Michael Uebel refers to a paper where they advise to start the video-tracking system and a timer set for 5 min when the rodent is placed in the maze.

What part of the maze you are most interested in you can analyze from your video tracking. I have used a linear maze with 11 possibilities to make errors and the mouse spent about 2 min in the maze the first day and the last of 5 days he could even run the maze in 12 sec. I add a picture of our maze (and open field).

We recorded the errors and time by hand and found that it was a convenient way to do that and cheap, as we did not have much money.