Scatterhoarding birds that cache food items have become an important model system for the study of spatial memory and its correlates in the brain. In particular, it has been suggested that through adaptive specialization, species that cache food have better spatial memory and a relatively larger hippocampus than their non-caching relatives. Critics of this approach, dubbed neuroecology, maintain that neither of these hypotheses has been confirmed. Here, we review the evidence pertaining to a correlation between food-storing capability and the relative volume of the hippocampus. Hippocampal volume has been related to food-storing behaviour in comparisons between species, within species, or within individuals, but the evidence is not consistent. There are several possible reasons for this inconsistency, including: (1) food-hoarding birds may not always use memory for retrieval, (2) there may be systematic differences between data from North American and Eurasian species that affect the analysis, and (3) sample sizes have in many cases been too small. In addition, both the independent variable (degree of food-hoarding specialization) and the dependent variable (relative volume of the hippocampus) are not clearly and consistently defined. Alternatively, it is possible that the neuroecological hypothesis is false. Systematic empirical research is necessary to determine whether or not food-storing birds have evolved adaptive specializations in brain and cognition.