Hello,

I was wondering what exactly the "baseline hits" matrix does under the statistics tab? Is this just informational (proportion of correct hits for each target), or is it used in the adaptive algorithms?

When I was using the standard mu rhythm RJB task where there may be, for example, 4 possible targets of equal size on the right side of the screen, there would be a row in the matrix for each target where the first column would have the target number and the second column would have the probability (0.25). However, now I'm doing tasks where the targets may be of different sizes and distance from the starting point. This would not lead to an equal probabily that the subject could hit each one. So, I'm not sure what to do with the baseline hits matrix, or if it even matters. I also noticed that sometimes I will set it up with values and when I check it after a couple runs it will have cleared out.

A related question... does having targets of different size and distance from the starting point make a difference for other adaptive parameters?

I appreciate any help or comments you have!

Thanks,

Elizabeth Felton

University of Wisconsin - Madison

## baseline hits, etc.

### Re: Adaptation parameters ...

Elisabeth,

As always, you are asking great questions

BaselineHits is informational, but is also used in the adaptive algorithm if you set X/YTrendControl to 2. In this case, you need to have one entry for each target. If you set X/YTrendControl to 3, then you also need to correcly set the adaptation values in the TargetPos matrix. See the User Manual for more details on this.

If target probabilities are not equal, I would change the BaselineHits appropriately. Otherwise I do not think there should be any problem with this scenario, although I am not sure we ever tried it.

The Gerv

As always, you are asking great questions

BaselineHits is informational, but is also used in the adaptive algorithm if you set X/YTrendControl to 2. In this case, you need to have one entry for each target. If you set X/YTrendControl to 3, then you also need to correcly set the adaptation values in the TargetPos matrix. See the User Manual for more details on this.

If target probabilities are not equal, I would change the BaselineHits appropriately. Otherwise I do not think there should be any problem with this scenario, although I am not sure we ever tried it.

The Gerv

### Who is online

Users browsing this forum: No registered users and 1 guest