Cafeteria feeding systems and diet selection behaviour in ruminants: A comprehensive review

Introduction

Animals select feeds based on various motivations, aiming not only to maximise energy intake as proposed by optimal foraging theory. But also, to meet specific nutrient needs like proteins and minerals (Ceacero et al., 2010) [10] . Ruminants exhibit 'nutritional wisdom', recognising beneficial or harmful feed components through sensory cues such as taste or smell (Duncan and Gordon, 1999) [25] , enabling them to choose diets that support rumen health or alleviate illness (Cooper et al., 1995; Villalba et al., 2010) [14,73] . Studies show that ruminants, whether housed or grazing, often select diverse and sometimes suboptimal diets beyond what optimal foraging theory predicts (Prache and Damasceno, 2006) [60] . However, modern feeding practices like total mixed rations or monoculture pastures may hinder natural foraging behaviour, leading to imbalances and health issues (Lynch et al., 1992) [49] .

Usage & Application

Choice feeding experiments evaluate an animal's willingness to consume specific feeds when offered separately or in combination, thereby helping to determine feed preferences and palatability. These experiments can be conducted either indoors or on pasture. The primary objectives (Meier et al., 2012) [50] are to assess intake proportions, consumption rates and feeding behaviour. According to Meier et al. (2012) [50] , the term "cafeteria experiment" has two main applications: 1. To assess animal preference or the palatability of various feeds, and 2. To evaluate the animal's ability to meet its nutritional requirements, avoid harmful effects, or engage in self-medication to support health.

Intensions of Choice Feeding Experiments

Choice feeding experiments are employed to evaluate several adaptive abilities in animals. These include maintaining an optimal rumen environment by selecting feeds with varying energy densities (Cooper et al., 1995) [14] or physical forms such as chopped, ground, or pelleted feeds (Cooper et al., 1996) [16] . Animals may also select feeds based on their nutritional requirements (Fedele et al., 2002) [29] or engage in self-medication to counteract adverse health conditions (Villalba et al., 2010) [73] . Additionally, such experiments investigate how prior experiences influence feed preferences (Catanese et al., 2010) [12] and how physiological stages, such as growth or reproduction, affect feed selection (Parsons et al., 1994) [57] .

Theoretical Models Explaining Diet Selection in Ruminants

According to Milne (1991) [51] , multiple factors influence the feeding behaviour and diet selection of ruminants, leading to the development of various models and theories to explain the mechanisms underlying these choices. a) Euphagia: The Euphagia model describes an animal's innate ability to select a diet that meets its nutritional requirements (Cooper, 1995) [14] . This concept is supported by the idea of nutrient specific appetite, exemplified by pica -an instinctive drive to consume specific substances to correct deficiencies (Rozin, 1976) [65] . Such behaviour suggests that most livestock species possess an inherent capacity to choose nutritionally appropriate diets. For instance, Danford (1982) [17] observed that phosphorus-deficient cattle may attempt to rectify the imbalance by consuming bones. b) Hedyphagia: Cooper (1995) [14] proposed the hedyphagia model, which suggests that animals select diets based on sensory appeal. Sensory-driven preferences are evident in sheep, which tend to favour young, soft, leafy forages over older, fibrous ones, often guided primarily by their senses (Milne, 1991) [51] . Animals learn to associate sensory traits like taste and smell with post-ingestive effects, shaping their food preferences over time (Distel & Villalba, 2018) [21] . While sight plays a key role in orientation (Kilonzo, 2003), sheep can still identify and select forage using smell, taste, and touch even when vision is impaired (Forbes, 1995) [30] . Secondary plant compounds also influence intake, either attracting or deterring herbivores depending on their sensory properties (Pfister, 1999) [59] . c) Optimal Foraging Theory (OFT): This theory suggests that herbivores evaluate food based on its intrinsic quality and decide whether to consume or avoid it accordingly (Provenza et al., 2002) [61] . Because foraging demands both time and energy, animals adopt strategies that maximize energy intake relative to effort. Natural selection favours behaviours that enhance survival, including efficient foraging (Cooper, 1995) [14] . Large herbivores, for example, tend to minimize foraging time and energy expenditure while selecting energy-rich diets (Norberg, 2021) [55] , and may shift locations to sustain optimal energy intake as environmental conditions change (Bergman et al., 2001) [8] . d) Morphophysiology: Animal-related factors such as body size, sex, breed, and morphology significantly influence diet selection. Larger ruminants are more efficient at digesting fibrous feeds due to their greater rumen capacity, which allows longer retention and more thorough microbial breakdown (Milne, 1991; Cooper, 1995) [51,14] . In contrast, browsing species with smaller rumens tend to pass indigestible particles more quickly, as browse materials often contain highly lignified, less digestible cell walls.

