Carnivores: Provide food on a random temporal schedule

How is the evidence assessed?
  • Effectiveness
  • Certainty
  • Harms

Study locations

Key messages

  • Three replicated, before-and-after studies in Switzerland, Ireland and Canada, UK, Ireland, Namibia and South Africa and one replicated, controlled study in Ireland found that an unpredictable feeding schedule reduced the frequency of stereotypic pacing behaviours in tigers and cheetahs.
  • One replicated, before-and-after controlled study in the USA found that an unpredictable feeding schedule increased territorial behaviour in coyotes but did not affect travelling or foraging.
  • Two before-and-after studies in Switzerland and the USA found that an unpredictable feeding schedule increased behavioural diversity in red foxes and alertness in a black-footed cat.

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A small replicated, before-and-after study in 1998 of Amur tigers Panthera tigris altaica in an outdoor enclosure in a zoo in Switzerland [1], found that feeding boxes with random opening times reduced the frequency of stereotypic pacing. When housed separately and fed at a random time and place of the enclosure, time spent stereotypic pacing by the female tiger (1%) was reduced compared to a conventional feeding schedule (16%) but not by the male (random placement: 0%; conventional feeding: 3%). When housed as a pair, stereotypic pacing was reduced in both tigers (0%) compared to conventional feeding (female: 7%; male: 10%). Two tigers were studied both when housed alone and as a pair. Several electronically controlled feeding boxes were installed at different places within the outdoor areas of the enclosure. Before 09:00 h, meat was distributed to all boxes and the doors were closed. The tigers were only able to open the boxes during two random 15 minute periods between 09:00 h and 17:30 h when the doors were unlocked. Tigers were fed on three day feeding regimes including a three-day baseline (fed at 14:30 h for two days and one fast day), three days conventional feeding (fed at 14:30 h) and lastly box feeding. Behaviour was recorded for six hours on the third day of each regime using focal sampling. The study was replicated when housed as a pair.

    Study and other actions tested
  2. A replicated, before-and-after controlled study in 2009 of coyotes Canis latrans in a research centre in the USA [2], found that when fed using automated feeders on an unpredictable schedule, marking and howling behaviour increased compared to a predictable schedule. Frequency of marking behaviour (118 events/observation session) and howling behaviour (81 events/observation session) was higher when fed using an unpredictable regime compared to a predictable regime (marking: 42; howling: 24). There were no differences in time spent foraging, travelling, resting or standing. Twelve coyotes were housed individually in 0.1 ha experimental pens. Using automated feeders, one group of coyotes were fed daily on a predictable schedule at 08.00 h and 08.05 h and one group were fed twice daily on an unpredictable schedule. Observations included both pre- and post-feeding activity and non-feeding times using continual focal sampling for two hours per individual per day for ten days.

    Study and other actions tested
  3. A small before-and-after study in 2005 of red foxes Vulpes vulpes in a wildlife park in Switzerland [3] found that when foxes were provided with unpredictable automated feeds, behavioural diversity and activity increased compared to scheduled feeding but not in relation to other feeding enrichment methods. Behavioural diversity (Shannon index: 2) and time spent in active behaviours (45%) increased when fed unpredictably, compared to predictable feeds before (Shannon index: 1.5; activity: 14%) and after (Shannon index: 1.8; activity: 26%) enrichment was presented. Activity was defined as all behaviours except resting and sleeping. Four adult foxes, housed together, were fed daily except Saturdays, on 400 g of meat, 200 g of fruits and 200 g of dried dog food, raisins, sunflower seeds and nuts. Behaviour was observed for one-hour sessions four times a day. During each observation hour, each fox was continuously observed for 15 minutes to assess behavioural diversity while instantaneous scan sampling at 2.5 minute intervals was used to assess general activity. Data were collected for five days in six conditions: 1) scheduled feeding times; 2) electronic feeders, each randomly dispensing 1/3 of all meat feed between 10:00 h and 18:00 h; 3) electronic feeders and a self-service food box; 4) electronic feeders plus scattered and hidden food; 5) electronic feeders and an electronic dispenser which dispersed food around the enclosure; and 6) a second period of scheduled feeding times

