Feeding is not one fixed behavior across a species
A wild fallow deer population shows that willingness to seek food from humans is not random. Some individuals repeatedly take risks to beg and obtain food, while shyer individuals avoid human contact and receive none, even though all were equally exposed to people from birth and lived in the same herds [5]. The bolder deer also intercepted food offerings early, which prevented others from getting supplemental feeding at all [5].
That matters for captivity because it means a group of wild animals can contain individuals with very different food-seeking tendencies. A shy animal that avoids novelty or human contact in the wild may be the one most likely to hold back from unfamiliar food in an enclosure, while a bold animal may adapt quickly. The deer study measured begging from humans, not eating in captivity, so it supports a general point about individual variation rather than a direct explanation of refusal.
Captivity removes the work of finding food
One critique of zoos argues that keeping wild animals in captivity deprives them of liberty, including the right to obtain food [4]. That is a philosophical argument about welfare, not a feeding experiment, and it does not measure appetite or food intake. Still, it points to a concrete difference: in the wild, an animal searches, chooses, and works for food; in captivity, food is usually provided.
If feeding is tied to active seeking rather than to the mere presence of food, then a captive animal faced with a food bowl has lost the sequence that normally leads to eating. The evidence here does not show that this loss causes refusal, but it is consistent with refusal being a response to changed conditions rather than a sign that the animal is simply not hungry.
Stress can fade as animals settle, but the evidence is indirect
Wild black rats studied under semi-captive conditions, with food freely available and social contact, showed no signs of pain or stress, and their sleep patterns stabilized within two days [3]. Feeding behavior was monitored for over a month in that setup [3]. This is a study of sleep recording, not of food refusal, so it cannot tell us why an animal might stop eating. What it does show is that a carefully designed captive setting can keep wild animals feeding and behaving without obvious distress.
The implication is limited but useful: when wild animals are given free food, social contact, and a setting close to their natural environment, the conditions that might suppress feeding are reduced. The rats' stable feeding does not prove that stress causes refusal elsewhere, but it is consistent with the idea that unfamiliar or restrictive captivity is part of the problem.
What the evidence cannot settle
Two of the supplied papers are only partly relevant. One examines bait shyness, the learned avoidance of toxic food in wild animals such as coyotes, which is a different phenomenon from refusing familiar food in captivity [1]. The other compares micronutrients and feeding patterns in wild versus captive donkeys but does not examine refusal to eat [2]. Neither can be used to explain a captive animal's appetite loss.
So the honest answer is that the supplied evidence describes pieces of the picture, not the whole. It shows that food-seeking varies between individuals [5], that captivity changes how food is obtained [4], and that wild animals can feed normally in a well-designed captive setting [3]. It does not establish which of these matters most, or whether refusal is mainly stress, unfamiliarity, loss of control, or something else.
Sources used in this answer
Conditioning food—illness aversions in wild animals: Caveant Canonici
The paper examines learned avoidance of toxic food (bait shyness) in wild animals and does not address why wild animals might refuse food in captivity.
Vitamin E and selenium concentrations of wild donkeys and their diets in the extreme desert ecosystem of Death Valley, California, versus captive donkeys
The paper compares micronutrient concentrations and feeding patterns between wild and captive donkeys but does not explore refusal to eat in captivity.
Bridging the gap between lab and field sleep studies: a proof-of-concept for studying wild rats in semi-captive environments
Wild black rats in a semi-captive setup with free food and social contact showed no signs of pain or stress, and their sleep patterns stabilized within two days while feeding was monitored for over a month.
Against zoos
The paper argues that captivity deprives wild animals of liberty, including the right to obtain food, though it does not investigate feeding behavior or appetite.
Artificial selection in human‐wildlife feeding interactions
In a wild fallow deer population, willingness to beg for food from humans was a repeatable individual trait, with bolder deer obtaining more food and shyer deer avoiding human contact and receiving none.