ANY-maze Help > The ANY-maze reference > The Protocol page > The elements of a protocol > Tracking > Tracking options

Tracking options

Introduction

ANY-maze includes some tracking options which are designed to address some specific situations which can arise in certain apparatus and/or tests. These are set using the Tracking options element of the protocol.

Tracking when the animal will already be in the apparatus at the start of the test

Before choosing to use this option, you should check whether including an accustomisation period in your tests would be a better solution.

In most experiments, you will put the animal into the apparatus at the start of the test - this is the case for the plusmaze, water-maze, etc. However, you might encounter a situation in which you want to start the test with the animal already in the apparatus. For example, you might want an animal to be in the apparatus for 24 hours before you actually perform the test.

As you may know, ANY-maze works by comparing images of the apparatus at different times, and it uses the difference in the images to detect the animal. At the very beginning of a test, the system needs to have an image of the apparatus with no animal in it at all - this means it can quickly determine the animal's position and start tracking without any delay (it can in fact start tracking even if the animal is present, but this can lead to inaccuracies in the first few seconds of the test).

So, the system needs to have an image of the apparatus with no animal in it, and normally it captures this image for itself before you start a test. However, if the animal will already be in the apparatus, then obviously this isn't going to work and so a different method must be adopted - that's what this option is for.

If you switch this option on, then before you can begin a test ANY-maze will ask you to 'capture' an image of the empty apparatus. You do this by clicking the button which will be shown above the apparatus image in the Tests pane. The important point about this is that you can capture this image at any time before the test - and after you've captured the image, you can exit ANY-maze or even switch your computer off.

So in the case of the example I gave above, in which the animal will be in the apparatus for 24 hours before you perform the test, you would simply capture the image before putting the animal in the apparatus. You could then switch off the computer for the 24 hour period, and then switch it on again when you want to run the test.

Actually performing a test in this situation is just the same as normal, except that the 'Auto-start' feature is usually unnecessary as the animal is already in the apparatus. So all you have to do is click Start and the test (and the tracking) will begin. By the way, you can use auto-start if you want to - this is useful if you need to make some kind of alteration to the apparatus at the start of the test. For example, if you need to open a door in the apparatus then you could use auto-start so that when you walk away from the apparatus after opening the door, the test would start automatically.

Retaining the background picture between consecutive tests of the same animal

As explained above, when manually 'capturing' images of the empty apparatus, you have to click the button before each test, and you have to do this when there is no animal in the apparatus. This is usually not a problem for the first test in an experiment, but what if you intend to run a series of tests on the same animal, and you don't intend to remove the animal from the apparatus between one test and the next?

Clearly, in this case you won't be able to 'capture' a new empty-apparatus image before each test, so instead you can select the option to Retain the background picture from one test to the next. This will cause ANY-maze to automatically use the background image from one test as the 'captured' background image for the next test, just as if you had 'captured' the image yourself.

There are a few points to note about this:

•

You can 'over-ride' the automatically retained background image by clicking the    button in the usual way.

Not trying to adapt to lighting changes that occur during tests

Usually if ANY-maze detects an overall change in the light levels during a test it will reset the tracking system, because otherwise the tracking will simply stall (as the apparatus background will have changed significantly). This reset usually means that the tracking stutters for a moment then starts tracking the animal again.

However, some lighting changes, such as a rapid change to be much brighter followed immediately by a rapid change back to the original light levels, can actually cause the exact problem the reset is intended to resolve, i.e. ANY-maze stops tracking the animal. This sort of issue is most commonly seen when a door is opened and closed, momentarily making the apparatus more brightly illuminated. To resolve this you can tell ANY-maze not to try to adapt to lighting changes, in which case it will not reset the tracking system when light levels change.

Changing the tracking system to match that in a previous version of ANY-maze

We are continually improving ANY-maze, both in terms of the features it offers and also the way that it tracks animals. When such advancements are made, it may cause the tracking of the animals to behave slightly differently to previous versions of the software. Usually this is a good thing as the tracking should be more accurate; however, if you want to compare the results of experiments made in this version of ANY-maze with a previous version, you will probably want to ensure that the tracking algorithm is as similar as possible in both experiments. Also in some rare situations, the tracking in your apparatus may have been better in a prior version.

To address this, you can alter the tracking engine that ANY-maze should use to one from a previous software version. Note that this will only take effect for future tests and won't affect the tracking in any tests that you have already performed in the experiment. For this reason we generally don't recommend changing the tracking system once you have started performing the tests in an experiment. If you do change it, you may want to consider re-performing any tests that you had already performed.

If you do find yourself wanting to revert to a previous tracking version because you feel it provided better results than the updated tracking, please let us know by emailing ANY-maze Support. If you can send us videos of tests in your apparatus and an explanation of what aspects of the tracking you're unhappy with, we'll do what we can to improve it.

Tracking when the apparatus is inside an enclosure

This feature is a legacy option which was removed from ANY-maze in version 5.1. It will still appear if you open a file that was created in an earlier version and which used this function.

If your apparatus is inside an enclosure such as a soundproof box, then ANY-maze will need to know this. The main reason is related to the changes in light levels which will probably occur when you open the box to put the animal inside, although there are other technical considerations as well, including the fact that the auto-start feature will work in a different way as you are unlikely to actually appear inside the box.

Note that you don't need to switch this option on if the apparatus isn't entirely enclosed. For example, if you have a light-dark box in an open fronted cupboard (i.e. one with no doors), then you don't need to switch this option on. If, however, the box is inside a cupboard which does have doors, and you close the doors during tests, then you should turn this option on.

Although most of the issues related to tracking inside an enclosed box are addressed by setting this option on, there are a few other things you should be aware of and these are listed in Testing do's and don'ts

© Copyright 2003-2026 Stoelting Co. All rights reserved

ANY-maze help topic T0314