    Study and other actions tested
  4. A replicated, before-and-after study in 2010 of cheetahs Acinonyx jubatus in a wildlife park in Ireland [4] found that varying the time of feeding decreased pacing behaviour. Pacing frequency was reduced when food was temporally varied (0.02 mean proportion of scans) compared to predictable feeding (0.17 mean proportion of scans). Ten cheetahs were housed in five enclosures (solitary and mixed housing). Instantaneous scan sampling was used at 5-minute intervals and a total of 48 scan samples were carried out per enclosure each day. Eight days of data were collected per enrichment technique, including eight baseline days. Temporal variation included feeding at an alternative time to 16:00 h.

    Study and other actions tested
  5. A replicated, controlled study in 2012 of cheetahs Acinonyx jubatus in Canada, UK, Ireland, Namibia and South Africa [5] found that an unpredictable feeding schedule decreased the levels of stereotypic pacing compared to a predictable regime (results of statistical models). Increasing size of enclosure also decreased stereotypic pacing, whilst being solitary and having the ability to view other cheetahs in adjacent enclosures increased it. Factors that did not have a significant effect included visual barriers, presence of raised areas, visitor numbers, sex, vehicle disturbance, presence of enrichment, age and diet diversity. One hundred and twelve cheetahs in 88 enclosures, maintained in nine zoological institutions were studied. Data were collected on stereotypical pacing using instantaneous scan sampling with five-minute intervals. Each day was divided in to two-hour periods consisting of eight 15-minute time periods.

    Study and other actions tested
  6. A replicated, before-and-after study in 2010 of cheetahs Acinonyx jubatus in a wildlife park in Ireland [6] found that varying the time of feeding decreased levels of locomotion and stereotypical behaviours and increased time devoted to other behaviours. In two out of five comparisons the proportion of scan samples in which locomotion was observed was lower (0.13–0.15) than under a fixed schedule (0.22) and in one out of five comparisons the proportion of scan samples in which stereotypical behaviour was observed was lower under a varied (0.03) than under a fixed schedule (0.09). However, in four out of five comparisons the proportion of scan samples in which ‘other’ behaviours were observed was higher under a varying (0.15–0.22) than under a fixed feeding schedule (0.05–0.09). Eight cheetahs housed in five enclosures were studied, three housed solitarily, two males housed together and one female with two cubs. Cheetahs were fed whole rabbits or chickens once daily at the same time, for six days a week. Data was collected over 16 days, during which food was temporally varied in eight randomly selected days. Instantaneous scan sampling was used every five minutes in varying degrees of sampling effort (ranging from 24 to nine scan samples) within a four-hour time period. Behavioural categories included exploratory, inactive, locomotion, stereotypical and vigilance. ‘Other’ behaviour category included aggression, allo-grooming, feeding, playing, standing and vocalisation.


    Study and other actions tested
  7. A before-and-after study in 2013 of a black-footed cat Felis nigripes in a zoo in the USA [7] found that when feeding schedule was unpredictable, alertness increased compared to the baseline of no enrichment. Alertness increased (24%) compared to a baseline of no enrichment (16%). Additionally, locomotion (48%) and investigatory (4%) behaviours increased and alertness decreased (24%) compared to when diet was changed from dry and wet commercial cat food and mice to a low starch diet (locomotion: 28%; investigating: 1%; alertness: 44%). Instantaneous focal sampling was used every 30-seconds during 20-minute periods to record behavior of an individually housed cat. Data was collected at least once per week over 10 months. Four treatments included: 1) baseline, 2) random feeding times twice daily, 3) increased exhibit complexity, 4) changed diet to completely wet food (lower starch).

    Study and other actions tested
Please cite as:

Jonas, C.S., Timbrell, L.L., Young, F., Petrovan, S.O., Bowkett, A.E. & Smith, R.K. (2020) Management of Captive Animals. Pages 527-553 in: W.J. Sutherland, L.V. Dicks, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2020. Open Book Publishers, Cambridge, UK.


Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Management of Captive Animals

This Action forms part of the Action Synopsis:

Management of Captive Animals
Management of Captive Animals

Management of Captive Animals - Published 2018

Captive Animal Synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.

Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust