Maze performance and immunity costs in mice

It has long been recognized that learning can be impaired when animals are infected or immunodepressed. Several things could account for this, including deficits in exploratory activity or attention due to stress, metabolic burden or parasitic manipulation. With the exception of parasitic manipulation, such effects may reflect an adaptive trade-off in investment between learning and either maintaining steady-state immune responsiveness or mounting a response to infection. One prediction of the adaptive trade-off hypothesis is that increased learning effort, which might be reflected as an increased level of performance or maintaining a given level of performance in more difficult learning tasks, will lead to reduced immunocompetence and thus resistance to infection. We tested this prediction in outbred BKW mice using a simple maze learning task. Mice required to learn more variable food locations in a radial maze showed a reduced proportion of correct responses and a positive relationship between learning performance and the severity of a subsequent infection with the blood protozoan Babesia microti. While a general increase in total IgG concentration (a measure of peripheral immune responsiveness) over the experiment was damped in high variability treatments, this did not relate directly to learning performance. While neither learning performance nor infection correlated with intercurrent changes in steroid hormones (corticosterone and testosterone), infection was positively associated with pre-experimental corticosterone concentration, and learning performance weakly so. The results thus suggest some predisposition towards spatial learning and infection among mice with higher starting concentrations of corticosterone, but no modulation of corticosterone in relation to the learning experience itself.