Factors Involved in Cache-Retrieval in Birds: A Review

Amanjit Dhatt

Writer’s comment: When Dr. Demory assigned us a review in English 104E (Scientific Writing), I did not initially have a topic in mind. I wanted the topic to be interesting and new enough so writing the paper would be fun as well as educational. Then I remembered learning in Dr. Clayton’s Animal Behavior class (NPB 102) about some birds’ amazing ability to store food in hundreds of places. I was particularly intrigued that these birds could accurately recover the food months after storing it. Writing this paper on the factors involved in birds’ cache-retrieval was quite a challenge, but Dr. Demory’s excellent instruction and valuable feedback on the rough drafts made the task easier. (I also extend special thanks to Dr. Clayton without whose enthusiastic help this paper would not have been possible.)
- Amanjit Dhatt

Instructor’s comment: One of the objectives of English 104E: Scientific Writing is for students to become familiar with scientific journal articles—their purpose, audience, and language. To that end, the course focuses first on reading and analyzing such articles and then, finally, writing a scientific review paper, in which students must synthesize recent research on a specific topic. In tackling cache-retrieval behavior, Amanjit worked her way through an impressive number of articles, going beyond the assignment’s requirements. This breadth meant she became an expert on the phenomenon; it also required an organizational strategy to make the material meaningful to her readers—no mean feat—and she succeeded admirably. The resulting paper demonstrates her understanding of cache-retrieval behavior and of the complex nature and purpose of scientific writing itself.
- Pamela Demory, English Department


     Food caching is a widespread phenomenon observed in many bird species. Many of these birds hoard food at dozens to thousands of sites during autumn and then successfully retrieve their caches in the winter, days, weeks, or months later. This remarkable phenomenon of accurate cache-retrieval, given the time gap of weeks or months between caching and recovery, has prompted many scientists to investigate the factors underlying this behavior. In this review paper, I will summarize and evaluate studies of two factors that contribute to successful cache-retrieval in birds: hippocampus involvement and spatial cues in cache-site choice. I will also discuss the possible role of food-dispersal pattern in cache-recovery.
     The hippocampus, a brain structure known to play an important role in spatial memory formation, is central to the examination of cache-retrieval in birds. Food-storing bird species have larger hippocampal volumes than their non-storing counterparts (Volman et al. 1997). There is evidence of enhanced hippocampal development in young birds exposed to caching experience versus the ones deprived of it (Patel et al. 1997). Also, decreased accuracy of cache-retrieval in birds with hippocampal lesions strongly implicates the role of the hippocampus in food hoarding and retrieval (Hampton et al. 1996a). Furthermore, food recovery is facilitated by birds’ use of certain local and global landmarks as spatial cues to cache and then locate their food (Gould-Beierle et al. 1996). Finally, patterns of cache-dispersal may provide further insight into the methods by which birds easily and correctly remember their cache-sites. For example, some species store food in clumps and this strategy may decrease memory load. A review of these factors will provide an integrated picture of the ways in which birds accomplish the enormous task of locating their hidden food in hundreds of places.


Association between cache-recovery and hippocampal volume
     The hippocampus is known to play a central role in spatial memory. As birds rely so heavily on memory for accurate cache recovery, several researchers have tested whether the hippocampus is important in memory for food caches. Hampton et al. found that the food-storing black-capped chickadees’ (Parus articapillus)hippocampal volumes were significantly larger than those of the non-storing dark-eyed juncos (Junco hyemelis)(1996b). These results are supported by a study of four species of woodpeckers in which red-bellied (Melanerpes carolinus)and red-headed (Melanerpes erythrocephalus)woodpeckers showed relatively larger hippocampal volumes than did their non-caching or minimally caching counterparts, hairy (Picoides villosus)and downy (Picoides pabescens)woodpeckers (Volman et al. 1997). In a comparative study of four food-storing species of corvids, Basil et al. also found hippocampal differences, indicating a positive correlation between hippocampal volume and degree of reliance on cached food, as well as duration of spatial memory. In this work, Clark’s nutcrackers (Nucifrage columbiana),not only had the largest hippocampal complex of the four species but also cached more seeds. According to a study by Kamil et al., Clark’s nutcrackers also retained spatial memory for cache-retrieval for longer periods of time (cited in Basil et al. 1996). In yet another study, red-bellied woodpeckers, which place their individual food items in dispersed distribution, had greater hippocampal volumes than did the red-headed woodpeckers, whose cache-sites are clustered (cited in Volman et al. 1997).
     Taken together, these findings demonstrate a close association of increased hippocampal size with cache-retrieval as well as degree and scatter of caching in birds. Research focusing on the question of what causes the hippocampal volumes to differ between the storing and non-storing birds provides insights into the relationship between the hippocampus and spatial memory involved in storing and recovering food.

Spatial learning induces hippocampal development
     Whether hippocampal enlargement precedes or follows the onset of storing behavior has received considerable attention. A study by Healy and Krebs reveals that hippocampal sizes are similar in food-storing magpies (Pica paca)and non-storing jackdaw (Corus monedula)nestlings. However, when nestlings leave the nest and begin caching, the number of hippocampal neurons increases in the former and decreases in the latter, thus leading to an observed difference in adult hippocampal volume between the two species. These results indicate that structural differences in the hippocampus are very closely related to behavioral changes requiring spatial memory tasks such as caching food (1993).
     This relationship is solidified by studies on the effects of spatial learning experience on hippocampal growth. For example, Patel et al. observed that initial hoarding and cache-recovery experiences are crucial to neuronal proliferation of the hippocampus. In a comparison of the hippocampal development in marshtits (Parus palustris) that were allowed to cache and retrieve food with that of the experience-deprived control group, the former showed a significant increase in hippocampal volume compared to the latter (1997). However, a causal relationship confirming the function of the hippocampus in caching remained to be investigated.

Hippocampal lesion studies
     Studies in which the hippocampus is lesioned and the subsequent food storing and retrieval behaviors are observed have been very important in defining the role of the hippocampus in spatial memory. One experiment tested the effects of hippocampal damage on the ability to recognize a stimulus by spatial location versus its color in black-capped chickadees and dark-eyed juncos. It was found that in both species lesions of the hippocampus impaired the spatial location ability to a greater extent than color memory when compared to the respective pre-lesion results (Hampton and Shettleworth 1996). In another study, Sherry et al. also concluded that hippocampal lesions disrupt learning of spatial discrimination of objects while leaving the black-white discrimination almost unaffected. Furthermore, damage to the hippocampus hampers the black-capped chickadees’ accuracy in cache-retrieval (1989). These results, combined with the fact that hippocampal development and size are highly correlated with the use of spatial memory, confirm the role of the hippocampus in enabling birds to recover their caches.

Spatial cues

     In addition to relying on the hippocampus, birds also use certain spatial cues to facilitate a successful search for caches. In a study by Herz et al., black-capped chickadees were allowed to cache in the presence of both distal (objects placed on the walls of the chamber, away from the caching area) and proximal (objects placed closer to the caching area) cues. The birds’ cache-recovery was unaffected when the proximal objects were removed prior to cache-retrieval, but it declined significantly in the absence of distal objects, which suggests that these birds are more likely to use global versus local cues for caching as well as cache-relocation (1993). Gould-Beierle and Kamil’s study of Clark’s nutcrackers produced similar results (1996). However, generalizations cannot be made about the specific types of spatial cues used by birds to remember their caches. While European jays prefer proximal and vertical objects (Bossema 1979), Clark’s nutcrackers rely more on tall landmarks to search their caches (Basil 1993) (cited in Gould-Beierle and Kamil 1996).

Influence of cache-dispersal patterns on cache-retrieval

     Patterns of cache-dispersal, in addition to spatial cues, may also enhance cache-recovery accuracy. Some hippocampal studies show that birds which distribute their caches over a wider area have relatively larger hippocampal volumes than those whose caches are clustered (cited in Volman et al. 1997). Even given the function of the hippocampus in spatial memory, these findings do not necessarily indicate that the former retrieve their caches more accurately than do the latter. In fact, as Barnea and Nottebohm suggest, birds probably use clustered caching to maximize their cache-retrieval accuracy while minimizing the energy expended in food-storage. Their study on the black-capped chickadees reported that these birds consistently clumped their caches in an area with a specific orientation relative to the food source. Barnea and Nottebohm hypothesize that in directional clumped caching, chickadees visit the storing area more than once before cache-retrieval. Thus repeated exposure to the general food storing area may help the birds easily remember their cache-sites. Such caching patterns may also reduce the birds’ memory load as they have to remember fewer directions and landmarks for their caches than do birds that scatter their caches in many directions (1994).


     From the research evidence so far, the role of hippocampus in the development of spatial memory essential for food caching and recovery is very clear. Furthermore, some birds use certain spatial cues and others may cluster their caches in specific directions to remember their cache-sites, thus facilitating easy and accurate cache-retrieval.
     However, research on spatial cues has used only adult birds to determine which species prefers what types of spatial cues. To clearly delineate the spatial cue-types for a specific species, experimental work needs to be carried out in which young birds of a species are raised to cache in the presence/absence of different spatial cues. The subsequent observations of these birds’ cache-retrieval capacities and degree of hippocampal development can provide insights into the types of cues used by the species in question.
     Experience in caching can have a great effect on birds’ food storing and food recovery behavior. However, control studies to understand the effect of caching experience on cache-dispersal patterns have not been done yet. Therefore, experiments that involve depriving some young birds of caching while allowing others to store food and then observing their food scattering patterns will shed some light on the influence of caching experience on the degree and directionality of cache-dispersal. Also, experiments comparing directional cache clustering species with the ones scattering food in many directions over wider areas should be conducted to test whether cache clumping in a particular direction enhances cache-retrieval.
     While the role of the hippocampus in caching and cache-recovery in food-storing birds has been clearly established, the hippocampal and other neural mechanisms underlying spatial memory formation still remain to be investigated. Future research directed towards this goal will enable us to understand exactly how caching birds recover their food. Moreover, it has been found that mammalian and bird hippocampal complexes are similar in structure and neuroanatomical organization (cited in Clayton 1998). Therefore, a model explaining the ways in which the bird hippocampus functions during caching and cache-recovery holds potential applications for humans. Such a model can help us understand human memory formation as well as memory deficits.

Literature Cited

Barnea, Anat and Fernando Nottebohm. 1994. Patterns of food storing by black-capped chickadees suggest a mnemonic hypothesis. Animal Behavior, 49, 1161-1176

Basil, Jennifer A., Alan C. Kamil, Russel Balda, and Katherine V. Fite. 1996. Differences in hippocampal volume among food storing corvids. Brain Behavior and Evolution, 47, 156-164

Clayton, N.S. 1998. Review: Memory and the hippocampus in food-storing birds: a comparative approach. Neuropharmacology, 37, 441-452

Gould-Beierle, Kristy L. and Alan C. Kamil. 1996. The use of local and global cues by Clark’s nutcrackers, Nucifraga columbiana. Animal Behavior, 52, 519-528

Hampton, Robert R. and Sara J. Shettleworth. 1996a. Hippocampal lesions impair memory for location but not color in passerine birds. Behavioral Neuroscience, 110, 831-835

Hampton, Robert R. and Sara J. Shettleworth. 1996b. Hippocampus and memory in a food-storing and in a nonstoring bird species. Behavioral Neuroscience, 110, 946-964

Healy, S.D. and J.R. Krebs. 1993. Development of hippocampal specialization of a food-storing bird. Brain Research, 53, 127-131

Herz, Rachel S., Liana Zanette, and David F. Sherry. 1993. Spatial cues for cache retrieval by black-capped chickadees. Animal Behavior, 48, 343-351

Patel, Sanjay N., Nicky S. Clayton, and John R. Krebs. 1997. Spatial learning induces neurogenesis in the avian brain. Behavioral Brain Research, 115, 115-128

Sherry, D.F. and S.J. Duff. 1989. Effects of hippocampal lesions on spatial memory and cache-retrieval. Journal of Experimental Biology, 199, 165-172

Volman, Susan F., Thomas C. Grubb, Jr., and Kristofor C. Schuett. 1997. Relative hippocampal volume in relation to food-storing behavior in four species of woodpeckers. Brain, Behavior and Evolution, 49, 110-120