Escape latency

The question

Next stop on the tour, and it's a busy one. Why does a short swim time not always mean a good memory? You're watching Morris Water Maze Software Guide, over at morriswatermazesoftware.com.

Meet the escape latency

Escape latency is the time from when the animal is released to when it finds the platform, or to when the trial ends if it never does. It is the oldest and most reported water maze score. Latency falls as animals learn, but it also falls when they swim faster, stop hugging the wall, or adopt a non-spatial habit such as chaining. That is why latency is a first look at learning, not the whole answer.

How it runs

Here's how it plays out. First, start the clock at the moment of release, from a recorded start point. Next, stop the clock when the animal stays on the platform for the set dwell time, or when the trial limit is reached. Then, score animals that never find the platform at the trial limit, and note that they failed. Finally, plot latency across days and start points, not just as a single group average.

Strengths and limits

So what's in it for you? It is simple to collect and easy to compare with published work. It gives a clear learning curve when things go well. Labs have reported it for decades, so readers expect it. Where does it bite? Swim speed, floating and wall-hugging all change it without any change in memory. A short latency can come from a habit such as circling at the right distance from the wall.

What to measure

Now for the numbers, my favourite bit. Mean latency per day: the classic learning curve, best shown with the spread across animals, not only the mean. Latency to a dwell on the platform: stops a brief touch from counting as a find. Trials at the time limit: how often the animal never found the platform. Latency beside path length and speed: so a motor change is not read as a memory change. A treated group reaches the platform ten seconds sooner than controls by day four. Then you look at speed: they also swim faster. Path length is the same. The latency gain was motor, not spatial.

The takeaway

Short version: latency to a dwell on the platform and its neighbours tell you how the task was done, not just how fast. The full transcript is on this page, with sources.