On a quick skim, an element that seems to be missing is that having emotions which cause you to behave ‘irrationally’ can in fact be beneficial from a rational perspective.
For example, if everyone knows that, when someone does you a favor, you’ll feel obligated to find some way to repay them, and when someone injures you, you’ll feel driven to inflict vengeance upon them even at great cost to yourself—if everyone knows this about you, then they’ll be more likely to do you favors and less likely to injure you, and your expected payoffs are probably higher than if you were 100% “rational” and everyone knew it. I believe this is in fact why we have the emotions of gratitude and anger, and I think various animals have something resembling them. Put it this way: carrying out threats and promises is “irrational” by definition, but making your brain into a thing that will carry out threats and promises may be very rational.
So you could call these emotions “irrational” or the thoughts they lead to “biased”, but I think that (a) likely pushes your thinking in the wrong direction in general, and (b) gives you no guidance on what “irrational” emotions are likely to exist.
Thanks for the feedback @localdeity. I agree that my article could be read as implying that emotions are inherently irrational and that, from an evolutionary perspective, emotions have underlying logics (for instance anger likely exists to ensure that we enforce our social boundaries against transgression). This reading does not reflect my views however.
My scheme follows decision theory by assuming that we can assign an “objective” utility value to each action/option. This utility value should encompass everything—including whatever benefits may be reflected in the logics underlying emotions. Thus, there shouldn’t be any benefit that emotions provide that is not included in these utility values. There are times when our emotions are aligned with those actions that maximize expected utility, but this is not guaranteed. Whenever an emotion goads us to act in line with utility maximization we can call that emotion “rational.” When the emotion spurs us to act in a way that conflicts with our best interest (all things considered), we can call that emotion “irrational”.
My goal in this article was not to argue that emotions are fundamentally irrational. Emotions operate according to their own internal rules. These rules are more akin to pattern-response than to the sorts of calculations prescribed by decision theory. This article tries to integrate these effects into decision theory to create a model of human behavior that is qualitatively more accurate.
On a quick skim, an element that seems to be missing is that having emotions which cause you to behave ‘irrationally’ can in fact be beneficial from a rational perspective.
For example, if everyone knows that, when someone does you a favor, you’ll feel obligated to find some way to repay them, and when someone injures you, you’ll feel driven to inflict vengeance upon them even at great cost to yourself—if everyone knows this about you, then they’ll be more likely to do you favors and less likely to injure you, and your expected payoffs are probably higher than if you were 100% “rational” and everyone knew it. I believe this is in fact why we have the emotions of gratitude and anger, and I think various animals have something resembling them. Put it this way: carrying out threats and promises is “irrational” by definition, but making your brain into a thing that will carry out threats and promises may be very rational.
So you could call these emotions “irrational” or the thoughts they lead to “biased”, but I think that (a) likely pushes your thinking in the wrong direction in general, and (b) gives you no guidance on what “irrational” emotions are likely to exist.
Thanks for the feedback @localdeity. I agree that my article could be read as implying that emotions are inherently irrational and that, from an evolutionary perspective, emotions have underlying logics (for instance anger likely exists to ensure that we enforce our social boundaries against transgression). This reading does not reflect my views however.
My scheme follows decision theory by assuming that we can assign an “objective” utility value to each action/option. This utility value should encompass everything—including whatever benefits may be reflected in the logics underlying emotions. Thus, there shouldn’t be any benefit that emotions provide that is not included in these utility values. There are times when our emotions are aligned with those actions that maximize expected utility, but this is not guaranteed. Whenever an emotion goads us to act in line with utility maximization we can call that emotion “rational.” When the emotion spurs us to act in a way that conflicts with our best interest (all things considered), we can call that emotion “irrational”.
My goal in this article was not to argue that emotions are fundamentally irrational. Emotions operate according to their own internal rules. These rules are more akin to pattern-response than to the sorts of calculations prescribed by decision theory. This article tries to integrate these effects into decision theory to create a model of human behavior that is qualitatively more accurate.