Virtual Morris water maze

The question

Here's a puzzle worth a few minutes: Can we put a person in the same test we give a rat, without the water? You're watching Morris Water Maze Software Guide, over at morriswatermazesoftware.com.

Meet the virtual Morris water maze

The virtual Morris water maze recreates the pool and its room as a 3D environment that a person explores on a screen or in a headset. A hidden target replaces the platform. Because the logic is the same as the rodent task, it's a bridge between animal and human studies of spatial memory, ageing and disease.

How it runs

Here's the routine, in order. First, build a virtual arena with distinct distal cues and a hidden goal; keep the layout fixed across sessions. Next, give the participant a short practice phase so the controls aren't what you're measuring. Then, run learning trials from varied start points, then a probe trial with the goal removed. After that, record position and view direction many times a second, with time stamps. Finally, in immersive systems such as HVS 6D VR, the person can physically walk or turn, which brings body-based cues back into the task.

Strengths and limits

What's it good for? The same analyses can be applied as in the rodent pool, so findings translate. No water, no stress from swimming, and it's quick to run. Cues and goals can be moved precisely for reversal or cue-conflict tests. Now, the small print. Desktop versions remove body movement cues, which matter for real navigation. Gaming experience and comfort with controls vary a lot between people. Some people get motion sickness if the display isn't well designed.

What to measure

Now for the numbers, my favourite bit. Heading error: the angle between the first movement and the ideal line to the goal, a clean readout of whether the person knows where to go. Path efficiency ratio: actual path length divided by the direct path; one is perfect. Proximity measures: average distance to the goal over the trial, comparable across humans and rodents. Time slices: splitting a trial into segments to see a strategy change mid-trial. Dual-strategy and reversal analyses: whether the person follows a moved cue or the remembered place. Imagine an older adult who reaches the goal as fast as a student, but whose first heading is 40 degrees off every time. The latency hides it. The heading error shows it.

The takeaway

If you remember only one thing, make it this: path efficiency ratio and its neighbours tell you how the task was done, not just how fast. Everything we covered is written up below.