I reckon this thread is more informative about PEM than some of the published papers out there!



It's also been very helpful to me in trying to answer some of my own questions. As I said earlier, I used to think I got PEM from both cognitive and physical exertion and that the two were much the same. On reflection I started to have doubts, especially because all the PEM events I had considered cognitively-induced had actually involved a mix of different types of exertion.
After a lot of head scratching, here are my tentative conclusions (with all the usual reservations around recall bias):
rapid-onset localised symptoms after very minimal exertion (ROLS)
These start during an activity and necessitate a relatively brief rest before being able to continue. Symptoms relate directly to the activity.
Example (physical): meal preparation has to be broken into several sessions with rest between, symptoms are tiring arm muscles plus orthostatic discomfort from sitting upright with feet down.
This physical activity-rest cycle can be safely repeated a few times throughout the day but repeating too often leads to delayed PEM, presumably due to cumulative effects. This means the
PEM threshold is crossed before symptoms appear (apart from the temporary sort like in the example above that don't by themselves lead to PEM).
Example (cognitive): reading an article has to be broken up by rest every few paragraphs as comprehension starts to fail.
This cognitive activity-rest cycle can be safely repeated a few times throughout the day; repeating too often leads to increasing mental fatigue until the brain simply shuts down - but it
does not lead to delayed PEM, as far as I can tell.
Some of these ROLS could be mistaken for deconditioning except for the fact that the level of exertion needed to trigger them varies too much and too fast to be fully explained by any training effect.
rapid-onset localised symptoms plus generalised exhaustion* after minor exertion (ROLS++)
Similar to ROLS but after a slightly higher level of exertion. Several hours of rest are needed for symptoms to subside again. Symptoms are more severe but still relate directly to the activity. In addition there is generalised exhaustion.
Example (physical): walking 50m all at once leads to shaky, straining legs, breathlessness and generalised exhaustion. These symptoms gradually improve over some hours of rest but will
always be followed by delayed PEM even if the activity is immediately followed by complete rest.
Example (cognitive): I experimented with lying down and listening to a podcast in a language I'm not entirely fluent in while also doing a mental simalteneous translation of it. Predictably that fried my brain in no time at all. Major brain fog and generalised exhaustion for several hours after but
no delayed PEM.
* With "generalised exhaustion" I mean more than fatigue. It can include feeling unsteady, dizzy, nauseous and being unable to think straight, a bit like a not quite fit enough marathon runner collapsing after staggering across the finish line, but without the endorphins.
delayed PEM, early phase
Starts 4-6 hours after exertion. Symptoms -
irrespective of type of overexertion - feel like immune activation, much like coming down with a cold with a sore throat, feeling in turn hot and cold, and generally feeling off. Timing of early PEM symptoms can overlap with the ROLS++ which can make them difficult to tell apart. The main PEM symptom cluster invariably follows the next day, no amount of rest can stop it (though not resting will make it worse).
Definite triggers: physical & orthostatic overexertion
Likely* triggers: stress & emotional exertion - interestingly ROLS & ROLS++ symptoms after stress & emotional exertion are indistinguishable from those after orthostatic stress
Unlikely* triggers: sensory overload & pure cognitive effort, refer my experiment** above which resulted in major ROLS++ but not in delayed PEM whereas the same level of ROLS++ from physical activity always leads to PEM
* Can't be certain about degree of likely/unlikely because activities involving those potential triggers typically also involve some level of physical and/or orthostatic exertion at the same time.
** Of course my little experiment is an anecdote about a single, not a cumulative, effort of a very specific type of cognitive exertion. Interpret with caution. Oh, and don't experiment yourself, please. Your outcome may be a lot worse than mine. Do as I say, not as I do!
delayed PEM, peak phase
Continuation of the above, symptoms peak 24-36 hours after exertion but take a lot longer to reduce again. Symptoms are much more severe and cover the whole spectrum from flu-like to autonomic and cognitive etc etc.
In summary:
For me, both physical and cognitive exertion lead to localised symptoms immediately or soon after the activity. Symptoms can be significant but I'm not convinced they're PEM. I suspect they reflect the sort of fatiguability a lot of people with other conditions struggle with, too. However many people, within as well as outside the ME community, call these symptoms PEM.
With respect to "classic" delayed PEM, I can only be certain it happens after physical (and orthostatic) overexertion. Symptoms of "classic" delayed PEM are qualitatively different from the ROLS/ROLS++ ones. I can't be sure about "classic" delayed PEM after cognitive exertion but I'm inclined to say I don't get it.
Also, this difference seems important:
If I exert physically to the point of getting ROLS++, I will always get PEM later; the PEM threshold is passed even before appearance of symptoms.
If I exert cognitively to the point of getting ROLS++, I will recover over a few hours and not get PEM the next day.