If I remember (and understood) correctly, the ‘effort preference’ was measured using a test that’s designed to measure ‘effort preference’ in patients with e.g. depression.
You play a game with a time limit where you have to repeatedly choose between an easy and a hard task of quickly pressing a button X times within Y seconds. For easy tasks, both X and Y are low. For hard tasks, X and Y are higher - you have press the buttom more times and for a longer duration. The values for X and Y are not constant, so you have to figure out if the given task is relatively trivial or difficult for its category.
Each task gives a reward. The easy tasks give a lower reward than the hard tasks. However, the rewards are also randomized. So you might end up with a trivial ‘easy’ task that gives a higher reward than the previous difficult ‘easy’ task.
Your job is to maximize your reward. So they measure ‘effort preference’ in terms of how ‘motivated’ you are to play strategically and try to pick the tasks that gives the most bang for your buck, so to speak.
Additionally, one very important premisse is that the participants have to be able to complete all tasks. But as we know, people with ME/CFS might tire out more quickly, and might not be able to complete the hard tasks because they are physically too demanding. Not surprisingly, the data showed that some of the ME/CFS patients were unable to complete the hard tasks, so they scored way lower because of that. This data was NOT excluded, but the data shows that people with ME/CFS that are able to complete the test actually has a marginally higher ‘effort preference’ than the healthy controls.
Furthemore, the test has a fatal flaw. The participants get real life money from playing, but not based on the total score. They get the money from just two randomly chosen tasks that they managed to complete.
This means that the optimal strategy is to only complete high value hard tasks. If you don’t get a high value hard task that you believe you can complete, you always pick the easy (and shorter task) and fail it on purpose.
(Edit: I forgot to say that this strategy gives you a lower score in the test. So playing optimally is worse - which makes no sense at all.)
There’s also something about how they intepreted the results in a different way than the creators of the test, but I can’t remember the details.
To sum it up: They used a fatally flawed test that isn’t suitable for physically ill people. They also deviated from the data intepretation protocol without explaining why.
