Skip to main content
Comparisons and Troubleshooting

Differences in Stress Response in the Morris Water Maze and the Barnes Maze

By September 30, 2015February 20th, 2017No Comments

Last time on the Maze Engineers blog we covered sex differences in performance on the forced swim trial. Today’s special will also involve water-based tasks, namely the Morris water maze. Like last time, we’re going to compare and contrast facets of maze studies. This time, however, we’re going to compare mazes themselves, and to see if the water maze is more stressful than another, different maze: the Barnes maze.

Barnes Maze vs. Water Maze

Which is more stressful: water, or open spaces?

So what’s the Difference?

Intuitively, the names should automatically highlight the difference between both mazes, but don’t let their names fool you: there are in fact many similarities between the two. Here’s a brief rundown of both the Barnes maze and the water maze.

Barnes Maze

  • Used to test spatial memory and learning
  • Built according to rodents’ aversion to open spaces
  • Circular construction with 20 sockets arranged in a clock-like fashion
  • Built with visual cues in plain sight
  • Contains an escape hatch under one of the sockets
  • Performance is measured based on number of errors made
  • Other performance parameters include length of path taken to escape hatch

Morris Water Maze

  • Also used to study spatial memory and learning
  • Also circular in nature, but filled with water
  • The objective here is to swim to an unseen platform
  • This is done through a variety of strategies, including using visual cues (reference points)

While there is much in common between these mazes, there are several differences between them. For one, the Barnes maze doesn’t use aversive stimuli, which would make it the least stressful of the two, but is this really so?

The answer is, yes, and it comes to us from Harrison, Hosseini and McDonald (2009) over at Vanderbilt University.

The Study

To properly assess stress levels, Harrison et al. focused on collecting and measuring levels of the stress hormone corticosterone in mice, which was extracted following the test.

Before they were placed in either maze, they were tested for anxiety in an elevated plus maze and the light-dark task during the first week. For the second week, they were tested on social dominance using the tube test, in which the mice had to scare their cagemates out of the cage.

Following that, the mice were placed in one of three groups based on their performance in these tasks, and were tested for five days:

  • Barnes Maze Condition (BM)
  • Morris Water Maze Condition (MWM)
  • No Further Testing (NFT)

For the BM group, spatial learning was assessed on the amount of errors made, escape latency, and length of the path. For the MWM group, escape and path latency were also evaluated, along with swimming speed and swimming in the maze’s perimeter.


  • During testing in both the water and Barnes mazes the mice improved on spatial learning
  • Escape latency decreased after training, and by day 5 both groups in both mazes were reported to have spent the same amount of time in their environments
  • As expected, corticosterone levels varied between each group: the mice in the MWM group had the highest stress levels of the three, while the NFT group had the lowest
  • Corticosterone levels in blood plasma had no relation to any of the anxiety measures, but was a strong predictor for water maze performance
  • This suggests that stress does facilitate spatial learning. However, mice that took longer to swim were also highly stressed
  • Other behaviors such as passive floating and peripheral swimming may have also influenced stress (and corticosterone) levels

Overall, this evidence is correlational and still warrants further investigation, but the punchline remains: the Morris Water Maze is more stressful than the Barnes Maze.


Harrison FE, Hosseini AH, & McDonald MP (2009). Endogenous anxiety and stress responses in water maze and Barnes maze spatial memory tasks. Behavioural brain research, 198 (1), 247-51 PMID: 18996418

Close Menu