Visual environment and delay affect cache retrieval accuracy in a food-storing rodent

Abstract

Many scatter-hoarding species use spatial memory to relocate their food caches. Two factors can affect spatial memory: the availability of landmarks in the environment, and the latency between learning and recall. Using a 2 x 2 factorial design, we determined the effect of these factors on cache retrieval accuracy in Merriam's kangaroo rats (Dipodomys merriami). Kangaroo rats cached seeds in an arena under conditions that varied by retrieval interval (lor 10 days) and by number of landmarks (0 or 16 landmarks). After 1 day, they recovered equal proportions of caches in both landmark conditions. After 10 days, they recovered more caches in the 16-landmark condition than in the bare condition. This is the first study to show that landmarks are necessary for accurate cache recovery after long delays. This result is consistent with maze studies that have shown that landmarks reduce proactive interference and that delay increases proactive interference.

Memory in Food Caching Animals

“…If landmarks are important, this should produce substantial decrements in the ability to relocate caches, and it does (e.g., nutcrackers, Balda and Turek, 1984;parids, Herz et al, 1994). Barkley and Jacobs (1998) took a slightly different approach, allowing Merriam's kangaroo rats to cache and recover with either no or with many landmarks present. While the number of landmarks had no effect after a 1-day retention interval, there were large effects after 10 days.…”

An Invasive Grass and a Desert Adapted Rodent: is there an Effect on Locomotory Performance and is it Modified by Prior Experience or Familiarization?

“…This result was contrary to my predictions. If the difference is attributed solely to spatial cognition (i.e., the non-Veldt animals learned the physical locations of the tussocks) (Langley 1994;Sherry et al 1992;Barkley and Jacobs 1998;2007), Veldt and non-Veldt performance should have also been the same in the field, since the particular templates of Veldt grass they encountered were novel to both groups. In other words, both groups should exhibit equal levels of spatial cognition.…”

“…I therefore used running performance in novel habitat as an analogy for running performance during a dispersal phase. However, a kangaroo rat in an established home range encounters familiar habitat patches regularly (high spatial cognition has been demonstrated in kangaroo rats [Langley 1994;Sherry et al 1992;Barkley and Jacobs 1998;2007], suggesting they have high cognizance of the physical landscape in their home ranges). Therefore I also explored kangaroo rat performance in Veldt grass in a familiar setting, as a proxy for running performance in an individual's home range.…”

“…Based on previous spatial cognition studies (Langley 1994;Sherry et al 1992;Barkley and Jacobs 1998;2007), I…”

“…I predict no interaction between habitat of origin and the Veldt grass density treatment, however, I do predict a trend of amelioration in performance from field (unfamiliar habitat patch) to lab trials (familiar habitat patch) for kangaroo rats of both habitats of origin. Based on spatial cognition studies on kangaroo rats (Langley 1994;Sherry et al 1992;Barkley and Jacobs 1998;2007), the animals should learn the landmarks in their environment and be able to navigate more efficiently through them with experience.…”