Experiences and methods for observing, testing and tracking cognitive fatigueability and PEM - discussion thread

Trish

Trish

Moderator
Staff member
This arose on another thread and I thought worth starting a new thread:

Are there changes we notice to cognitive function that indicate we are cognitively fatiguing and may be crashing or crashed?

Can we develop ways of tracking them easily to help with pacing?

This thread is for sharing your experiences of cognitive changes you experience, and for suggesting ways of tracking them.
 
@forestglip has started a thread about using photos of the face to track changes. Go to that thread if you want to discuss it:
Idea for machine learning model to track fatigue accurately

Also from that thread, Peter Trewhitt has suggested tracking how long a string of numbers you can recall, eg phone numbers, which reduces when cognitively fatigued and when in PEM.

Peter Trewhitt said:
Added Note - another free association is that my digit span reduces dramatically when in PEM and when in an ME crash. This is the telephone number memory, the ability to immediately repeat a string of digits, what we use to go from a number in a phone directory to dialling the number. The norm is repeating seven digits plus or minus two. Pre-morbidly I could happily repeat eight or nine digits (as a former psychology student I was very aware of this) now generally I am only reliable with four or five, but when in PEM it can go down to one or two, so that I can only dial a written phone number one digit at a time whilst marking the current digit physically with my finger. I am very aware of this at present as I am currently in a battle with my mobile provider who are failing to transfer my service from an ancient mobile to a new smart phone. Lots of texted security codes, passwords and PIN numbers in numerous phone calls to automated systems and unhelpful customer service operatives, as well as now having to manually copy my stored numbers from the old to the new phone. So I am having to consciously use my digit span in ongoing PEM inducing activity.

It would not be too complex, I imagine, to create an app to measure your current digit span, though having to keep breaking to do so would be disruptive and could add to triggering PEM.
Click to expand...

Also suggested: Changes in reasoning ability.
 
I notice I'm cognitively fatiguing when my typing error rate rockets up. If it's really bad, and I want to type something, I switch from touch typing on my laptop where I can't see the keys, so I'm typing from memory of key positions, to typing one fingered on my smartphone where I am looking at the keyboard.
 
Speech. Word-finding difficulties, errors or slowness in speech, more "er"s and pauses, and perhaps tangentiality or a lower level of coherency as well.

I'm not sure if speech-to-text is good enough to measure the error rate here, yet, and I doubt there's an open-source system that will measure the timings between words to determine if they're atypical. ASR systems are obviously improving, though - I think OpenAI released a new one, Whisper, recently?
 
Here's another possibility: typing patterns. I'd bet that if I typed a paragraph or two and inter-keystroke timing was measured you'd get a correlation with how I was feeling on that particular day. (If you wanted to observe usual activity, rather than keep asking the user to type a sample paragraph, you'd need a kernel-mode driver or a DLL injected into every process that receives textual input on Windows. I've written such software in the distant past, albeit for a completely different purpose.)
 
I think observing usual typing activity would be problematic, as different tasks require more thought, so would happen more slowly even when relatively less fatigued, and would be much more complex to track.

A short typing task about the length of this post, with a document to copy would be OK, I think, and could be used as a test piece. The speed, length of pauses, number of errors, whether you notice and correct them, would be quite revealing.
 
Music - keeping time, coming in on time, and hitting pitch when singing.

Speech, linguistic, and also spectral, tonal, and rhythmic analysis.

General ability to discriminate auditory and visual input.

Memory and concentration.

Coordination, accuracy, and speed of typing.

Handwriting.

Any motor-sensory-cognitive action or activity, especially repeated or sustained.
 
I think observing usual typing activity would be problematic, as different tasks require more thought, so would happen more slowly even when relatively less fatigued, and would be much more complex to track.
Click to expand...
You could track how the user's speed, error rate etc changes over the course of working on the same task (document/web form/etc) - does it worsen over time, and by how much? Are they fatiguing quickly, or slowly? You could also try to do this a bit more intelligently by tracking the user's activity at the time in an attempt to avoid task-specific differences: for instance, you could track data entered into web forms separately from typing into Microsoft Word.

There are possible non-cognitive ones too: the one I thought of straight away was gait analysis. I know I often feel less steady, and more prone to dizziness and vertigo, around PEM episodes. I'm not sure if the cameras on mobile devices and modern computers are good enough to operationalise this in an app, though. Several years ago there were attempts to commercialise gait analysis as a form of biometric authentication, so the technology definitely exists.
 