e) e. Learning through consequences: The model illustrates how dietary preferences are shaped by both pre ingestive experiences and post ingestive effects, often acquired through social learning or trial and error. Herbivores, when exposed to diverse diets, can learn the benefits of specific plant secondary compounds (Villalba et al., 2010) [73] and natural selection has equipped them with the ability to acquire such knowledge. Sheep typically sample unfamiliar foods in small amounts and increase consumption if no negative consequences follow. Young lambs can develop taste aversions or preferences through observation and interaction with role models, particularly their mothers, although this does not ensure identical dietary choices (Kilonzo, 2003) [47] . [2,12] and goats (Duncan et al., 2006) [26] . Fewer studies have involved cattle, with some experiments conducted on heifers (Sandoval-Castro et al., 2005) [67] and dairy cows (Van Dorland et al., 2008) [72] . Cattle are more commonly used in outdoor feeding studies (Hirata et al., 2010) [42] , with sheep (Harvey et al., 2000) [39] and goats (Penning et al., 1997) [58] [2,11,20] , though some use open-air enclosures for penned animals (Ben Salem et al., 1994) [7] . Group housing with individual intake measurement requires advanced equipment, making it rare (Van Dorland et al., 2008) [72] . As a result, intake is often recorded at the group level (Askar et al., 2006; Kalio et al., 2006) [3,46] . Experimental feeds are typically offered separately in troughs (Scharenberg et al., 2007) [68] or buckets (Kalio et al., 2006) [46] to minimize losses. To prevent positional bias, feed locations are randomized daily (Van Dorland et al., 2007) [71] , since animals may show preferences for certain spots (Forbes, 2007) [31] .

Experimental Design for

Methodologies and Measurements Used in Choice Feeding

Experiments Measurement of Individual Feed Intake: The primary variable for evaluating preference or palatability in choice feeding experiments is the intake of the specific feed item(s) tested. In indoor settings, intake can be measured manually or with computer-assisted systems that use feeding troughs equipped with weight scales (Forbes, 2007) [31] . These systems, when combined with electronic identification units and access-controlled feeding doors, allow preference assessments even in group housing and can detect diurnal feeding patterns (Van Dorland et al., 2007) [71] . Outdoors, intake is typically estimated based on intake rate (amount per time) and grazing duration (Harvey et al., 2000) [

, with intake rate derived from jaw movement or grazing time records (Prache and Damasceno, 2006) [60] , alongside body weight changes and feed dry matter content. Alternative methods include using oesophageal fistulae to collect extrusa for bite weight estimation, while bite rate can be recorded using sound-based systems (Distel et al., 1995) [22] . A simpler approach involves recording the forage offered and remaining, with biomass growth in protected areas assessed using exclosure cages (Chapman et al., 2009) [13] , though this method limits the ability to collect data for individual animals.

Measurement of feeding behaviour

Feeding behaviour traits may include (i) eating time, rate and meal size per feed in indoor settings; (ii) grazing time, intake rate, meal size and position in pasture outdoors; and (iii) frequency of switching between feeds, applicable in both environments. Behaviour can be assessed through direct observation (Chapman et al., 2009; Rutter et al., 2004) [13,66] , video recording (Hirata et al., 2006) [40] , or photographic recording (Prache and Damasceno, 2006) [60] . Automated systems like the 'Ethosys' collar (Ginane et al., 2003) [34] , elastic noseband devices (Parsons et al., 1994) [57] , and the IGER recorder (Rutter et al., 2004) [66] can track behaviour, including at night (Gordon, 1995) [36] . While direct observation is common outdoors, it can also be done indoors, though less frequently due to alternative methods. This method involves tracking animal location or specific behaviours over time, but observer presence must not influence behaviour, and observations may be made from outside the plot or a concealed elevated position (Prache and Damasceno, 2006; Rutter et al., 2004) [60,66] . This method allows for detailed tracking of behaviour, detecting unusual events, and analysing temporal patterns in feed preference (Morand-Fehr, 2003) [52] . Scan sampling is often used to record group behaviour at intervals (Hirata et al., 2008) [41] , capturing traits like ruminating and lying. Video recording, useful for both short-and long-term behaviour analysis, allows for playback and manipulation with specialised software (Weigensberg and Fairbairn, 1994) [74] .

Measurement of Secondary Response Variables

Feeding strategies are designed not only to maximize feed intake but also to influence secondary outcomes such as animal performance, digestion, metabolism, product quality, and gaseous emissions.

