The reference class problem is your problem too

The reference class problem arises when we want to assign a probability to a proposition (or sentence, or event) X, which may be classified in various ways, yet its probability can change depending on how it is classified. The problem is usually regarded as one specifically for the frequentist interpretation of probability and is often considered fatal to it. I argue that versions of the classical, logical, propensity and subjectivist interpretations also fall prey to their own variants of the reference class problem. Other versions of these interpretations apparently evade the problem. But I contend that they are all “no-theory” theories of probability - accounts that leave quite obscure why probability should function as a guide to life, a suitable basis for rational inference and action. The reference class problem besets those theories that are genuinely informative and that plausibly constrain our inductive reasonings and decisions. I distinguish a “metaphysical” and an “epistemological” reference class problem. I submit that we can dissolve the former problem by recognizing that probability is fundamentally a two-place notion: conditional probability is the proper primitive of probability theory. However, I concede that the epistemological problem remains.     

Answering the Bayesian Challenge

This essay answers the “Bayesian Challenge,” which is an argument offered by Bayesians that concludes that belief is not relevant to rational action. Patrick Maher and Mark Kaplan argued that this is so because there is no satisfactory way of making sense of how it would matter. The two ways considered so far, acting as if a belief is true and acting as if a belief has a probability over a threshold, do not work. Contrary to Maher and Kaplan, Keith Frankish argued that there is a way to make sense of how belief matters by introducing a dual process theory of mind in which decisions are made at the conscious level using premising policies. I argue that Bayesian decision theory alone shows that it is sometimes rational to base decisions on beliefs; we do not need a dual process theory of mind to solve the Bayesian Challenge. This point is made clearer when we consider decision levels: acting as if a belief is true is sometimes rational at higher decision levels.

---

Approval-directed agency and the decision theory of Newcomb-like problems

Decision theorists disagree about how instrumentally rational agents, i.e., agents trying to achieve some goal, should behave in so-called Newcomb-like problems, with the main contenders being causal and evidential decision theory. Since the main goal of artificial intelligence research is to create machines that make instrumentally rational decisions, the disagreement pertains to this field. In addition to the more philosophical question of what the right decision theory is, the goal of AI poses the question of how to implement any given decision theory in an AI. For example, how would one go about building an AI whose behavior matches evidential decision theory’s recommendations? Conversely, we can ask which decision theories (if any) describe the behavior of any existing AI design. In this paper, we study what decision theory an approval-directed agent, i.e., an agent whose goal it is to maximize the score it receives from an overseer, implements. If we assume that the overseer rewards the agent based on the expected value of some von Neumann–Morgenstern utility function, then such an approval-directed agent is guided by two decision theories: the one used by the agent to decide which action to choose in order to maximize the reward and the one used by the overseer to compute the expected utility of a chosen action. We show which of these two decision theories describes the agent’s behavior in which situations.

     

On decision making without likelihood judgment

Subjective expected utility, prospect theory and most other formal models of decision making under uncertainty are probabilistic: they assume that in making choices people judge the likelihood of relevant uncertainties. Clearly, in many situations people do indeed judge likelihood. However, we present studies suggesting that there are also many situations in which people do not judge likelihood and instead base their decisions on intuitively generated, non-probabilistic rules or rationales. Thus, we argue that real-world situations are of two types. In situations eliciting a probabilistic mindset, people rely on judgments of likelihood. In situations eliciting a non-probabilistic mindset, they neglect judgments of likelihood. We suggest three factors that may influence the tendency towards either probabilistic or non-probabilistic mindsets. We also outline how extant probabilistic theories may be complemented by non-probabilistic models.     

The role of moral utility in decision making: An interdisciplinary framework

What decisions should we make? Moral values, rules, and virtues provide standards for morally acceptable decisions, without prescribing how we should reach them. However, moral theories do assume that we are, at least in principle, capable of making the right decisions. Consequently, an empirical investigation of the methods and resources we use for making moral decisions becomes relevant. We consider theoretical parallels of economic decision theory and moral utilitarianism and suggest that moral decision making may tap into mechanisms and processes that have originally evolved for nonmoral decision making. For example, the computation of reward value occurs through the combination of probability and magnitude; similar computation might also be used for determining utilitarian moral value. Both nonmoral and moral decisions may resort to intuitions and heuristics. Learning mechanisms implicated in the assignment of reward value to stimuli, actions, and outcomes may also enable us to determine moral value and assign it to stimuli, actions, and outcomes. In conclusion, we suggest that moral capabilities can employ and benefit from a variety of nonmoral decision-making and learning mechanisms.     

