Quantifying Accuracy and Bias in Motive Introspection
Date of Award
Spring 2023
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Psychology
First Advisor
Crockett, Molly
Abstract
How well do people know their own motives in moral life? As veridical as introspection may seem intuitively, psychologists often question introspective reports, deeming them to be biased, inaccurate, or no more revealing than inferences made by outside observers. Such concerns about introspection are often cited, but less often tested. In this dissertation, I argue that introspective reports in the moral domain can be more accurate than often suggested, and showcase a method for quantifying peoples’ degree of introspective accuracy. Chapter 1 distinguishes between two related often confused capacities that guide our moral self-knowledge. Introspection generates knowledge about ongoing mental states, such as our conscious motives, and retrospection retrieves prior self-knowledge stored in long-term memory, such as memories of our past behavior. I distinguish the scope of introspection and retrospection, setting the stage for research on the accuracy and uniqueness of each as a source of self-knowledge. Chapter 2 examines the accuracy of retrospection by testing how well people recall their own past behavior in a morally-relevant setting—deciding how to allocate money between oneself and different strangers (i.e., dictator games). Here, I document a motivated bias in retrospection. People whose behavior violates their own moral standards tend to later recall being more generous than they really were. Chapter 3 then builds on this work by examining how accurately people’s introspective reports of their motives compare to the motives revealed through their actual choices. Here, I show robust evidence that people can accurately report their motives, even in moral settings, and show that such introspective accuracy can be dissociated from accuracy achieved through retrospective processes. At the same time, I document a similar motivated bias in introspection. Among people whose motives deviate from their own moral standards, I observe a tendency to distort one’s level of prosocial motivation. A key insight from Chapters 2 and 3 is that motivated biases are not driven by behavior alone, but by whether people believe they acted unfairly. However, one surprising pattern was how much variation people show in what they consider to be a fair amount to keep in dictator games. Chapter 4 further investigates the subjective basis for people’s personal standards, and the qualitative side of their motives. The research in this chapter makes consistent and powerful discovery: from 2020-2021, many people keeping money in dictator games reported being driven by dire financial circumstances. These data not only illuminate variation in subjective fairness standards, but more broadly suggest that behaviors in online economic games might be motivated not by greed, but financial need: a reality fundamentally missed by looking at people’s behavior in isolation. In Chapter 5, I close by considering future paths by which psychologists might deepen our understanding of the moral and introspective aspects of the mind, drawing on wisdom from the history of psychology.
Recommended Citation
Carlson, Ryan W., "Quantifying Accuracy and Bias in Motive Introspection" (2023). Yale Graduate School of Arts and Sciences Dissertations. 890.
https://elischolar.library.yale.edu/gsas_dissertations/890
