Frequently Asked Questions about the Morris Water Maze

URL: http://www.poirrier.be/~jean-etienne/notes/mwm/index.php

Questions and answers initially collected/written on May 6th, 2009
Updated on May 24th, 2009 (3D representation of pool and lightning)

Since I published some videos of Morris Water Maze (MWM) experiments, I received questions about the set-up of the maze and its concept in general. I tried my best to answer them (if you have some more, feel free to ask). Before going on, please note I don't have any interest in any of the cited companies. OK, here are the questions ...

General questions

Q: What is the Morris water maze?

A: You can read this Wikipedia article.

Questions about the pool

Q: Is the maze higher than the person? The maze we have is on ground level therefore the animal will be able to see the operator. In your video can the animal see the operator?

A: Our maze is lower than the person for the simple reason it is easy to handle the animal in the maze (at the beginning and the end of the trial). As soon as the animal is released in the maze, the scientist (myself) bends so he is "sitting" below the maze and the animal can't see him. This can be seen on this video (around the 6-7th seconds). From some places in the room, the scientist can see the computer screen (the animal can never see it). But when the scientist is at the opposite side of the room, we designed a very simple system connected to the computer and Ethovision. It consists of LEDs connected to the computer parallel port. Ethovision can send signals to them and turn them on or off when, for example, the animal found the platform or is in the maze for more than x seconds. The device can be seen below (or on Flickr).

LED device

Q: Are you painting the water or is just that your pool is black and the water is just normal water?

A: I add non-toxic black painting to the water and the pool is painted in black too (because our rats are white). Of course, painting the water in black will make the part of the rat below the water surface not visible (or less visible). But when we just left the water non-painted, we have bad reflections from the lighting system (preventing the video-tracking system to work properly). "Metalic" borders of the maze that are out of the water are also painted in black.

Q: What paint do you use to colour the water?

A: For the paint, we are using a liquid Redimix poster paint 265 (Lefranc-Bourgeois).

Q: How do you heat the water? do you have an external circuit or just an internal heater with thermostate?

A: I heat the water with simple, external aquarium heating systems like the one on the picture below. We put 4 of them while filling in the pool with water and it takes approximately 2 hours to reach the ad hoc temperature (once the pool is full ; because we add water at 10.0°C). We remove them before beginning any experiment and the water stays at the correct temperature for about 0.75 - 1.0 hour. It was merely a question of money because 4 times 15€ is cheaper than building a proper thermostatic system for our pool (connections on the bottom of the pool, flat enough so rats can't touch it with their tails, easily cleaned, etc.).

heating system

Q: How is your room lighted? Do you use any special system of ilumination? I am currently trying to use halogen uplighters as the lights on the ceiling are too reflective on the pool when filled with water.

A: We don't use any special illumination system. Lights are coming from 4 simple neon tubes on the ceiling of the room. Between them and the pool we put long cardboard deflectors in order to avoid direct lightning on water. Indirect lightning is still present so the video tracking can work efficiently, see these videos (you can still see some reflection in the pool, see figure below). A better system would have been to put the lights on the floor, next to the pool.

3D representation of a Morris water maze pool with and without direct lights
3D representation a Morris water maze pool with and without direct light. A: view of the pool from the side without deflectors (direct light). B: view of the pool from the side with deflectors (indirect light). C: view of the pool from the top without deflectors (direct light). B: view of the pool from the top with deflectors (indirect light). Somethins like a rat is represented on each picture. Image C doesn't allow correct tracking of the animal (D does).

Question about the software

Q: What software do you use?

A: we use Noldus Ethovision 3 as software. At the time we started considering buying one videotracking system (around 2002-2003), it was the best one (we compared it with Smart and 2100 Plus from HVS Image). We didn't follow their respective evolutions since then.

Q: How do you find using Ethovision for the MWM? e.g easy, challenging?

A: I find Ethovision rather easy to use once you followed the initial training Noldus gave when they came to deliver/install the system (just the computer, Ethovision 3 and the camera ; they didn't sell maze at that time). The system is flexible enough to process different kinds of behavioural experiments. And it doesn't need to be "re-programmed" if you switch from one experiment type to the other all the time. In version 3 (we never tried the XT version), the data summary is very powerful but also rather difficult to master. Once you have a "template" presenting your data, I advise you to save it before trying something else because it may take some time to revert the way data is presented (but data is never altered, fortunately!). Apart from this minor difficulty, it's a very good software.

Q: Can you send me the unedited videos of MWM you made on YouTube so I can train my video tracking package on them?

A: Unfortunately, these videos are screen captures: if they have the tracking information, it means I recorded what the tracking software was showing. None of these videos were post-processed for tracking. So I can't send you a video with tracking without it.