One of the things that worries me is what Mithriel mentioned. PWME go in to this test motivated by wanting to produce an abnormal result - maybe the second result to look different from the first. Knowing just how much frame of mind affects one's physical performance on the running track I wonder how problematic this is.
I seem to have missed this thread a week and a half ago.
Frame of mind has no impact at the levels of exertion involved because it is a ramped exertion test. You cannot deliberately put in more or less effort because workrate is set externally and anything else leads to failure of the test.
This puzzles me too. How do we now a voluntary reduction does not occur? We know almost nothing of the mechanism of 'voluntary effort'. If I am on an exercise bike and someone asks me if I am doing maximum voluntary effort I don't think I can say. One can get somewhere near to a sense that instantaneously but over a period of many seconds all sorts of things modulate the sense of effort being maximal. Effort isn't actually something we have any way of measuring. And if you feel ill you are in a completely different mindset.
Effort is controlled during a ramped workrate exercise test. That is the strength of this method. The perceived effort on the other hand is proportional to the signal generated upstream of the motor cortex and is related to how our proprioceptive system works (dating back to Helmholtz).
Notably, CPET participants (regardless of healthy or athletic) have a nonlinear shift in perceived effort (on Borg scale) at the ventilatory threshold. It is not the baseline level of perceived effort or slope that we should care about, it is when there is a nonlinear shift in the curve.
Also, for what it's worth, I apparently reported lower perceived exertion (muscle exertion, not breath) during the 40w warm up on the second day. But as the workrate ramped up, it seemed much harder than the first test - any power I felt I had on the previous day was kind of missing - I can't remember when I first noticed this, but it was likely from around the VT1 and above based on the power (workrate) figures.
Regardless of whether it was kinetic limitations of metabolic sources of energy that required an increased shift in effort or not, the fact is this nonlinear shift will lead to a sudden increase in activation of (higher threshold, higher force) muscle fibres which have lower capillarisation, lower myoglobin, lower oxidative capacity, but higher glycolytic capacity. Hence there will be a noticeable shift in metabolic sources from this alone.
The 'bias' of the system is not the upstream signal, for if that changed significantly from day to day, there would be major problems with proprioception, which we do not see in patients.
Instead, the bias is determined by the feedback from type III/IV muscle afferents which in turn reduce the excitability of the motor cortex output ("central fatigue"). The purpose of this biasing system is simply to increase the level of ventilation for a given level of motor cortex output to account for fatigue related shifts in metabolism.