MST and Sports

[Alessandro Santirocchi]

Hi Dr. Stark

We're currently designing a study in which we would use MST in participants who practice different sports. One of the aim of this study is to understand if practicing a specific sport affects the pattern separation effect. We would like to use the sets 1-4 and the default 2s Dur, 5s ISI, 64 per condition, self-paced. We would be very grateful if you could kindly give us your opinion about the suitability of these parameters and sets for our study?

Thanks in advance for your kind help and cooperation

Best

Alessandro 

[Shauna Stark]

Those parameters all look good here. My only suggestion would be to randomize the sets across conditions. We have made a strong effort to match difficulty across sets, but we usually still randomize participants across repeat testing to ensure that set effects aren't a factor.

Good luck with your research study! I'll be interested in your results!

Best,

Shauna

[Alessandro Santirocchi]

Dear Shauna,

 

thank you very much for your answer and for your methodological advice that we’ll follow. We’ll keep you update about our study and results.

 

Best regards

 

Alessandro

[alessandro....@uniroma1.it]

Dear Shauna,

My research is ongoing, however I had to stop data collection due to the Covid-19 emergency and I wanted to know if there exists an online version of the MST so that I can keep going forward with the administration with athletes.

Best regards

Alessandro

[Craig Stark]

Yes, we do have online versions up and going and have been using them in-house.  All are built using jsPsych to really handle stimulus delivery, but I've been working with both Cordova and JATOS for other aspects of management.  Cordova is nice in that you can easily get onto things like iPads as well, but JATOS gives a very nice platform to let experimenters manage studies, their data, etc.  The way I have it setup, though, is that for each experiment I want to run, I have all the MST code and resources (e.g., images) and I have a "setup.html" file that does all of the things like assigning people to conditions.  That setup code changes across experiments while the main MST code remains the same.

Shifting to online, though, means you'll need something to host the code.  I've had good luck even running a simple virtual machine (e.g., VirtualBox) that runs a small Linux web server.  

In the end, here -- I've got several variants of the MST up and going and we're using them just fine.  Being remote does mean a bit more setup on your end, so it's not quite as simple as just handing out the standalone.  I can put a simple JATOS study-test MST up on the GitHub page if people like, though.

Craig

[san...@gmail.com]

Hi everyone,

We have finished data collection for our project but I am curious about your thoughts on administering the MST as we would any other computer-based clinical task. I have been using Zoom and the remote access feature and this has been working well. 

For the MST, I imagine I would run the task on my end like normal, give the participant remote control through the Zoom platform, and they make the choices for the task on their end. This process has been working well for other tasks so far.  The most common issue I would plan for would be Internet related delays and this may be troublesome depending on which version of the MST is being administered, but would this method work otherwise?

Curious to hear your thoughts.

Sanam

[Craig Stark]

Right - you could either give them control or you could do a quick install of the MST on their machine and run it that way.  Since we don't really use reaction time, I doubt it'd be a real issue in terms of the timing.

Another option would be to setup a Zoom with them so you can consent, give instructions, etc. and then give them a link to the online version.  With JATOS, you can either install JATOS onto some web server (I've even used a simple virtual machine) or you can just run JATOS on some computer in the lab / at home that can be exposed to the outside world (might have to work with IT to open a port if there's a university firewall).  But, the simple method would be to just run JATOS on some Windows / Mac / Linux box and then, when you're in Zoom, you'd send them a link (e.g.,  http://127.0.0.1:9000/publix/161/162/start?srid=425 -- except not 127.0.0.1 but the actual IP of your machine).  Their web browser and your computer would together run the experiment and the data would end up on your machine.

Craig

[alessandro....@uniroma1.it]

Dear all, 

Thanks a lot, for your support, this is really helpful. 

We’ll keep you update about our study.

Best regards 

Alessandro