Key Aspects to Consider for the Successful Conduct of Choice Feeding Experiments

A. Animal-Specific Aspects 1. Importance of animal species: Although ruminant species such as cattle, sheep, and goats share many behavioural traits and abilities (Ekesbo, 2011) [27] , species-specific characteristics can influence outcomes in choice feeding experiments. For instance, sheep are highly social animals and may become distressed when isolated, potentially affecting feeding behaviour. Feed preferences can also vary with feeding type: cattle and sheep are primarily "grass and roughage eaters," whereas goats are considered "intermediate types" (Hofmann, 1989) [43] . Rogosic et al. (2006) [64] reported that goats consumed more shrubs than sheep, although the order of preference was similar between the species. 2. Physiological stage of the animals: Individual nutritional needs (Dikmen et al., 2009) [20] and physiological stages such as growth, pregnancy, lactation or the dry period can significantly influence feed preferences. Hunger is a major confounding factor in choice feeding studies, as hungry animals often consume the first available feed regardless of their actual preference (Scharenberg et al., 2007) [68] . Newman et al. (1994) [53] reported that hungry sheep grazed for longer periods and at a faster rate but exhibited less preference for clover compared to well-fed counterparts. Feed selection can also be shaped by specific macronutrient deficiencies; lambs fed imbalanced diets selected feeds that corrected the nutritional shortfall, regardless of physical form (pellets, rolled, or ground). Fedele et al. (2002) [29] found that goats given free-choice access to six different feeds adjusted their intake across reproductive stages, indicating that diet selection was aligned with physiological demands. Similarly, Cooper et al. (1994) [15] showed that pregnant ewes displayed a greater preference for protein-rich feeds than nonpregnant ones, especially when energy availability was high. In contrast, Parsons et al. (1994) [57] observed minimal and inconsistent effects of lactation on feed preferences in sheep, and Morand-Fehr (2003) [52] reported no clear association between physiological stage and feed preference in goats.

Health status may also influence feed choice. Parasiteinfected animals may alter their preferences if familiar with palliative or medicinal compounds. For instance, Villalba et al. (2010) [73] observed an increased preference for a tannin-alfalfa mixture in infected lambs. The role of sex in feed preference remains uncertain, with Rodríguez et al. (2008) [63] reporting inconclusive findings.

Number of animal replicates: Individual animals may

exhibit variation in feed preferences and their perception of palatability (Dikmen et al., 2009) [20] . Due to the high resource demands associated with choice feeding trials, these studies often involve a limited number of animals per treatment group (Catanese et al., 2009) [11] .

In most indoor experiments, animals are either individually housed or tethered and treated as distinct experimental units. [63,72] and eight (Dikmen et al., 2009) [20] . Even when animals are group-housed (Van Dorland et al., 2007) [71] , intake measurements are frequently conducted using five to six individuals.

In pasture-based studies, group sizes vary widely from as few as two animals (Rutter et al., 2004) [66] to approximately 30 (Hirata et al., 2008) [41] . Some studies use a single 'focal' animal for observation (Penning et al., 1997) [58] , while others treat the entire group (Rutter et al., 2004) [66] . as a replicate. However, statistical independence may be compromised in group settings due to the influence of social dynamics on feeding behaviour (Hirata et al., 2008) [41] , although other studies have observed relatively independent feeding actions (Ginane et al., 2005) [35] . Latin square designs are often employed to enhance resource efficiency; nonetheless, the potential for carry-over effects between treatments may introduce bias.

Animal Adaptation and Consideration of Previous Experience

Adaptation Periods in Choice Feeding Studies: In indoor choice feeding experiments, adaptation periods to the experimental environment typically range from one week (Dikmen et al., 2009) [20] to two weeks (Hadjigeorgiou et al., 2003) [38] , with Morand-Fehr (2003) [52] using 15-18 days for goats. A minimum of six days is considered necessary for animals to adjust to new feeds in terms of intake and digestion (Demarquilly et al., 1995) [18] , as initial neophobia may suppress intake, which gradually increases with familiarization.

Young animals generally adapt more readily to novel feeds compared to adults (Lynch et al., 1992) [49] . For accurate palatability assessment, Kaitho et al. (1996) [45] recommended more than one day of feed exposure, while Kyriazakis et al. (1998) [48] suggested a 10-day period to allow animals to associate feed characteristics with post-ingestive effects.

In outdoor experiments, prior grazing experience is often assumed (Hirata et al., 2006) [40] , though some studies have implemented adaptation periods for instance, 12 days (Penning et al., 1997) [58] or over 14 days (Newman et al., 1994) [53] . Learning plays a crucial role, particularly when animals respond to post-ingestive feedback. For example, goats have been shown to recognise phenolic compounds within an hour (Papachristou et al., 2003) [56] and parasitised lambs learned to prefer tannin-rich feeds over a few days (Villalba et al., 2010) [73] .