On bad decisions and disconfirmed expectancies: The psychology of regret and disappointment

Decision outcomes sometimes result in negative emotions. This can occur when a decision appears to be wrong in retrospect, and/or when the obtained decision outcome does not live up to expectations. Regret and disappointment are the two emotions that are of central interest in the present article. Although these emotions have a lot in common, they also differ in ways that are relevant to decision making. In this article we review theories and empirical findings concerning regret and disappointment. We first discuss how regret and disappointment differ with respect to their antecedent conditions, appraisals, and phenomenology. We also discuss possible behavioural consequences of experiencing these emotions. Next, we consider how the anticipation of regret and disappointment may influence decision making. We use regret and disappointment theory, developed by the economists Bell (1982, 1985), and Loomes and Sugden (1982, 1986, 1987), as a framework for our discussion. Finally, we argue that combining the theoretical approaches and research paradigms of behavioural decision theory with emotion theories will significantly increase our knowledge of antecedents and consequences of emotions.     

Family conflict shapes how adolescents take risks when their family is affected

Numerous studies have established that the social context greatly affects adolescent risk taking. However, it remains unexplored whether adolescents' decision‐making behaviors change when they take risks that affect other individuals such as a parent. In the current study, we sought to investigate how the social context influences risky decisions when adolescents' behavior affects their family using a formalized risk‐taking model. Sixty‐three early adolescents (M_age = 13.3 years; 51% female) played a risk‐taking task twice, once during which they could make risky choices that only affected themselves and another during which their risky choices only affected their parent. Results showed that adolescents reporting high family conflict made more risky decisions when taking risks for their parent compared to themselves, whereas adolescents reporting low family conflict made fewer risky decisions when taking risks for their parent compared to themselves. These findings are the first to show that adolescents change their decision‐making behaviors when their risks affect their family and have important implications for current theories of adolescent risk taking.     

A Peircean Theory of Decision

It is sometimes argued that the fact that possession of perfect knowledge about the future is impossible, means that it is impossible for decisions to be rational. This reasoning is fallacious. If rationality is given a new interpretation, then decisions can be considered rational. A theory of decision that has as its basis Peirce’s theory of abduction can provide a new way of understanding decisions as rational processes. The Peircean theory of decision (i) considers decisions as part of a complete strategy, and (ii) shows that decision making is governed by the same rules as scientific abduction. These rules are neither permissive rules like rules of deductive inference nor predictive like laws of nature, but rather genuine laws of conduct that determine what step should be made, if a given end is to be reached. This revised version was published online in June 2006 with corrections to the Cover Date.     

Integrating Criminological and Decision Research Theory: Implications for Understanding and Addressing Crime in Marginalized Communities

Minority groups are significantly overrepresented in crime. Theories of racial differences in crime developed using two separate and distinct approaches that highlight either increased exposure to criminogenic factors at the individual level or greater risk of crime due to disadvantaged neighborhood conditions. Neighborhood theories describe how structural disadvantage disrupts neighborhood social processes and produces oppositional street cultures. In the article, we advance theorizing on race and crime by linking the neighborhood experience to individual-level decision making via new conceptualizations of culture. Rather than a “values as goals” view of culture, culture may include a “tool kit” of ways to solve problems and this cultural toolkit may, in turn, influence how an individual makes decisions. Specifically, culturally learned toolkits may increase flaws in the decision process (e.g., fast and intuitive rather than deliberate decision processes, the use of decision heuristics) to produce more crime, which would explain the association between race and crime. We integrate this conceptualization of culture and these flaws in the decision-making process into rational choice theory at the individual level and describe how they may be exacerbated in disadvantaged neighborhood contexts. Implications for understanding race and crime and directions for future research are discussed.     

Adult Age Differences in Dual Information Processes: Implications for the Role of Affective and Deliberative Processes in Older Adults' Decision Making

ABSTRACT— Age differences in affective/experiential and deliberative processes have important theoretical implications for judgment and decision theory and important pragmatic implications for older-adult decision making. Age-related declines in the efficiency of deliberative processes predict poorer-quality decisions as we age. However, age-related adaptive processes, including motivated selectivity in the use of deliberative capacity, an increased focus on emotional goals, and greater experience, predict better or worse decisions for older adults depending on the situation. The aim of the current review is to examine adult age differences in affective and deliberative information processes in order to understand their potential impact on judgments and decisions. We review evidence for the role of these dual processes in judgment and decision making and then review two representative life-span perspectives (based on aging-related changes to cognitive or motivational processes) on the interplay between these processes. We present relevant predictions for older-adult decisions and make note of contradictions and gaps that currently exist in the literature. Finally, we review the sparse evidence about age differences in decision making and how theories and findings regarding dual processes could be applied to decision theory and decision aiding. In particular, we focus on prospect theory ( Kahneman & Tversky, 1979 ) and how prospect theory and theories regarding age differences in information processing can inform one another.     

