A modified Barnes maze for an accurate assessment of spatial learning in mice

Background: The Morris water maze (MWM) and the Barnes maze (BM) are among the most widely-used paradigms for assessing spatial learning in rodents, with specific advantages and disadvantages for each apparatus. Compared with the intense water-related stress exerted during the MWM, the BM exhibits a milder light-induced stress, while suffering from biasing animals towards non-spatial strategies such as serial search, a heuristic non-spatial search strategy. To overcome this problem, we have developed a modified Barnes maze (MBM) apparatus that recapitulates natural environments more accurately without inducing undesirable exploration strategy bias. New method: Apparatus. A circular 122 cm-wide table with 40 randomly placed holes. One target hole is leading to an escape chamber. Task. Three target locations were examined, varying in their distance from the center. C57BL6/j male mice were given three trials per day to find the target. Following acquisition, a probe test was performed by removing the escape chamber. Results: Spatial-encoding-depended reduction in latency to reach the target was observed, along with improvement in path efficiency with test progress. Mice tested with peripheral and distal targets outperformed mice tested with a central target. A robust exploration pattern was identified in the probe test. Comparison with existing method: The MBM mimics natural environment to a higher degree of accuracy than the BM, without eliciting bias towards non-spatial searching strategies. Conclusions: Spatial learning in the MBM is a target-location sensitive process, providing flexibility in task difficulty. Along with overcoming biases towards non-spatial strategies, the MBM represents an improvement over the well-validated BM.