Influence of Prior Experience on Feed Preference: Prior exposure to experimental or similar feeds, especially during early life stages, significantly influences feed preference and intake behavior in choice feeding trials (Parsons et al., 1994) [57] . Such exposures can even occur prenatally (Simitzis et al., 2008) [69] or via social learning, where maternal influence tends to be stronger than that of peers (Lynch et al., 1992) [49] . Early-life feed experiences can exert long-lasting effects on feeding behaviour. Familiar feeds or forms, such as hay, may temporarily increase intake due to novelty when replacing a previous diet (Favreau et al., 2010) [28] , potentially biasing preference evaluations. However, prolonged exposure can lead to greater acceptance and intake of initially unpalatable feeds (Ngwa et al., 2003) [54] .

Assessments in Individually Housed Animals Versus Group-Housed Animals

Social dynamics within a flock can significantly influence grazing behaviour, potentially obscuring individual feed preferences in choice feeding experiments (Hirata et al., 2008) [41] . The tendency to remain close to conspecifics may override feed selection, particularly in small groups. For gregarious species such as sheep, using group sizes larger than three to six individuals has been suggested to help mitigate social interference (Dumont and Boissy, 2000) [24] . On pasture, experimental designs should allow animals to express their preferences freely, without being constrained by social pressures. In indoor settings, providing individual access to feeders within group housing can help reduce dominance-related effects; however, limited feeder access may still result in intake suppression due to social hierarchies (Fraser and Broom, 1997) [32] . While individual housing effectively eliminates social influences, it may reduce feed intake in some species, such as cattle, due to isolation stress.

Feed-Specific Aspects 1. Number of Feeds Tested Simultaneously:

In outdoor choice feeding studies involving ruminants, the number of feed options offered can vary widely from as few as two (e.g., Catanese et al., 2009) [11] to as many as ten forages though most studies typically employ dual choice designs (Hirata et al., 2008) [41] . While increasing the number of feed choices can complicate the decisionmaking process for animals, it may also lead to higher overall intake.

Characteristics of Experimental Feeds:

The physical characteristics of experimental feeds can significantly influence animal preferences (Morand-Fehr, 2003, for goats) [52] . To avoid introducing bias into choice feeding trials, the physical properties of the feeds such as form, length and density should be as similar as possible, unless those traits are the variables under investigation. For forages, key considerations include dry matter content and particle length, with physical form being particularly important for feeds with low palatability.

Small ruminants may also show a preference for concentrates containing coarsely ground ingredients (Baumont et al., 2000) [6] . The palatability of finely ground or dusty feeds can be improved through moistening or pelleting. Sensory attributes such as odour and taste also play a crucial role in feed selection (De Rosa et al., 2002; Distel et al., 2007) [19,21] . Moreover, the quality and digestibility of experimental feeds can affect rumen fill and feeding behaviour (Ginane et al., 2003) [34] . Fibre characteristics influence intake rate (Favreau et al., 2010) [28] , and feeds with different potential intake rates may lead to distinct choices. The presence of toxins or plant secondary metabolites can also impact intake.

Set-up Specific Aspects Duration and Timing of Exposure to Experimental Feeds:

Access to experimental feeds in choice feeding studies can be provided either continuously throughout the day and night (Distel et al., 2007) [21] or restricted to specific periods.

Restricted access may vary in duration, from short periods of 30 minutes or less (Morand-Fehr, 2003) [52] to extended periods ranging from over 30 minutes to as long as 8 hours (Alonso-Díaz et al., 2009) [2] . The decision regarding access duration may also be influenced by practical considerations such as resource availability. Extended exposure times are especially important when animals are offered multiple feed options, as they require sufficient time to evaluate and select the most suitable diet. Short-term preference assessments may not reliably predict long-term feeding behaviour. Preferences observed during brief exposure periods may differ significantly from those seen over extended periods (Parsons et al., 1994) [57] . Furthermore, the timing of feed exposure can influence feeding outcomes. Diurnal preference patterns have been documented when animals were given prolonged access to feeds (Catanese et al., 2009; Van Dorland et al., 2008) [11,72] . Typically, animals show higher grazing activity during early morning and late evening hours (Lynch et al., 1992; Albright and Arave, 1997) [49,1] , likely due to more favorable environmental conditions, such as lower solar radiation and temperatures.

Length of choice feeding periods

The duration of the choice period in feeding studies can be classified into short-term (less than one day; e.g., Distel et al., 1995) [22] , medium-term (one day to one week; e.g., Hirata et al., 2006) [40] , and long-term (several weeks; e.g., Hirata et al., 2010) [42] , often preceded by adaptation or learning phases. Some studies utilize a single exposure to experimental feeds over several days (e.g., Atwood et al., 2006) [4] , while others employ repeated exposures either within a single day (Distel et al., 1995; Distel and Vallaba (2018) [22,23] or across multiple days (e.g., Kyriazakis et al., 1998) [48] . Short-term durations are commonly used for their cost-effectiveness, particularly in assessing palatability or initial feed preferences (Morand-Fehr, 2003) [52] .