Last edited:
Trish said:
I notice I'm cognitively fatiguing when my typing error rate rockets up. If it's really bad, and I want to type something, I switch from touch typing on my laptop where I can't see the keys, so I'm typing from memory of key positions, to typing one fingered on my smartphone where I am looking at the keyboard.
Click to expand...

Yep, I'm totally hitting backspace much more frequently when I'm very fatigued.

Nightsong said:
Here's another possibility: typing patterns. I'd bet that if I typed a paragraph or two and inter-keystroke timing was measured you'd get a correlation with how I was feeling on that particular day. (If you wanted to observe usual activity, rather than keep asking the user to type a sample paragraph, you'd need a kernel-mode driver or a DLL injected into every process that receives textual input on Windows. I've written such software in the distant past, albeit for a completely different purpose.)
Click to expand...

This seems potentially useful. A keylogger that captures each button press and the time between. It might return something like this:

P - 7 ms
D - 5 ms
BACKSPACE - 3 ms
U - 6 ms
P - 5 ms
P - 2 ms
Y - 4 ms

Though practically, there's a huge privacy issue with this if its logging everything you type. It'd probably have to be made open-source and work completely locally (not send any data to another computer) to be trustworthy.

I think there are devices you can connect between a keyboard and computer to catch all the buttons you press.
 
You could measure all kinds of variables. Backspace use, yes, and inter-word as well as inter-keystroke timings (the time it takes to find a suitable word). Spelling & grammar error rates relative to a baseline. Mouse movement patterns. As for privacy - as long as you don't store the text on disk or send it over the network - just do the analysis and zeroise the memory that contains the analysed text. In order to work out whether any of this would be useful you'd need a small trial with a group of cooperating ME patients: measure absolutely every statistical & timing property you can think of, see which ones correlate with visual analogue scales / questionnaires.

Another few "fatigue/PEM detection" ideas (cognitive & non-cognitive):

- Eye-movement tracking. Frequency of eye movements, blink rate, etc, may correlate with inattention or fatigue (?)

- Breathing patterns may change as exertion becomes harder, and there are probably various measurable physiological correlates of that.

- HR relative to recumbancy time and step count. I think in some patients postural tachycardia may be worsened during episodes of PEM, so tracking how HR responds not only to going from supine to standing & sitting to standing but also by how much it increases based on the number of steps taken compared to a baseline. (I know some patients are already tracking their sitting/standing HRs & step counts but I'm thinking more about the relationship between the two.) I'd guess that data would be accessible through some smartwatch/wearable APIs.

ETA: I wonder if there are subtle movements - or lack thereof - associated with fatigue or PEM that could be picked up by a wearable's accelerometer?
 
Last edited:
Anything that needs rhythm (walking, playing music), balance (walking, moving from standing to sitting), spatial relationships (going through a doorway), and grip force calculation (picking up a plate) will show that I'm off.

Also having to make up nouns and adjectives because the real ones have disappeared behind the brick wall. Eg, reducing people to helpless giggles by describing the window in a tin whistle as the music hole.

I don't know whether facial expressions would work for me, as mine are very contrived due to autism. I had to learn to do them as deliberately as the way children learn to pronounce a sentence in another language.

Skin tone probably would, though, it's how colleagues at work used to know. It might only be because I have the fairest skin type humans are issued with and my colouring changes a lot during conversation. When I'm in PEM it apparently doesn't, it just looks grey.
 
While I relate to many symptoms above, my wife has other experiences. Could we consider starting multiple threads on the lesser symptoms occasionally to bring out the differences or commonality between sufferers? For instance the International criteria lists poor breathing / breathlessness, what does that mean for us as individuals and our experience? I could add IBS or allergy, night sweats, urine urgency, poor temperature control. While many of this non exhaustive list gets mentioned in passing I cannot see patient experience in a list per minor symptoms. Not sure if it should be public, but it’s very reassuring to read others have the same experience.
 
yME said:
While I relate to many symptoms above, my wife has other experiences. Could we consider starting multiple threads on the lesser symptoms
Click to expand...
By all means start or join other symptoms threads. There are lots of threads already, so you may find one that fits, or you can start another one if you don't find one.
The symptoms forums are all listed here:
Symptoms and signs discussions
And the members only forum is here:
Symptoms and Signs (Members Only)
 
You must log in or register to reply here.
Back
Top Bottom