Dieting that resulted in exorbitant glucose variability, perhaps, yeah. But there are a lot of other factors at play than just glucose/insulin - tiny example eg. leptin ~> activation of anorexigenic POMC neurons ~ melanocortin ~ glucocorticoids/appetite/obesity/sexual behavior/sleep + higher fasting insulin levels

This would explain why the ketogenic diet is so effective, beucase your brain starts using ketones instead of glucose for energy.

I know for me, as a person with ADHD, I have MUCH better self control when I'm on keto then when I'm not. It's the main reason I stick to the diet.

This is an interesting point. As others addressed the main issues, I would just add that dieting does not need to be continuous exercise of will power. It may just take will power in establishing an environment not surrounded by unhealthy food, e.g. when shopping.

I would not relate this point to serum glucose in any way, however.

Dieting is possible... but you should expect to have significantly decreased self control in all other areas of life at least for as long as active self control is necessary to refrain from eating poorly.

Is this an attested phenomenon? How does it work?

I was just reasoning from the studies that exerting self control depletes your ability to exert more self control. In the study most often cited there is actually an interesting link to this exact phenomenon, though. The participants were asked to refrain from eating chocolate (although there is no indication that these people were trying to diet for any reason). Afterward they were tested on standard decision-making tasks, at which those who had to resist eating chocolate performed significantly worse than those who didn't. Then, they were given a glucose drink and their performance returned to normal. A person dieting would be both exerting self control to resist eating things and also resisting things that would provide more glucose for the brain to be using.

However, I seriously doubt this is specific to self control. I would imagine performing any action that depletes glucose in the brain would have the same effects. Prioritizing tasks and deciding tasks to delegate are a couple I've read which are supposed to be very energy-hungry (I've been reading 'Your Brain At Work' recently and its mentioned in there).

Not impossible, but difficult. And more difficult for some than others depending on initial glucose level.

No but it would explain why so many have difficulty. This isn't all that interesting to be honest. We've known for a long time that energy levels have a direct effect on brain function. That self control is a "function of the brain" is obvious and subject to the same principals.

Garbage in, garbage out as the saying goes. Proper diet will improve brain function and "self control".

From the paper:

"Resisting the temptation to eat the cookies presum- ably depleted an energy resource that could otherwise have been used to persist on the subsequent task"

To your point, the best way to diet then would be to avoid the temptation which would mean less use of glucose in the resisting of it.