AbstractThe importance of habitat type for the diversification of spatial strategies and telencephalon morphology was investigated in threespine stickleback fish (Gasterosteus aculeatus). Field-caught samples from sea-run (ancestral) and ecologically diverse freshwater (derived) populations were studied. Freshwater habitats included shallow, structurally complex lakes with benthic-feeding stickleback, and deeper, structurally simple lakes with limnetic (planktivorous) forms. While stickleback employ a variety of learning strategies to navigate within their environment, natural variation of spatial learning was studied because, unlike other strategies, it covaries with the size of a neuroanatomical structure. Spatial learning in relation to inferred ecology, ancestry, and experience was investigated using a T-maze. Benthic and limnetic stickleback populations were compared, and benthics exhibited superior spatial learning compared to limnetics. In another study, a sea-run population was used to infer the ancestral condition for spatial learning in stickleback. Spatial learning was probably present in the ancestor and retained in freshwater populations as an adaptation for benthic foraging. However, lab-bred lacustrine fish performed poorly compared to their field-caught counterparts, indicating that experience is important for spatial learning. Using field-preserved stickleback populations, I tested the hypothesis that relative size (adjusted for overall brain size) of the telencephalon is larger in benthics that occupy spatially complex habitat compared to limnetics from habitats with less structure. Contrary to expectations, field-preserved benthic populations did not consistently have larger relative telencephalon sizes than limnetic populations. However, the telencephalon of field-preserved sea-run and benthic populations was more convex laterally than that of limnetics. Although relative telencephalon size was not always larger in benthics compared to limnetics, convex telencephalon shape may indicate enlargement of the dorsolateral region, which is homologous to the tetrapod hippocampus, and greater relative size of the hippocampus is associated with superior spatial learning. Building on these results, the importance of genetic and environmental factors to telencephalon morphology was studied. Field-preserved, lab-held (i.e., field-caught fish held in aquaria for 90 days), and lab-bred fish from benthic, limnetic, and anadromous populations were compared. An ecotypic pattern for telencephalon shape differences that was similar to previous results was detected in field-preserved and lab-held fish, but these differences disappeared in lab-bred fish. Relative telencephalon sizes of field-preserved fish were larger than those of lab-held and lab-bred ones. Taken together, these results suggest that experience, like spatial learning, is important to telencephalon morphology. Thus, freshwater threespine stickleback appear to possess considerable telencephalon plasticity that may have been retained since the ancestor.