Decisions and plans (Part III of Baron, 1994)

In a previous post I said I would cover parts II and III of Baron’s guide to the psychology of human decision-making entitled Thinking and Deciding (1994 2nd edition). Part III is about decision making, including both the decisions that affect only the decision maker and the decisions that affect others, that is, decisions that raise moral questions. It also examines long-term planning, with special emphasis on the choice of personal goals. The part is particularly concerned with the inference of a course of action from our goals and from evidence concerning the consequences of our options for achieving them.

Normative theory of utility    … concerns the theory of how we should choose among possible actions under ideal conditions.  It is concerned with inference not search. The basic idea is that utility – or desirability of outcomes – should be maximized, so the best decision is the one that best helps us achieve our goals. In the case of conflict it is a matter of maximizing total utility. The concept of utility represents whatever people want to achieve: virtue, productive work, enlightenment, respect, love. It is not the same as monetary reward, happiness or satisfaction. wide range of human goals. This model serves as a theoretical ideal that we can use to justify prescriptive models for decision analysis, whether in business, government, or medicine. There are three main aspects: (1) expected-utility theory, which considers the trade-off between probability and utility; (2) Multiattribute Utility Theory (MAUT), which considers the trade-off between different goals; and (3) decision that involve conflict among the goals of different people. Normative utility theory helps justify the prescriptive model of actively open-mindedness, because it could find a neglected possibility with a higher expected utility. The same applies to evidence, goals, and biased weighing of evidence. Ideally, belief formation should take place after a thorough search and unbiased evaluation of evidence. Search thoroughness must be in proportion to the expected benefits. (Ch. 16 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Decision analysis       … is normative utility theory of decision making applied to real-life situations such as airport location, nuclear waste disposal, and school desegregation. Decision analysis, as discussed in this chapter, attempts to assess the utilities and probabilities of outcomes to calculate the expected utility of various options. It is not a simple cost-benefit analysis, because that would confuse market value with subjective value. Instead, outcomes are represented by how much they are estimated to help achieve a given set of goals. Decision analysis can also be used as a second opinion or to evaluate decisions after they are made or to resolve conflict among different decision makers. To be useful, he need for thorough search must be respected. The result of the analysis must be contrasted to the initial intuition to see why they differ. Various techniques of utility measurement are discussed, including prediction, direct scaling, difference measurement, a standard gambling approach, conjoint measurement, and Multiattribute Utility Theory (MAUT). Special emphasis is placed on aspects such as trade-offs and the value of human life. When it comes to teaching decision making to 8th-grade students a promising approach may be to give additional emphasis to the types of errors that people make in the absence of decision analysis, such as short-sightedness (neglect of relevant goals), single-mindedness (neglect of alternatives and evidence), impulsiveness, and neglect of probability. (Ch. 17 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Choice under uncertainty       This chapter provides descriptive theory on the way in which we respond to risk or uncertainty about the outcome of an option. We tend to be averse to risks when the alternative is a gain that is certain. Biases in decisions under uncertainty include: (1) neglect of probability; (2) preference reversals and contingent weighting; (3) the Allais paradox; (4) prospect theory, which provides the simplest possible ‘explanation’ for observed violations of expected-utility theory; (5) the certainty effect; (6) anticipated emotional effects of outcomes, including regret, rejoicing, disappointment, elation and their rationality in decision making; (7) the ambiguity effect in the case of unknown risks, such as those of nuclear power and genetic engineering; (8) aversion to missing information; (9) deferral under conditions of uncertainty or conflict. It is concluded that some apparent violations, such as those caused by regret, are not necessarily violations at  all, since an overly simple analysis could have neglected real emotional consequences. Other violations might result from using generally useful heuristics in situations where they are harmful in achieving our goals. In the case of ambiguity deferral points to the need for acquiring more information. All of the violations occur in holistic judgment, but can be avoided if utility theory and decision analysis are used. (Ch. 18 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Choice under certainty        … concerns decisions in which we analyse as if we knew what the outcomes would be. Many biases found in such decisions involve an incomplete search for goals. In the extreme form only a single goal is considered: the “bottom line,” the defeat of communism, the protection of abstract “rights.” Such singlemindedness often involves social roles, as in the case of soldiers, scholars, politicians, lawyers and the like, but can be cured by actively open-minded searching for other goals that may on some occasions outweigh their professional goals. Attention to a single, dominant dimension can lead to unnecessary disagreement between groups and to violations of each individual’s true preferences. Making trade-offs helps – when we have the time to do it. Biases (or heuristics) in making trade-offs include: (1) the prominence effect; (2) attitude to (non-)compensatory decision strategies; (3) the status quo (endowment) effect; (4) omission (inaction) bias; (5) opportunity costs (easily neglected); (6) reference point framing; (7) integration vs. segregation in mental accounting; (8) reference price manipulation; (9) intransitivity of preferences; (10) elimination by attributes; (11) asymmetric dominance; and (12) compromise. Most of these heuristics save time, so we can use them provided we do so knowingly. Often a good way to avoid getting in trouble is to develop counterheuristics (“What would we do if …”). (Ch. 19 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Quantitative judgment       This chapter examines further our ability to make consistent quantitative judgments without the aid of formal theories. Quantitative judgment seeks to improve on unaided holistic judgment by evaluating cases on the basis of a set of evidence and given goals with respect to a set of criteria or attributes, somewhat