Factors Influencing Diet Selectivity

• Type of Livestock: Bailey et al. (2019) [5] classified herbivores into grazers, browsers, and intermediate feeders based on anatomical traits and foraging behaviour. Large ruminants, such as cattle, possess broad muzzles and large rumens, enabling efficient utilization of grasses. The extended retention time of feed within the large rumen-reticulum complex supports microbial fermentation, enhancing nutrient extraction. Janis and Ehrhardt (1988) [44] observed morphological differences in muzzle shape between grazers and browsers, noting that grazers typically have broader muzzles adapted to consuming short grasses. Similarly, Coopers (1995) [14] emphasized that a large rumen volume enables effective digestion of fibrous plant materials. • Breed: Frost and Mosley (2020) [33] noted that livestock breeds vary in body size and production characteristics, which in turn influence their nutritional requirements. Larger breeds typically require higher dry matter intake and possess greater digestive capacity. Breed-specific grazing behaviour also plays a role in feed selection and resource utilization. According to Brand (2000) [9] , Dorper sheep are less selective grazers compared to Merino sheep, consuming a wider variety of plant species, including shrubs. This broader dietary adaptability allows Dorpers to gain weight more rapidly, shortening the time to market and thereby potentially reducing greenhouse gas emissions associated with prolonged production cycles. • Age: Grings et al. (2001) [37] reported that metabolic rate declines with age, resulting in reduced feed requirements in older animals, while younger animals exhibit more active foraging behaviour to satisfy their higher nutritional demands. Frost and Mosley (2020) [33] further observed that diet selectivity and tolerance to plant secondary metabolites diminish as animals age. Younger animals, due to their limited foraging experience, tend to explore a wider range of feeds, including those that may have previously caused discomfort. • Body Condition: Body condition, an indicator of an animal's fatness or thinness, plays a significant role in regulating feeding behaviour and intake levels. Frost and Mosley (2020) [33] reported that animals in poor body condition typically exhibit increased feed intake as a compensatory response, whereas well-conditioned animals tend to reduce their intake. Similarly, Newman et al. (1994) [53] observed that undernourished or fasted sheep are more likely to consume greater quantities of available feed, including less preferred or even potentially toxic plants, in an effort to address nutritional deficiencies. • Stage in Production Cycle: Animals adapt their dietary choices in response to changing nutritional demands across different life stages (Frost and Mosley, 2020) [33] . Lactating females, for instance, require higher levels of protein and energy to support milk production, with energy requirements fluctuating depending on the specific phase of production due to metabolic changes. However, certain plant species may be toxic during pregnancy and pose risks such as fetal harm or pregnancy loss, highlighting the importance of avoiding the consumption of such plants during gestation (Stegelmeier et al., 2020) [70] . • Selective Grazing: Foraging animals meet their seasonal nutritional demands by selectively choosing plant species and parts that support digestion and provide comfort, while avoiding those that cause discomfort. Exposure to unpalatable or discomfort-inducing feeds can lead to avoidance behaviours and repeated consumption of poorquality feed may suppress future intake. • Provenza et al. (2002) [61] explained that when animals are exposed to a range of diets with differing nutrient and toxin levels, they regulate their intake to balance nutrient needs and minimize toxin ingestion. Toxins can impair physiological functions by reducing feed intake, hindering digestion, causing nutritional imbalances and negatively affecting reproduction and immunity. • External Plant Attributes (EPA): Raufirad et al. (2015) [62] suggested that external plant attributes (EPAs) are the first features animals perceive while grazing, significantly influencing their preference and selectivity. Traits such as spines, inflorescence, stem structure, resistance to grasping, height, leaf position, succulence, leaf-to-stem ratio and branch density all play a role in determining plant selection by sheep.

Conclusion

Cafeteria feeding provides a practical method for meeting the dynamic nutritional needs of livestock and poultry through self-selection. Research indicates that animals have the ability to instinctively select balanced diets when given appropriate feed options. This approach can enhance nutrient efficiency and reduce protein overuse, contributing to more sustainable production practices. However, the success of cafeteria feeding depends on well-designed feeding systems and effective management practices. While initial costs and the learning curve may present challenges, the long-term benefits in terms of performance and feed economy are promising. Overall, cafeteria feeding serves as a valuable tool in advancing precision nutrition within modern animal agriculture.

