Broiler chickens are typically housed in barren environments and at high stocking densities in order to minimize production cost, which has the potential to compromise broiler welfare[1,2,3,4]. In conventional housing systems, broilers spend approximately 80% of their time budget lying down, and a positive association between time spent lying down and lameness has been found, which negatively influences broiler welfare. Environmental enrichment (complexity) has a positive impact on animals’ biological functioning and behavior[1,4,6,7,8,9,10]. Broilers with access to elevated platforms experienced improved gait through an increased occurrence of low (good) gait scores, lower (better) flock mean gait score, and lower occurrence of tibial dyschondroplasia compared to broilers without access to platforms. Stocking density is another important environmental factor that can impact broiler chicken welfare aspects, such as leg health, level of bruising and scratches, lameness, and behavioral suppression[12,13,14]. While housing broilers at high stocking densities maximizes profit for the producer, it has the potential to compromise bird health and welfare as seen through decreased final body weight and feed conversion ratio, as well as increased occurrence of footpad dermatitis and mortality[15,16,12,47]. Furthermore, high stocking densities reduce space use and therefore activity level, and increase disturbances that lead to decreased plumage and carcass quality[2,15,18,20,49]. These and other studies show that conventional housing of broiler chickens has some negative effects on bird health and welfare, although their effects on affective state are unknown.
The ability to perform highly motivated behaviors is important for good animal welfare[1,21,22,23]. Three species-specific behaviors have been identified for broilers, whose deprivation may cause negative affective states: perching, dustbathing, and foraging[23,24,25,26,27]. Perching is a natural behavior for jungle fowl (broilers’ ancestors), which seek elevated resting spaces possibly to avoid predation while sleeping[28,29]. Broilers have maintained motivation to perch when given easily accessible perches. When provided perches at a low stocking density, broilers perched for longer (nearly 25% of the observation periods) and were less aggressive than control birds. Dustbathing is another highly-motivated behavior for broilers, as jungle fowl who are deprived of substrate to dustbathe in will exhibit excessive, compensatory dustbathing behavior when eventually provided with a substrate or even dustbathe on wire flooring, highlighting the need to perform this behavior regardless of access to a suitable substrate. Broiler chickens dustbathe throughout their life, even chicks have been shown to dustbathe during the first week of life (Jacobs, 1.7% of observed time, unpublished data). Furthermore, jungle fowl and broilers have retained the motivation to forage. Even when regularly fed, semi-wild junglefowl foraged for 61% of the time they spent active, as well as spent 34% of their time scratching, a behavior considered to be associated with foraging. Without access to a preferred substrate or environment, domestic fowl forage in feces[34,35], suggesting that this behavior is highly motivated. Commercial broilers can be deprived of all three of these high-motivation behaviors; they do not have access to perches or proper substrate to dustbathe and forage in. While they do have access to litter, this is often reused between flocks, and therefore contains moisture and feces that fosters rapid bacteria growth and ammonia production[35,36]. Because broilers are inactive for large portions of the day, the risk of developing contact or footpad dermatitis increases with provision of dirty litter, and could lead to a negative affective state due to pain associated with dermatitis[37,38,39].
Life experiences, including environmental conditions, can elicit short-term emotions which are defined as functional states elicited by reward or punishment (stimuli that animals work to either gain or avoid). These emotions are adaptive and help animals appropriately respond to changes in their environment. Emotional responses shape an animal’s mood, which can be defined as long-term, diffuse states that reflect the cumulative valence of emotions over time. Both emotion and mood contribute to an animal’s affective state, which can be measured along a spectrum between positive and negative (valence). An animal that experiences more positive than negative emotions throughout life, for instance induced by the ability to express play behaviors, will be in an overall positive affective state, compared to animals that have more negative experiences. When an animal experiences more negative than positive emotions, such as chronic or excessive fear and anxiety, the animal will be in an overall negative affective state[6,43,44]. Typically, negative experiences tend to have a stronger influence on affective state than positive experiences[45,46,47]. In order to achieve good welfare, the induction of positive experience must be considered in addition to preventing negative experiences. However, most published studies for agricultural species focus on avoidance of negative affective states such as level of fear[7,48], or physiological measures, like lameness[8,50].
Affective states are closely associated with cognition–the mechanisms by which animals acquire, process, store, and act on information from their environment. Affective states influence cognitive processing and cognitive processing impacts affective states[51,52,53]. When emotions and affective states impact aspects of cognition, such as judgement, attention, and memory, we call this “cognitive bias”[54,55,56]. Cognitive biases can be used as an indicator of animal welfare[41,55,57,58,59].
Humans experiencing a negative affective state (depression, anxiety) tend to interpret ambiguous events negatively and have a pessimistic outlook, whereas humans who are in a more positive affective state tend to interpret the same ambiguous events positively and have a more optimistic outlook[60,61,62,63]. Animal responses to ambiguous situations can be quantified using a judgement bias test, assessing cognitive bias. Judgement bias testing is used to determine levels of optimism and pessimism of subjects based on responses to ambiguous cues during testing. Judgement bias cues can be spatial, visual, auditory, olfactory, tactile or a combination of these (multimodal). Shorter latencies to approach ambiguous cues would indicate optimism, whereas longer latencies to approach ambiguous cues would indicate pessimism[9,41,69,70,71,72,73,74]. A meta-analysis of 71 judgement bias studies on 22 species showed optimistic and pessimistic responses to ambiguous situations resulting from positive or negative affective states, respectively. Thus, judgement bias is considered the “gold standard” for evaluating affective states in animals.
One previous study has evaluated broiler chicken affective state through a judgement bias test. Iyasere et al. (2017) trained broilers on a spatial, go/no-go judgement bias task to discriminate between a reward- (mealworms) and punishment-associated (air puff) cone. Following discrimination, birds treated with corticosterone had longer latencies to displace cones at all cue locations compared to control birds, suggesting a pessimistic bias. A similar test approach was conducted on laying hens, where ‘exploratory’ layer hens (categorized based on novel object and open area test responses) housed with enrichments showed more optimistic responses than exploratory hens housed without enrichments. Therefore, judgement bias tests could be a valuable tool to assess positive affective states in broilers housed under varying environmental conditions. Including an evaluation of lameness in conjunction with judgement bias testing is warranted, as birds were required to walk in judgement bias training and testing.
In the present study, we used a judgement bias test to assess the effect of environmental complexity and stocking density, manipulated in a factorial experiment, on broilers’ affective states. Gait was quantified as a potential confounding factor for the judgement bias test. We hypothesized that birds housed in high-complexity (HC), low-density (LD) environments would respond more optimistically, indicating more positive affective states, compared to birds from low-complexity (LC), high-density (HD) environments. Birds housed in LC/LD or HC/HD environments were predicted to show intermediate levels of optimism.