As documented in several publications, active locomotion can strongly modulate cortical activity. I explained earlier how Scanbox uses a quadrature encoder to be able to control, in real time, an external stimulus that depends on the state of locomotion. In this way, we can design experiments in which stimuli are delivered during periods of locomotion or rest.
However, if either locomotion or rest are not sampled often enough by an individual subject, one may want to use an active treadmill to have some control over their duration. The same controller used to move the motors of the microscope has additional axes that can be used for this purpose, allowing us to slowly ramping up and down the velocity of the platform with arbitrary velocity profiles. The video below shows the smooth acceleration and deceleration we can achieve. The Scanbox interface keeps track of commands issued to the treadmill and monitors the actual physical movement through the quadrature encoder.
Scanbox 
Dario, do you find that the 8 kHZ (or whatever it is for your resonant galvos) affects behaviour? One of the things we have found with our behaviour work (visual detection task) is that the quieter environment and less distractions, the better. But maybe you have your galvos sound-insulated and can’t hear them (we haven’t even attempted to do that yet).
Our scan head is fully enclosed, but you can easily hear the 8kHz of the resonant mirror. We don’t have any visual/detection or discrimination tasks at the moment, so I can’t tell… but we do acclimate them to the platform and under the microscope a few days before imaging. Maybe training to do the task under the same conditions that you will image (including the 8kHz tone) will help them ignore the noise?
Thanks Dario. At the moment our 2P imaging and mouse behaviour are separate endeavours (we’re using Neuronexus MEAs with the behaving mice). We do want to bring the two streams together at some point though, hence our thinking about it. Of course, we could do this on one of our non-resonant scopes. We don’t know that the 8 kHz will cause a problem – it was just the intuition of our mouse behaviour people (who may themselves end up instrumentally conditioned…) that this would be a problem. I’ll let you know if/when we try it.