While I don't think the comparison is perfect, Daniel Kahneman's "Thinking Fast and Slow" illustrates several examples of cognitive biases where the expected su

While I don't think the comparison is perfect, Daniel Kahneman's "Thinking Fast and Slow" illustrates several examples of cognitive biases where the expected suffering based on "area under the curve" (e.g., duration of pain in an non-anesthetized colonoscopy, or pain in a cold water torture test) is greater than the actual reported pain, mostly because we tend to report based on the later feelings (liberation, easing of pain) rather than peak pain or the integral of total pain. Perhaps another way of putting this is that confirmation of hope is a pretty powerful antidote to prior pain.