Open bastianilso opened 2 years ago
Should the field shown on the screenshot be replaced by a value? Or does it only serve as a title?
Should the field shown on the screenshot be replaced by a value? Or does it only serve as a title? Put the value next to "Reaction Time:" fx. Reaction Time: 0.53s
Should the field shown on the screenshot be replaced by a value? Or does it only serve as a title? Put the value next to "Reaction Time:" fx. Reaction Time: 0.53s
Alright, so what's the difference between "Reaction Time", "Best", "Median" and "Worst" ? Because I have only 3 columns for the reaction time in the database, and 4 in your preview. It's just a mistake ?
aaaaah, you're right. OK, in that case, Reaction time is the title. and best, worst, median are the places where we put values like i described above
Reaction Time Best: 0.2s Median: 0.4s Worst: 3.4s
the R shiny page currently looks like this, do you want any changes?
Thanks jules, looks good. 🚲
Some quick notes:
Currently I use something like this :
if the reaction time of the patient is in the first 50% of the histogram :
the color is green
if the reaction time of the patient is in the second 50% of the histogram :
the color is orange
else :
the color is red
the graph below (Cumulative Histogram) represents the MPC sessionMedianReactionTime for this settings : circleAmount=12, gameType=A, sessionLength=2
We can use something similar to select the color of the iPad (see above). -> For example :
if the reaction time of the patient is in the first 50% of the histogram :
the color is green
if the reaction time of the patient is in the second 50% of the histogram :
the color is orange
else :
the color is red
The problem is when the patient's session does not match with the game settings of MPC players, what are we supposed to compare ?
I can suggest you to use the MPC data when it's possible, and when it's not possible to use what I used before. I can add parameters so that the user can modify the range of colors as he wish.
what do you think of that ?
Hi Jules,
This is a good point. We don't know how much the reaction time will change, when we increase circleAmount from 12 to 18, to 30. We would need to collect data about this, in order to have a complete reference data set.
Perhaps you can ask some students here at CREATE to play Trail it for 2 minutes, in the other 2 difficulty settings? I think we can get a nice reaction time estimate if we can find 12 players.
If you need consent forms, they should be in my office.
so you would like to do that in these difficulties ?
So this also means that we assume that the session length does not influence the reaction time?
The main influence I would expect from session length would be:
We can probably overcome these effects by averaging across the whole session. Secondly, since the heatmap is a spatial visualization, I think it is less important in this context.
@Jules-CLERC What does the dashboard say? :-) If you plot reaction time, for each level, can you observe these effects?
I don't know exactly what to conclude from this, the reaction time doesn't seem to change especially in the first levels, Regarding the last levels, it is normal that the reaction time is shorter, because only the best players reach these levels.
@Jules-CLERC Thanks! I would show the y-axis from 0 to 1 second here - this is a bit too zoomed in (if you look at the y-axis, the difference is max ~150ms from the best to the worst reaction time, which is not all that much).
So yea, not that much difference, therefore we don't need to worry too much about the session length right now.
Code for the heatmap (use string replace to insert the right colors and reaction times in the right spots).