Psychomotor Vigilance Test - discussion and testing

For the record, I used my laptop with a mouse when I got my result upthread. Cognitive dysfunction, slow reactions, difficulty concentrating etc are among my most debilitating symptoms which are present (although to a lesser extent) even without PEM, so I'm not surprised. I did it again today, got 348 msec when I tried doing my best in a more "refreshed" state, still suboptimal. I'll try to do it again when I happen to be in PEM, to see how I score.

 

If there isn't a similar test already available as a program (avoiding internet delays), it should be fairly simple (for someone without ME) to write. The hardware and peripheral variations (mouse vs touch) is more of a problem, so I agree it's more important to to note the variations between ME severity for each individual. People without ME also will vary a lot on this test. I'm curious whether people with severe ME are reliably worse on the test.

 

If it's not a great example of how much brain fog impacts us, I don't know what is.

 

I checked and touchscreen sampling rates are fairly close to what most computer mice have. Shouldn't be much more than a 10-15ms difference, and there is a difference with the quality of mice as well so there's variation all around.

It's usually between 5-20ms in both cases.

 

Touchscreens must vary considerably, then. Both from android to ios and likely from device to device. I am very loyal to the ipad mini - this is my third - but the OS is glitchier than other macs, or perhaps devs don’t test on it sufficiently.

 

Oh definitely on cheaper devices there will be a larger difference. But other than expensive gaming mice, many touchscreens are actually faster than a regular mouse. Especially smaller screens.

It's not just glitches or anything like that, the sampling rate is an engineering choice based on costs.

 

I'm going to get all boring and make it complicated but I think this tool for ME/CFS might be more interesting used in a different way than necessarily it is/just doing the 120s and taking an average.

Following the obvious 'learning effect' ie where I'm rubbish geting my eye in for the first few goes, I noticed myself getting tired and worse half way through yesterday and had to change hands at some point and noticed an uptake at that point. Today I found I had to switch hands regularly and the time was drastically faster with the fresh hand and then tailed off after about 4/5 goes then picking up with other hand repeated almost same time pattern (reaction time doubling after 5 goes)

I guess for pwme that varies with severity and PEM but also what condition you are generally in that day (sensory, cognitive, motor) but I wonder whether the more interesting measure (which might vary based on this) is when the 'tail off' effect ie starting to get more rubbish instead of better with practice, kicks in as much as the 'fatigue' idea of just being sluggish. So I'd be intrigued if there were finer-grained options on tests for us.

Of course this is where the 'if anyone really actually wanted to 'help' and do something useful for us' part is (and why I don't believe the 'good intentions' fop from so many)... if this type of software was used to experiment with different 2-day CPET type things and testing what difference taking breaks or chunking into smaller time periods has on different pwme vs 'norms' normal distribution then we might start having some useful advice that could help us manage our illness. I'd of course want the caveat that it wasn't about 'function at the time' ie take more breaks and perform better that day, or for weeks, but also checking to see whether in the months that followed such 'performing at limit' (ie learning to 'cheat the PEM'/ride the limit) led to relapse. Which is where those averages certainly would also be interesting?

Plus of course I'm sure that things like whether you are lying down and comfortable vs the awkwardness of set-up that measn re having your finger in an easy position to see and click all matter in playing off health vs this test. But it surely could also easily be adapted to show the difference in reclining vs standing or the effect of sitting up for hours on certain people simply by changing to a hand-held trigger type clicker + google glasses for display. Heck someone from the product design area could do tests to look at different set-ups and colours/resolution/contrast that make it 'kinder' or easier on us - which could have implications for future developments that might help future pwme.

With tech going where it is there seem fewer excuses for these simple things not to be converted into well-thought-through tests by scientists who actually listen to patients so can indeed control for the various different variables we all have.

 

Results:
Test Duration= 121 seconds

Number of false starts= 0

Average response time= 438 msec over 25 attempts.

Your results show that your alertness may be suboptimal. Consider medical evaluation.

I didn't expect a good result. My reflexes are not great and never have been. Another possible factor - I'm also a very slow writer and always have been and I simply can't write fast without it just becoming a jerky meaningless scribble. I noticed that my results varied a lot because my concentration is poor and my eyes wandered. My best time was about 280 ms and my worst was over 800 ms.

I did it a second time and managed to reduce the number of times my eyes wandered, but it didn't change the opinion :

Test Duration= 121 seconds

Number of false starts= 0

Average response time= 363 msec over 24 attempts.

Your results show that your alertness may be suboptimal. Consider medical evaluation.

I think it would have been interesting for each session to see the slowest and fastest time and a measure of the variance of the results.

 

Test Duration= 125 seconds

Number of false starts= 0

Average response time= 516 msec over 25 attempts.

Your results show that your alertness may be suboptimal. Consider medical evaluation.

It is 4:25am, though.

 

Shouldn't there be an age adjustment for this kind of test?

Isn't it known that, on average, reaction time slows down as one gets older?

 

Oh good, a competition.

Results:
Test Duration= 122 seconds

Number of false starts= 0

Average response time= 266 msec over 25 attempts.

Your results show that your vigilance and alertness are excellent

 

Great that some of us can still manage decent numbers
Next we're going to hear that those numbers can be too good in ME/CFS, confirming hypervigilance and FND...

 

Graham McPhee was working on a program of modified Stroop test, which included a second test after about 20 mins I think. He tested it on some local group members and others. Any testing needs to show the "fade" effect, ie PEM.

 

I took the test last night around midnight, I'm a night owl. I was tired, but not ME tired. Thought I'd compare with midday, and managed a little better, but I think that's just because of practise, dogged determination and testosterone fuelled ambition:

Results:
Test Duration= 124 seconds

Number of false starts= 0

Average response time= 251 msec over 23 attempts.

Your results show that your vigilance and alertness are excellent

 

I don't know why they think it's an appropriate thing to have on a sleep disorders website, people like me will be up all night trying to beat their high score.

 

The latency you get in this test will be the sum of your reaction time and the computer's latency, most of which is display latency. Display latency varies widely, and is likely to depend on obscure details of your operating system, graphic card, and monitor. The lag from your mouse counts but is likely small in comparison. Thus, I would not trust the absolute numbers from this test unlessinput and display lag can be calibrated out. Regardless, the results will still be consistent as long as identical hardware and software are used.