In Daniel Kahnemahn’s Thinking, Fast and Slow, he draws a distinction between the ‘experiencing self’ and the ‘remembering self.’ Like most conceptual models it is imprecise, but it nonetheless illustrates a few useful points. The experiencing self is the self that is concerned with our present state of being. The remembering self looks back at our past experiences and evaluates them. (Of course, one can say that we are then experiencing a memory, but that misses the point, which is simply that our experience at the time of the event and our experience of the memory of the event will often judge the event very differently).
To illustrate this, Kahnemahn describes an experiment involving submerging one’s hand in a bucket of ice water. Each subject was told there would be three trials. For one of the trials, the subject’s hand was immersed in water cold enough to be painful for 60 seconds. The other trial the hand was immersed in the same temperature water for 60 seconds, but then after 60 seconds warmer water was allowed into the bucket, and then the hand was removed at 90 seconds. Those two trials were given in random order, such that half of the participants had the 60 second trial first, and half had the 90 second trial first. There was no actual third trial, but participants were told for the third trial that they would have a choice between the 60 second trial and the 90 second trial.
The chart illustrates the level of pain felt during the trial, with the blue line representing the 60 second mark. It’s not an exact replica of the pain recorded by participants, since Kahnemahn doesn’t provide that data, but the point is that the pain level decreases after 60 seconds, and the last 30 seconds are still painful, but not as painful.
From the perspective of the experiencing self the goal is generally to avoid pain. Even though the last 30 seconds are less painful than the first 60, they are still painful. Since the first 60 seconds of pain are identical in both the 60 second trial and the 90 second trial, there is no good reason to voluntarily endure an additional 30 seconds of pain. But that’s not how the experimental subjects responded:
Fully 80% of the participants who reported that their pain diminished during the final phase of the longer episode opted to repeat it, thereby declaring themselves willing to suffer 30 seconds of needless pain in the anticipated third trial.
This is because the remembering self is very bad at judging the duration of pain. Instead of looking at the area under the curve as the experiencing self does, the remembering self judges pain based on the peak amount of pain, and the amount of pain at the end.
Kahnemahn, for the record, has a self-admitted bias towards the experiencing self, although he recognizes that both should be taken into consideration. He also notes, for the economists, that “The evidence presents a profound challenge to the idea that humans have consistent preferences and know how to maximize them, a corner of the rational-agent model.”
The idea of the two selves also has implications for ethics. Utility, because of its focus on pain and pleasure, lends itself quite naturally to the experiencing self. While attempts have been made to broaden Utility beyond simple pain and pleasure (Mill’s famous argument that it is better to be a dissatisfied Socrates than a satisfied pig, among others), the basic outlook is still one that looks at the aggregate utility, defined primarily as pain and pleasure, of an individual during their lifetime.
Virtue ethics, by contrast, is much closer to evaluating a life well-lived from the perspective of one remembering their life. Rather than focusing on individual moments of pain and pleasure, one focuses on development into a virtuous human being.
It is fascinating the degree to which ethical systems are determined by the information they deem relevant. (Rawl’s veil of ignorance provides a very direct example of screening out information, while utility and virtue ethics provide examples of focusing on certain types of information to the potential exclusion of others).
None of this is to suggest that one perspective is correct and the other one incorrect. It is important to keep both in mind. My personal bias is towards evaluating based on the remembering self, and even knowing what’s going on I would choose the 90 second trial. That said, it’s usually best to have a relatively broad informational perspective (but not so broad as to be paralyzing and indecisive). Both the experiencing self and the remembering self should be considered in evaluating ethics and public policy.