Note: This article is truncated from the original.
For the full version of the Magnusson article, download our newsletter.

by Kathy Magnusson, DVM, PhD

The brain changes as we age. I am interested in finding ways to prevent or repair some of these changes, especially age-related declines of learning and memory.

On the other hand, I am not as young as I used to be, so you could also say that I am hoping to figure out how to improve memory problems before I forget what I am researching!

More than just a Swim Test

My laboratory uses specific tests in rodents to probe certain brain functions. Each test is designed to test one particular ability. For example, a test for memory in a laboratory animal is completely different than a test for anxiety. 

Memory tests often involve repeated trials to find food in an environment or to locate an object. Animal learn a particular location or task during a training phase, and then we can change something about the test and measure their cognitive flexibility – how well they can cope with that change.

As you might imaging, adaptability to new situations tends to decline as the animal gets older.

For different aspects of memory that involve spatial learning, we use the Morris Water Maze. In short, this is swim test for mice. We put them in a round pool filled with water that contains a platform. In order for the animals to get out of the water (which they are very eager to do), they must swim to the platform.

The difficulty is that the platform is hidden just under the surface of the water. The water is made opaque, so they can’t see the platform and swim toward it. When the animals are placed in the pool, they swim around until they find the platform. Every subsequent time that they are placed inside the pool, they will do the same thing: search until they find the platform.

To help the animals orient themselves in the round pool, objects or distinctive shapes are placed around the pool edge. If the animal finds the platform in a particular location, it will use the objects as landmarks to locate it faster on the next trial.

Once the mice learn that the objects tell them where the platform is, the faster the animal will swim directly to it. If the platform is removed, animals with a good memory will still search that area first before looking for the platform in other areas. Removing the platform helps us to see whether the animal knows where the platform is supposed to be and is not just swimming fast and running into it randomly.

We use this swim test extensively to link changes in the brain to memory declines with age. We can identify some older mice that perform almost as well as younger animals, while others perform much worse. Older animals with good memories have distinctly different protein markers in their brain than their poor-performing counterparts. This suggests they can compensate for the natural changes that occur with age.

Bringing it Home

Although research in rodents is the foundation of much of the scientific work on memory, there is a big push to bring this work out of animals and into people. To this, we have been developing a water maze for testing human subjects.

Unlike the rodent test, this is not performed in a pool of water but in a virtual computer environment. Otherwise, this task is designed to be similar to the water maze task that we use to assess memory in mice.

Using this ‘virtual water maze,’ we find that young adults learn the test in a very similar pattern as young animals. Like the mice in the pool, the participants use the landmarks on the computer screen to navigate the environment. Over repeated tests, their time to find the hidden platform quickens.

Older adults appear to learn much slower in water maze, even slower than the animal testing would predict.

Why? Well, there are several possibilities.

One is that older adults are not comfortable navigating a computerized environment while most young adults have been ‘training’ in this environment since they were children.

Another possibility is a motivation factor. Mice in a pool want to get out of the water as soon as possible. Navigating a computer environment means no reward for success, and no risks in failure.

However, we are able to divide participants into good performers and poor performers, just like we do for the mice. We can then monitor active brain regions that might be important to performance using equipment such as an fMRI.

We believe that using a virtual test will enhance our ability to transition from animals to humans. We are currently using the virtual water maze in a study of cognitive function in veterans from different war eras and in another study of older men taking multivitamin supplements.


Zamzow et al. eNeuro 6 (2019) doi: 10.1523/ENEURO.0310-18.2019

Zhong et al. Behavioral Neuroscience 131 (2017) doi: 10.1037/bne0000219