Analytic Confidence and Political Decision‐Making: Theoretical Principles and Experimental Evidence From National Security Professionals

When making decisions under uncertainty, it is important to distinguish between the probability that a judgment is true and the confidence analysts possess in drawing their conclusions. Yet analysts and decision‐makers often struggle to define “confidence” in this context, and many ways that scholars use this term do not necessarily facilitate decision‐making under uncertainty. To help resolve this confusion, we argue for disaggregating analytic confidence along three dimensions: reliability of available evidence, range of reasonable opinion, and responsiveness to new information. After explaining how these attributes hold different implications for decision‐making in principle, we present survey experiments examining how analysts and decision‐makers employ these ideas in practice. Our first experiment found that each conception of confidence distinctively influenced national security professionals' evaluations of high‐stakes decisions. Our second experiment showed that inexperienced assessors of uncertainty could consistently discriminate among our conceptions of confidence when making political forecasts. We focus on national security, where debates about defining “confidence levels” have clear practical implications. But our theoretical framework generalizes to nearly any area of political decision‐making, and our empirical results provide encouraging evidence that analysts and decision‐makers can grasp these abstract elements of uncertainty.     

The probabilistic nature of preferential choice

Previous research has developed a variety of theories explaining when and why people's decisions under risk deviate from the standard economic view of expected utility maximization. These theories are limited in their predictive accuracy in that they do not explain the probabilistic nature of preferential choice, that is, why an individual makes different choices in nearly identical situations, or why the magnitude of these inconsistencies varies in different situations. To illustrate the advantage of probabilistic theories, three probabilistic theories of decision making under risk are compared with their deterministic counterparts. The probabilistic theories are (a) a probabilistic version of a simple choice heuristic, (b) a probabilistic version of cumulative prospect theory, and (c) decision field theory. By testing the theories with the data from three experimental studies, the superiority of the probabilistic models over their deterministic counterparts in predicting people's decisions under risk become evident. When testing the probabilistic theories against each other, decision field theory provides the best account of the observed behavior.     

Knowledge of Our Own Beliefs

There is a widespread view that in order to be rational we must mostly know what we believe. In the probabilistic tradition this is defended by arguments that a person who failed to have this knowledge would be vulnerable to sure loss, or probabilistically incoherent. I argue that even gross failure to know one's own beliefs need not expose one to sure loss, and does not if we follow a generalization of the standard bridge principle between first‐order and second‐order beliefs. This makes it possible for a subject to use probabilistic decision theory to manage in a rational way cases of potential failure of this self‐knowledge, as we find in implicit bias. Through such cases I argue that it is possible for uncertainty about what our beliefs are to be not only rationally permissible but advantageous.     

Processes of patient decision making: Theoretical and methodological issues

Abstract

There is a paucity of research studying patients' decision making processes. Traditional normative approaches investigating human decision have presented rational, analytic processing as a ‘gold standard’ for decision making with decision aids developed to facilitate such thinking. This paradigm has been challenged by the recent emergence of naturalistic decision making. Naturalistic approaches argue for techniques modelled closer to how decisions are actually made in ‘real life’. Early work in this evolving field suggests the importance of more automatic, intuitive processing such as the use of heuristic short-cuts. This paper discusses the contribution of naturalistic decision theory and assesses its usefulness as an alternative to classical decision approaches in investigating patient decision making. Approaches for measuring underlying cognitive processes are also critically discussed.     

Free Will and Mental Powers

In this paper, we investigate how contemporary metaphysics of powers can further an understanding of agent-causal theories of free will. The recent upsurge of such ontologies of powers and the understanding of causation it affords promises to demystify the notion of an agent-causal power. However, as we argue pace (Mumford and Anjum in Analysis 74:20–25, 2013; Am Philos Q 52:1–12, 2015a), the very ubiquity of powers also poses a challenge to understanding in what sense exercises of an agent’s power to act could still be free—neither determined by external circumstances, nor random, but self-determined. To overcome this challenge, we must understand what distinguishes the power to act from ordinary powers. We suggest this difference lies in its rational nature, and argue that existing agent-causal accounts (e.g., O’Connor in Libertarian views: dualist and agent-causal theories, Oxford University Press, Oxford, 2002; Lowe in Personal agency: the metaphysics of mind and action, Oxford University Press, Oxford, 2013) fail to capture the sense in which the power to act is rational. A proper understanding, we argue, requires us to combine the recent idea that the power to act is a ‘two-way power’ (e.g., Steward in A metaphysics for freedom, Oxford University Press, Oxford, 2012b; Lowe (in: Groff, Greco (eds) Powers and capacities in philosophy: the new aristotelianism, Routledge, New York, 2013) with the idea that it is intrinsically rational. We sketch the outlines of an original account that promises to do this. On this picture, what distinguishes the power to act is its special generality—the power to act, unlike ordinary powers, does not come with any one typical manifestation. We argue that this special generality can be understood to be a feature of the capacity to reason. Thus, we argue, an account of agent-causation that can further our understanding of free will requires us to recognize a specifically rational or mental variety of power.