Exactly. And I think the effect can be even worse when examining our own behavior and state of mind. There are plenty of extremely well-known cognitive biases t

Exactly. And I think the effect can be even worse when examining our own behavior and state of mind. There are plenty of extremely well-known cognitive biases that tend to effect our own self-examination. It doesn't mean self-examination is bad, but it does mean that more methodical methods of reasoning (like scientific experiments) are important. The "sugar rush" case has another potential parallel here. It may in fact be the case that parents, when frustrated by their children's hyperactivity, give their children less sugar and in fact observe lower energy levels. Parents may even mistake this for "personal evidence" that "sugar causes hyperactivity" is a good explanation for their observations. But of course, they very well may have just lowered their child's calorie intake, and perhaps sugar itself has no unique effect on hyperactivity. Another thing to note regarding the sugar misconception is that studies have in fact shown that parents will report their children are more hyperactive if they have been told that their children recently consumed sugar. To me, that's the real nail in the coffin, and it's also not good news for the mode of thinking that goes like "I've observed this feeling in my own life therefore no evidence could possibly show otherwise."