Conflict of Interest Not available

Financial Support

References

1. Albright JL, Arave CW. The behaviour of cattle. CAB International, Wallingford, Oxon, UK, 1997.
2. Alonso-Díaz MA, Torres-Acosta JFJ, Sandovall-Castro CA, Hoste H, Aguilar-Caballero AJ, Capetillo-Leal CM. Sheep preference for different tanniniferous tree fodders and its relationship with in vitro gas production and digestibility. Animal Feed Science and Technology. 2009;151:75-85.
3. Askar AR, Guada JA, González JM, de Vega A, Castrillo C. Diet selection by growing lambs offered whole barley and a protein supplement, free choice: effects on performance and digestion. Livestock Science. 2006;101:81-93.
4. Atwood SB, Provenza FD, Villalba JJ, Wiedmeier RD. Intake of lambs offered ad libitum access to one of three iso-caloric and isonitrogenous mixed rations or a choice of all three foods. Livestock Science. 2006;101:142-149.
5. Bailey DW, Mosley JC, Estell RE, Cibils AF, Horney M, Hendrickson JR, et al. Synthesis paper: Targeted livestock grazing: Prescription for healthy rangelands. Rangeland Ecology and Management. 2019;72(6):865877.
6. Baumont R, Prache S, Meuret M, Morand-Fehr P. How forage characteristics influence behaviour and intake in small ruminants: A review. Livestock Production Science. 2000;64:15-28.
7. Ben Salem H, Nefzaoui A, Abdouli H. Palatability of shrubs and fodder trees measured on sheep and dromedaries: Methodological approach. Animal Feed Science and Technology. 1994;46:143-153.
8. Bergman CM, Fryxell JM, Gates CC, Fortin D. Ungulate foraging strategies: Energy maximizing or time minimizing? Journal of Animal Ecology. 2001;70(2):289-300.
9. Brand TS. Grazing behaviour and diet selection by Dorper sheep. Small Ruminant Research. 2000;36(2):147-158.
10. Ceacero F, Landete-Castillejos T, García AJ, Estévez JA, Gallego L. Can Iberian red deer discriminate among essential minerals in their diet? British Journal of Nutrition. 2010;103:617-626.
11. Catanese F, Distel RA, Arzadún M. Preferences of lambs offered Italian ryegrass and barley herbage as choices. Grass and Forage Science. 2009;64:304-309.
12. Catanese F, Distel RA, Rodríguez I, Villalba JJ. Role of early experience in the development of preference for low-quality food in sheep. Animal. 2010;4:784-791.
13. Chapman GA, Bork EW, Donkor NT, Hudson RJ. Performance and dietary preferences of white-tailed deer grazing chicory, birdsfoot trefoil or alfalfa in north central Alberta. Journal of Animal Physiology and Animal Nutrition. 2009;93:794-801.
14. Cooper SDB, Kyriazakis I, Nolan JV. Diet selection in sheep: the role the rumen environment in the selection of a diet from two feeds that differ in energy density. British Journal of Nutrition. 1995;74:39-54.
15. Cooper SDB, Kyriazakis I, Oldham JD. The effect of late pregnancy on the diet selections made by ewes. Livestock Production Science. 1994;40:263-275.
16. Cooper SDB, Kyriazakis I, Oldham JD. The effects of physical form of feed, carbohydrate source and inclusion of sodium bicarbonate on the diet selections of sheep. Journal of Animal Science. 1996;74:1240-1251.
17. Danford DE. Pica and nutrition. Annual Review of Nutrition. 1982;2:303-322.
18. Demarquilly C, Chenost M, Giger S. Pertes fécales et digestibilité des aliments et des rations. In: Jarrige R, Ruckebusch Y, Demarquilly C, Farce MH, Journet M, editors. Nutrition des Ruminants Domestiques. INRA Editions, Paris, France, 1995, p. 601-648.
19. De Rosa G, Moio L, Napolitano F, Grasso F, Gubitosi L, Bordi A. Influence of flavor on goat feeding preferences. Journal of Chemical Ecology. 2002;28:269-281.
20. Dikmen S, Ustuner H, Turkmen II, Ogan M. Fattening performance and feed source preference of native Awassi lambs fed individually in a cafeteria feeding system. Tropical Animal Health and Production. 2009;41:485491.
21. Distel RA, Rodríguez Iglesias RM, Arroquy J, Merino J. A note on increased intake in lambs through diversity in food flavor. Applied Animal Behaviour Science. 2007;105:232-237.
22. Distel RA, Laca EA, Griggs TC, Demment MW. Patch selection by cattle: maximization of intake rate in horizontally heterogeneous pastures. Applied Animal Behaviour Science. 1995;45:11-21.
23. Distel RA, Villalba JJ. Use of unpalatable forages by ruminants: The influence of experience with the biophysical and social environment. Animals. 2018;8(4):56.
24. Dumont B, Boissy A. Grazing behaviour of sheep in a situation of conflict between feeding and social motivations. Behavioural Processes. 2000;49:131-138.