similar to Multiattribute Utility Theory (MAUT). It consists of rating (numbers, grades), ranking (order), and classifying (assigning to one of two groups). Topics of quantitative judgment include: (1) multiple linear regression; (2) the lens model (a statistical prediction model that looks at decision making and goal attainment on the basis of cues for piecemeal judgments and their combination into an overall score); (3) the mechanism of judgment; (4) non-numerical impression formation; (5) averaging, adding, and number of cues; (6) representativeness in numerical prediction; and (7) classification. The main lesson of the research on quantitative judgment is that people are overconfident in their holistic judgments. Increased awareness of this ought to make jus more tolerant of one another, as well as more accepting of new (computer-based) method for decision making. This is less obvious when moral issues are involved (public policy, personal lives) we tend to disagree over. (Ch. 20 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Moral thinking       Certain issues obviously involve moral questions, including abortion, property rights, capital punishment, aid for the poor, and almost all other political issues, but wherever our choices affect the utilities of others, a moral decision must be made, including in our own lives. In the first part of this chapter, the nature of moral judgment is examined, with special attention to their universality. In the second part, it is argued that utilitarianism provides a way to understand and analyse moral issues in terms of the consequences of decisions. It also provides a way to think critically and constructively about our basic moral intuitions, the beliefs we acquired as children. The distinction between “intuitive” (normative) and “critical” (pre-/descriptive) moral thinking is examined in detail. According to R.M. Hare we should usually follow our moral intuitions, but we should also, however, be able to think critically about them and about decisions in particularly important cases. Kohlberg’s research suggest that Hare’s distinction corresponds to different “levels” of moral thinking, which suggests that many people are not capable of critical-level thinking about many issues. Of course, some of the biases in moral thinking are exactly the biases that occur in all thinking: failure to consider goals (in this case, moral goals), biased search for evidence and biased inference and failure to examine critically our own intuitive rules. (Ch. 21 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Fairness and justice         Moral issues range from the allocation of organs for transplantation to patients to the assignment of penalties for companies that harm the environment. The question is: who gets what in terms of rewards and punishments. Utilitarianism holds that fairness is whatever yields the best overall consequences in the long run, summed across everyone. Alternative normative theories hold that fairness is an extra consideration, beyond consequences. Such alternatives can even work prescriptively in a utilitarian context, provided people have a good understanding of the relation between fairness (to the individual) and utility (for the greater good). Experiments have shown that people desire to see fairness and will try to bring it about, even when utility is not maximized or when it demands a sacrifice on their part. Equity theorists also maintain that people want rewards and punishments to be proportional to “inputs” of some sort (effort, ability, end result). The key question is: what is the ultimate goal? Perhaps the most important problem is that people do not understand the justifications of the principles that they endorse, even when these principles can be linked to utilitarianism or some other theory. (Ch. 22 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Commons dilemmas        Because so many situations can be analysed as social dilemmas, much of the philosophy and psychology of morality is contained in decisions involving problem situations in which the narrow self-interest of each person in a group conflicts with the interest of the group as a whole. Examples include: cheating on one’s taxes, the arms race, accepting bribes, pollution, having too many children. Economic systems induce people to do their share of the labour and moderate their consumption. More cooperation would solve many human problems, including war. The prisoner’s dilemma is a standard example of why two rational persons might not cooperate. We often fail to recognize that we inappropriately decide to defect instead of cooperate. With a utilitarian approach it may be possible to avoid such biases. Prescriptively, we can approximate a normative decision theory by drawing on traditional modes of cooperative behaviour and by using certain heuristics that counteract the most common causes of defection and insufficient thought. The problems with three theories are discussed: cooperative theory, utilitarianism, and self-interest theory, which comes in two forms: (1) nobody cares about other people; and (2) individuals should look out for their own interest. All theories have limitations, even an explicitly more balanced one such as weighted utilitarianism. Important motives in social dilemmas include: altruism, competition, fairness, equality, envy, fear, and greed. (Ch. 23 of Baron, J. 1994 Thinking and deciding 2nd ed.)

Decisions about the future       Planning and the potential conflict between goals for future outcomes and goals for immediate outcomes are roughly analogous to those between self and others: we can think of our future selves as somewhat different people form our present selves. We tend to be biased in favour of our present self and need methods of self-control. A distinction is made between plans and policies. Some policies are made all at once, whereas others may evolve from precedents. Plans and policies create new goals or change the strength of old ones. The chapter deals mostly with personal goals, including the choice of personal goals. We must strike a balance between the risks of striving for high achievement and the principle that our goals should be achievable. There are four good reasons to stick to plans: (1) the involvement of others; (2) the value of dependability; (3) increased chances of success; and (4) the value of trust in ourselves. Bad reasons for sticking to plans include: (1) the status quo bias; (2) the sunk-cost effect; and (3) escalation of commitment. Other subjects include the rationality of discounting, methods of self-control, emotional aspects, identity formation and the importance of morality in choosing the basic goals for our lives. A key consideration is that the essence of morality is not self-sacrifice but benefiting others, and there are many ways in which it is possible for individuals to “do well by doing good.” (Ch. 24 of Baron, J. 1994 Thinking and deciding 2nd ed.)


About csl4d
This entry was posted in General. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s