expressions are quantified on a visual analogue scale. The resulting scores were analysed with Generalised Procrustes Analysis (GPA), generating two or more multi-variate dimensions of animal expression. Study 1 examined how 63 observers rated the same video clips of individual sheep during land transport, when these clips were interspersed with two different sets of video footage. Scores attributed to the sheep in the two viewing sessions correlated significantly (GPA dimension 1: rs=0.95, P<0.001, GPA dimension 2: rs=0.66, P=0.037) indicating that comparative rankings of animals on expressive dimensions were highly similar, however, their mean numerical scores on these dimensions had shifted (RM-ANOVA: Dim1: P<0.001, Dim2: P<0.001). Study 2 investigated the effect of being given different amounts of background information on two separate groups of observers assessing footage of 22 individual sheep in a behavioural demand facility. One group was given no contextual information regarding this facility, whereas the second group was told that animals were moving towards and away from a feeder (in view) to access feed. Scores attributed to individual sheep by the two observer groups correlated significantly (Dim1: rs=0.92, P<0.001, Dim2: rs=0.52, P=0.013). A number of descriptive terms were generated by both observer groups and used in similar ways, other terms were unique to each group. The group given additional information about the experimental facility scored the sheep’s behaviour as more ‘directed’ and ‘focused’ than observers who had not been told. Thus, in neither of the two studies did experimentally imposed variations in context alter the characterisations of animals relative to each other, but in Study 1 this did affect the mean numerical values underlying these characterisations, indicating a need for careful attention to the use of visual analogue scales.

The welfare consequences of long-distance transportation of animals remain a controversial topic. Animals that stand for most of the long journey (especially if additional muscular activity is required to deal with postural instability) are at risk of developing fatigue. Previous observational studies of behaviour and physiology suggested either that sheep do not become markedly fatigued by long journeys or that previous methods did not adequately identify fatigue. A range of behavioural and physiological measures were made on eight pairs of sheep during and after treadmill exercise. Within each pair of sheep, a treatment sheep was walked on a treadmill at 0.5 m/s for up to 5 h or until the sheep voluntarily stopped exercising or showed other signs of reduced performance, and a control sheep was exercised for two 10-min periods on either side of the exercise period for the treatment sheep. With the exception of one sheep that only walked for 4.5 h, all treatment sheep walked for 5 h without apparent difficulty. After exercise, the plasma cortisol concentration of treatment sheep was significantly greater than that of control sheep. However, there were no significant treatment effects on plasma creatine kinase activity or blood lactate concentration. After 5 h of exercise, there was a proportionate decrease in the median frequency of the electromyogram recorded over the m. semitendinosus, and this was significantly different from control sheep. There was no evidence that treatment sheep lay down sooner or for longer after treadmill exercise than controls. In sheep tested in a maze to examine whether there was increased motivation to rest after exercise, there was no significant difference between the times taken by treatment and control sheep to obtain a food reward. Qualitative behavioural assessment of the sheep by a panel of observers identified two main dimensions of sheep demeanour, but among descriptors elicited from observers only one person used a term associated with fatigue. No significant difference was found between the scores of treatment and control sheep on these two demeanour dimensions. Thus, there was little evidence that prolonged gentle walking exercise fatigues sheep. Further development of methods to both repeatedly induce and to identify fatigue in sheep is required.

The assessment of the health and welfare of farm animals is steadily gaining importance, not only from a human health perspective, but also for the sake of the animals themselves. Consumers are increasingly concerned with the effect of industrial farming practices on animal well-being, and so there is a growing need to develop scientifically validated methods which address the welfare experience of animals on farms. Qualitative Behaviour Assessment (QBA) is a novel method developed at Scottish Agricultural College (SAC), designed to make an animal’s welfare experience accessible through a ‘whole animal’ approach. From a ‘whole animal’ perspective, what animals feel is an integral part of what they do, and so their experience speaks to us through their behaviour, as an expressive ‘body language’. By integrating details of an animal’s posture, behaviour and of the context in which these occur, we can assess that body language, and judge whether the animal is calm, anxious, enthusiastic or agitated, whether it is generally content or in distress. The question is whether such judgements are reliable, and can be a legitimate part of scientific welfare assessment. Traditionally scientists are taught to be distrustful of applying terms such as ‘content’ and ‘distressed’ to animals, regarding use of such terms as ‘subjective’ or ‘anthropomorphic’. However, several years of Defra/SEERAD funded research at SAC have shown qualitative judgements of behaviour to be unfailingly reliable, repeatable, and well-correlated to quantitative behavioural and physiological welfare indicators, both for individual animals and for animals kept in groups (eg Wemelsfelder et al., 2000, 2001; Wemelsfelder & Lawrence, 2001; Rousing & Wemelsfelder, 2006; Wemelsfelder, 2007). Thus, when facilitated by an appropriate methodology, qualitative judgements of behaviour can and do attain scientific validity.

Most studies which provide qualitative assessments of animal behaviour use prefixed lists of adjectives (e.g. shy, bold, interested). However,Wemelsfelder et al. (2000, 2001), using Free-Choice-Profiling (FCP) methodology, allowed observers to develop their own descriptors for scoring pigs, and found this method to be highly reliable and repeatable. Repertory grid technique (RGT) is a frequently used method in human psychology and resembles FCP. Recently RTG has also been applied to assess personality in chimpanzees (Dutton et al., 1997). The purpose of the present study was to apply RGT to pigs and to correlate obtained pig scores with those previously obtained through FCP.

Qualitative assessment of behaviour is based upon the integration by the observer of many pieces of information which in conventional quantitative approaches are recorded separately. Behaviour is for example summed up as ‘fearful’, ‘anxious’ or ‘distressed’. Such interpretations form the core of the lay public's concern for animal suffering, and therefore are frequently dismissed as ‘anthropomorphic’ and unscientific. However, in theory it is possible that such assessments reflect observable aspects of behavioural organization, and may be liable to scientific analysis. A first step in considering this hypothesis is to investigate the interobserver reliability of qualitative assessments. The aim of this study was to determine the extent to which untrained observers show agreement in their spontaneous qualitative descriptions of interactive behaviour in growing pigs.