25. Duncan AJ, Gordon IJ. Habitat selection according to the ability of animals to eat, digest and detoxify foods. Proceedings of the Nutrition Society. 1999;58:799-805.
26. Duncan AJ, Ginane C, Elston DA, Kunaver A, Gordon IJ. How do herbivores trade-off the positive and negative consequences of diet selection decisions? Animal Behaviour. 2006;71:93-99.
27. Ekesbo I. Farm animal behaviour: Characteristics for assessment of health and welfare. CAB International, Wallingford, Oxfordshire, UK, 2011.
28. Favreau A, Ginane C, Baumont R. Feeding behaviour of sheep fed lucerne versus grass hays with controlled postingestive consequences. Animal. 2010;4:1368-1377.
29. Fedele V, Claps S, Rubino R, Calandrelli M, Pilla AM. Effect of free-choice and traditional feeding systems on goat feeding behaviour and intake. Livestock Production Science. 2002;74:19-31.
30. Forbes JM. Voluntary food intake and diet selection in farm animals. CAB International, Wallingford, UK, 1995.
31. Forbes JM. Feeding behaviour, voluntary food intake and diet selection in farm animals. 2nd ed. CAB International, Wallingford, Oxfordshire, UK, 2007.
32. Fraser AF, Broom DM. Farm animal behaviour and welfare. 3rd ed. CAB International, Wallingford, Oxon, UK, 1997.
33. Frost R, Mosley J. Diet selection in grazing animals. Rangelands Gateway, 2020.
34. Ginane C, Petit M, D'Hour P. How do grazing heifers choose between maturing reproductive and tall or short vegetative swards? Applied Animal Behaviour Science. 2003;83:15-27.
35. Ginane C, Duncan AJ, Young SA, Elston DA, Gordon IJ. Herbivore diet selection in response to simulated variation in nutrient rewards and plant secondary compounds. Animal Behaviour. 2005;69:541-550.
36. Gordon IJ. Animal-based techniques for grazing ecology research. Small Ruminant Research. 1995;16:203-214.
37. Grings EE, Short RE, Haferkamp MR, Heitschmidt RK. Animal age and sex effects on diets of grazing cattle. Journal of Range Management. 2001;54:77-81.
38. Hadjigeorgiou IE, Gordon IJ, Milne JA. Comparative preference by sheep and goats for Gramineae forages varying in chemical composition. Small Ruminant Research. 2003;49:147-156.
39. Harvey A, Parsons AJ, Rook AJ, Penning PD, Orr RJ. Dietary preference of sheep for perennial ryegrass and white clover at contrasting sward surface heights. Grass and Forage Science. 2000;55:242-252.
40. Hirata M, Kanemaru E, Tobisa M. Patch choice by cattle grazing tropical grass swards: a preliminary study. Applied Animal Behaviour Science. 2006;97:134-144.
41. Hirata M, Sakou A, Terayama Y, Furuya M, Nanba T. Selection of feeding areas by cattle in a spatially heterogeneous environment: selection between two tropical grasses. Journal of Ethology. 2008;26:327-338.
42. Hirata M, Yamamoto K, Tobisa M. Selection of feeding areas by cattle in a spatially heterogeneous environment: selection between two tropical grasses differing in accessibility and abiotic environment. Journal of Ethology. 2010;28:95-103.
43. Hofmann RR. Evolutionary steps of eco-physiological adaptation and diversification of ruminants: a comparative view of their digestive system. Oecologia. 1989;78:443-457.
44. Janis CM, Ehrhardt D. Correlation of relative muzzle width and relative incisor width with dietary preference in ungulates. Zoological Journal of the Linnean Society. 1988;92:267-284.
45. Kaitho RJ, Umunna NN, Nsahlai IV, Tamminga S, van Bruchem J, Hanson J, van de Wouw M. Palatability of multipurpose tree species: effect of species and length of study on intake and relative palatability by sheep. Agroforestry Systems. 1996;33:249-261.
46. Kalio GA, Oji UI, Larbi A. Preference and palatability of indigenous and exotic acid soil-tolerant multipurpose trees and shrubs by West African Dwarf sheep. Agroforestry Systems. 2006;67:123-128.
47. Kilonzo JM. Diet selection and nutrition of sheep and Grant's gazelles on Kapiti Ranch, Kenya. MSc Thesis, 2003.
48. Kyriazakis I, Anderson DH, Duncan AJ. Conditioned flavour aversions sheep: relationship between the dose rate of a secondary plant compound and persistence of aversions. British Journal of Nutrition. 1998;79:55-62.
49. Lynch JJ, Hinch GN, Adams DB. The behaviour of sheep: Biological principles and implications for production. CAB International, Wallingford, Oxon, UK, 1992.
