The Catching Arena is an open box made of black acrylic plastic, enclosed by white plastic curtains suspended from the ceiling and draped inside the arena walls. A wood collar supported a raised floor in the arena. This floor consisted of eight galvanized steel plates, divided into two parallel rows of four each, and the rows were separated by a wood divider. Each plate contained a 4 x 4 array of 4-cm drilled holes. Each hole was fitted with a cup that hung snugly from the cup’s rim. A plastic dish in the center of the divider acted as a feeder during cache sessions.

Maze Engineers offer the Caching Arena.

Price & Dimensions

Caching Arena

$ 1990

  • Acrylic Open Box: 118 x 179 x 47 cm
  • Floor: 8 galvanized steel plates 45 x 45 cm, divided into two parallel rows of four each.
  • Rows separated by a wood divider (29 x 179).
  • Each plate includes a 4 x 4 array of 4-cm drilled holes.
  • Each hole is fitted with a cup (4-cm interior diameter; 3.25 cm deep)



The Caching Arena is used to observe spatial learning and memory based on cache retrieval in rodents. Many animals such as rodents cache food items in their natural environments and leave them for later retrieval in thousands of widely scattered caches. These animals usually retrieve this stored food during food shortages, which can occur weeks to months after their initial placement (Hirsch, Kays, & Jansen, 2013). The Caching Arena is used to observe how rodents can remember these cached food sites. It consists of an open box with a raised floor consisting of steel plates. The steel plates are divided into rows and contain drilled holes in which fitted cups filled with sand are placed to serve as caching sites. During the task, the subjects must cache seeds and then recover them after a specific period. Recovery trials can either be conducted a day or ten days after cache session to observe the effect of retrieval interval on rodents’ long-term memory. In addition to retrieval intervals, the impact of landmarks on rodents’ learning and memory can also be evaluated since the Caching Arena is large enough to place multiple landmarks of different shapes and sizes. 

Other Apparatuses used to observe learning and memory in rodents include the Hole Board, the Hebb Williams Maze, the Lashley III Maze, and the Puzzle Box

Apparatus and Equipment

The Caching Arena consists of an open box made of black acrylic that measures 118 x 179 x 47 cm. White plastic curtains that are suspended from the ceiling and draped inside the arena walls enclose the box. The Caching Arena’s floor is raised and consists of eight galvanized steel plates supported by a collar. The steel plates measure 45 x 45 cm and are divided into two parallel rows of four each. The rows are separated by a divider measuring 29 x 179 cm. In the center of the divider, a plastic dish is placed, which acts as a feeder. Each steel plate contains a 4 x 4 array of 4 cm drilled holes in which a cup is fitted. The cup has an interior diameter of 4 cm and is 3.25 cm deep, and is filled with sand constituting 128 potential cache sites. 

Training Protocol

Clean the arena after every trial to avoid odor cues from previous trials. Appropriately light the arena. Place landmarks such as rocks, sticks, pine cones, and artificial cloth flowers in the Caching Arena. Place the landmarks in multiple places, such as the center of the divider, on the plates between cups, and around the plates’ edges. Randomly choose different landmarks for each trial. A tracking and recording system such as the Noldus Ethovision XT can be used to assist with observations.

The following is a sample protocol to test the effect of different visual environments and retrieval intervals on cache retrieval accuracy in rats using the Caching Arena.



Place 60 sunflower seeds in the feeder for 2-hour blocks. Place the subjects in the arena and allow them to cache the seeds in the sand-filled cups until all the seeds are cached or until 6 hours have elapsed. If the subjects don’t cache the seeds on day 1, conduct another caching session the next day for 6 hours. Return the subjects that successfully complete the caching session to the arena 24 hours later. Allow their caches in cups to remain intact but remove all the other seeds from the feeder, plates, or divider. 



Conduct habituation sessions in the same manner as pretest trials but allow each session to last for 2 hours. Conduct a retrieval session the following day for 80 min or until the subject recovers at least 60% caches. Conduct habituation trials until the subject makes two caches per session in at least two sessions with a minimum number of 4 trials given for each subject. Only allow subjects who make two caches in two sessions within seven habituation trials in the remainder of the experiment.


Caching Arena Test Trials

Place the subject in the arena. Allow the subject to cache the seeds for 20 minutes. End the trial if the subject makes a minimum of 2 caches or continue the trial for 35 or 50 minutes until two caches are made. If the subject fails to make at least 2 caches within the time frame, return it into its home cache and conduct another trial the next day. If the subject made one cache on the first day, cap the cache with a tightly fitting lid on the second day. Conduct trials for each of the following conditions in random order for each subject: delay between cache and recovery (1 day or 10 days) and two visual environment conditions (0 and 16 unique landmarks).


Caching Arena Recovery Trials 

Conduct recovery trials 1 or 10 days after test sessions. Replace the sites used as caches in testing trials with two seeds each. Place equal numbers of baited control sites containing two seeds each in the arena. Use the cups that the subjects used to store seeds in previous trials as the control sites. Conduct recovery sessions for 20 minutes. 

Literature Review

Investigation of the effect of different visual environments and retrieval intervals on cache retrieval accuracy in rats 

Barkley and Jacobs (1998) investigated cache retrieval accuracy in wild-caught Merriam’s kangaroo rats on the Caching Arena. The subjects were tested in two cache retrieval conditions (1 day or 10 days) and two visual landmark conditions (0 or 16 landmarks). Different landmarks consisting of natural shapes and textures were randomly chosen for each trial. In addition, extra apparatus cues were also placed in the testing room for the subjects to orient themselves in the arena easily. The experiment consisted of pretest, habituation, and test phases. A cache session and recovery session was conducted at each phase of the experiment. The feeder in the center of the arena was stocked with 60 sunflower seeds during cache sessions and was left empty during recovery sessions. Caches were determined by counting the number of seeds left in the feeder, the rat’s cheek pouches’ size, and whether the cups contained seeds. The presence of sand excavated from a cup and the presence of seeds in the cup indicated whether caching occurred at a site. During the pretest and habituation trials, 8 landmarks were used, and the subjects were checked after 2 hours to check if they at least made two caches. In contrast, in the test trials, the subjects were checked after 20 mins. Out of the 20 animals pre-tested, 12 animals successfully completed trials and moved onto habituation sessions. In the habituation trials, only 6 subjects passed trials and moved to the test phase. In the test phase, target and control caches sites were used in which the target sites consisted of caches used in the previous session that were replaced by two seeds. The control sites were cups that the animals stored seeds in the earliest trials and were placed around the arena and contained two seeds each. The control sites served to ensure that the recovery of target caches was not solely based on the search for seeds’ odor. The results indicated that there was no significant difference between seeds placed in each cache per condition. A significant effect of the visual environment on the number of caches made was observed in which the subjects made an average of one or more cache in the 16 landmark condition. In the test trial, it was observed that the subjects retrieved a more significant percentage of target caches than control caches under all conditions. A significant effect of retrieval interval and visual environment in the number of targets retrieved was observed. No significant difference between the two visual conditions in the percent of target retrieved was observed at the 1-day retention interval. However, a significant difference was observed at the 10-day retention interval. 

Request a quote

"*" indicates required fields