50. Meier JS, Kreuzer M, Marquardt S. Design and methodology of choice feeding experiments with ruminant livestock. Applied Animal Behaviour Science. 2012;140:105-120.
51. Milne JA. Diet selection by grazing animals. Proceedings of the Nutrition Society. 1991;50:77-85.
52. Morand-Fehr P. Dietary choices of goats at the trough. Small Ruminant Research. 2003;49:231-239.
53. Newman JA, Penning PD, Parsons AJ, Harvey A, Orr RJ. Fasting affects intake behaviour and diet preference of grazing sheep. Animal Behaviour. 1994;47:185-193.
54. Ngwa AT, Nsahlai IV, Bonsi MLK. Feed intake and dietary preferences of sheep and goats offered hay and legume-tree pods in South Africa. Agroforestry Systems. 2003;57:29-37.
55. Norberg RA. To minimize foraging time, use highefficiency methods when food is abundant but low-cost methods during shortages. Ecology and Evolution. 2021;11(23):16537-16546.
56. Papachristou TG, Nastis AS, Mathur R, Hutchings MR. Effect of physical and chemical plant defences on herbivory: implications for Mediterranean shrubland management. Basic and Applied Ecology. 2003;4:395403.
57. Parsons AJ, Newman JA, Penning PD, Harvey A, Orr RJ. Diet preferences of sheep: effects of recent diet, physiological state and species abundance. Journal of Animal Ecology. 1994;63:465-478.
58. Penning PD, Newman JA, Parsons AJ, Harvey A, Orr RJ. Diet preferences of adult sheep and goats grazing ryegrass and white clover. Small Ruminant Research. 1997;24:175-184.
59. Pfister JA. Behavioral strategies for coping with poisonous plants. Grazing Behavior of Livestock and Wildlife. 1999;70:45-59.
60. Prache S, Damasceno J. Preferences of sheep grazing down conterminal monocultures of Lolium perenne and Festuca arundinacea. Applied Animal Behaviour Science. 2006;97:206-220.
61. Provenza FD, Villalba JJ, Bryant JP. Making the match: from biochemical diversity to landscape diversity. In: Bissonette JA, Storch I, editors. Landscape Ecology and Resource Management. Island Press, New York, 2002, p. 387-421.
62. Raufirad V, Ebrahimi A, Azadi H. Animal preference and external plant attributes: application of principal component analysis. Journal of Biodiversity and Environmental Sciences. 2015;6(1):22.
63. Rodríguez AB, Bodas R, Prieto N, Landa R, Mantecón AR, Giráldez FJ. Effect of sex and feeding system on feed intake, growth, and carcass traits of Assaf lambs. Livestock Science. 2008;116:118-125.
64. Rogosic J, Pfister JA, Provenza FD, Grbesa D. Sheep and goat preference for and nutritional value of Mediterranean maquis shrubs. Small Ruminant Research. 2006;64:169-179.
65. Rozin P. Psychobiological and cultural determinants of food choice. In: Silverstone T, editor. Appetite and Food Intake. Berlin: Dahlem Konferenzen, 1976, p. 285-312.
66. Rutter SM, Orr RJ, Yarrow NH, Champion RA. Dietary preference of dairy cows grazing ryegrass and white clover. Journal of Dairy Science. 2004;87:1317-1324.
67. Sandoval-Castro CA, Lizarraga-Sanchez HL, SolorioSanchez FJ. Assessment of tree fodder preference by cattle using chemical composition, in vitro gas production and in situ degradability. Animal Feed Science and Technology. 2005;123:277-289.
68. Scharenberg A, Arrigo Y, Gutzwiller A, Soliva CR, Wyss U, Kreuzer M, et al. Palatability in sheep and in vitro nutritional value of sainfoin, birdsfoot trefoil and chicory. Archives of Animal Nutrition. 2007;61:481-496.
69. Simitzis PE, Deligeorgis SG, Bizelis JA, Fegeros K. Feeding preferences in lambs influenced by prenatal flavour exposure. Physiology and Behaviour. 2008;93:529-536.
70. Stegelmeier BL, Davis TZ, Clayton M. Plant induced reproductive disease, abortion and teratology in livestock. Animal Reproduction Science. 2020.
71. Van Dorland HA, Wettstein HR, Aeschlimann G, Leuenberger H, Kreuzer M. Preference of dairy cows for ryegrass, white clover and red clover, and its effects on nutrient supply and milk quality. Archives of Animal Nutrition. 2007;61:371-389.
72. Van Dorland HA, Kreuzer M, Leuenberger H, Wettstein HR. Eating behaviour of dairy cows offered fresh or ensiled white clover, red clover and ryegrass to choose from or in a mixture. Applied Animal Behaviour Science. 2008;111:205-221.
73. Villalba JJ, Provenza FD, Hall JO, Lisonbee LD. Selection of tannins by sheep in response to gastrointestinal nematode infection. Journal of Animal Science. 2010;88:2189-2198.
74. Weigensberg I, Fairbairn DJ. Conflicts of interest between the sexes: a study of mating interactions in a semiaquatic bug. Animal Behaviour. 1994;48:893-901.