Decision Making

LEADING FIGURES IN THE FIELD

Herbert Simon (1916–2001), who was associated with Carnegie Mellon University, was one of the first to point out the limits of human reasoning as it relates to decision-making. He argued against the normative model using the tenets of Information Processing theory and proposed the construct of bounded rationality, which specifies that the human mind is incapable of processing all aspects of all possible options in a given situation. Instead, the mind must simplify the process by reducing the number of options that are considered (e.g., five or less) and the number of attributes of these options (e.g., just the price, gas mileage, and color of potential cars to buy). He argued further that decision makers have a tendency to engage in “satisficing” when they think about their options; that is, instead of waiting to choose an option until all options have been carefully and exhaustively considered, decision makers consider options one by one in sequence until they reach one that is good enough. Once this good enough option is encountered, none of the remaining options is examined. This tendency to satisfice runs against the prescriptions of the normative model because it is possible that the best option that should have been selected was among those that were not yet evaluated. The notion of bounded rationality was extremely influential because of its applicability across disciplines. Scholars in the field of economics were particularly enamored with this notion and eventually nominated Simon for the Nobel Prize in economics (which he subsequently was awarded).

Simon's application of Information Processing theory to decision making influenced several generations of decision theorists that came along in the 1970s and 1980s. The idea that decision makers engage in cognitive shortcuts soon became pervasive in the literature. In some of the seminal studies in this area, Daniel Kahneman of Princeton University and the late Amos Tversky of Stanford University proposed a series of cognitive heuristics that people use when they process information relevant to decisions. These heuristics, in turn, produce systematic biases to respond in particular ways, and these biases affect their choices. For example, the so-called representativeness heuristic is operative whenever people are presented with a category of things (e.g., cars) or events (e.g., random sequences). All categories have prototypical instances that are assumed to be highly representative of the categories (e.g., the sequence 3, 19, 26, 29, 34, 40 in a lottery drawing for random events), as well as instances that are assumed to be less representative (e.g., the sequence 1, 2, 3, 4, 5, 6; in reality, both sequences are equally probable). Incorrect or not, the assumption that an instance is representative prompts people to think it is more likely to occur than the less representative instance. In the case of random number sequences, the representativeness heuristic may make gamblers decide to place a large bet right after observing the occurrence of a less representative sequence because the occurrence of the less probable sequence makes them think that the more probable sequence will occur next.

Kahneman and Tversky also gained prominence for discovering the fact that people will make different choices regarding the same information depending on how it is framed (e.g., “400 out of 1000 people will be saved by a drug” versus “600 will die”) and for their Prospect Theory. Other prominent scholars who have examined the prevalence and consequences of these and other reasoning biases on decision making are Baruch Fischhoff of Carnegie Mellon, Paul Slovic of Decision Research in Oregon, Jonathan Baron of the University of Pennsylvania, Hal Arkes of Ohio State University, Valerie Reyna of Cornell University, and Keith Stanovich of the Ontario Institute for Studies in Education.

In addition to stressing constructs such as memory limitations and cognitive shortcuts, the Information Processing (IP) approach also emphasizes other important aspects of performance that have a bearing on decisions such as cognitive strategies and metacognition. To illustrate one of these further applications, John Payne and James Bettman of Duke University examined the systematic strategies people use when they examine a number of options that are placed before them (e.g., 20 possible apartments that could be rented). The IP approach eventually developed a natural affinity to the idea that many aspects of human cognition (including cognitive shortcuts) were selected via evolutionary processes because of their adaptive value. Thus, the field took a 180-degree turn away from the original assumption that deviations from the normative model were a bad thing to the subsequent claim that these deviations were actually adaptive and likely to lead to environmental success.

Scholars such as Gerd Gigerenzer argued that in many situations, processing too many things could lead to serious negative consequences. For example, emergency room (ER) doctors could look for quite a number of symptoms when a cardiac patient enters the ER, but three particular signs are highly diagnostic of a heart attack. Spending time collecting data on the other less diagnostic symptoms would surely lead to an increase in ER deaths. Gigerenzer argued further that the most useful cognitive processes that people engage in during decisions operate at an unconscious level. Engaging in the conscious, systematic consideration of options would either be a waste of time or lead to lower levels of goal attainment, according to Gigerenzer.

Between the 1960s and 1990s, researchers who studied decision making in adults frequently were confronted with the criticisms that (a) much of the experimental work in decision making utilizes laboratory tasks that have little relevance to the real world, and (b) the IP inspired approaches seem to neglect the potentially important role of motivational or emotional factors in decision making. Perhaps in response to these criticisms, many of the most prominent figures in the field refocused their energies on explaining real world phenomena (e.g., actual home buyers assessing the risk of living near a nuclear reactor), and several also developed new theoretical models that give a more prominent role to emotions. Examples include Fischhoff, Slovic, Baron, and Arkes.

The portrait that has emerged since the 1990s is that of an adult decision maker who relies on cognitive shortcuts and emotional processing to make many ordinary decisions in ways that often lead to goal attainment. These same shortcuts and emotional responses can, however, lead to poor decisions in some circumstances. For example, there are times when it is a good idea to consider options more fully and not be distracted or misled by transient emotional states.

THE DEVELOPMENT OF DECISION MAKING

There are far fewer studies of the development of decision making in children and adolescents than studies of decision making in adults. Nevertheless, it is possible to draw some tentative conclusions about age changes that have the potential to affect the quality of decisions made by children and adolescents. Before doing so, however, it is important to note that developmentalists have not been concerned with age changes in the extent to which children make decisions (i.e., all children make many decisions per day at all ages), but rather, with age changes in the quality of children's decisions. Charting age changes in quality, of course, requires that one have a working model of what decision making competence entails.

Because of important shortcomings in the aforementioned normative model, some scholars have shifted from using the normative model as a guide to evaluating decision quality and moved towards defining decision competence in terms of environmental success. In particular, these scholars argue that a skilled decision maker is someone who knows the difference between options that are likely to lead to goal attainment (good options) and options that are unlikely to lead to goal attainment (not-so-good options). If so, the question then becomes one of identifying characteristics of decision makers that help them recognize or discover good options in a particular situation.

Some of the characteristics that have been proposed are (a) knowledge and experience (i.e., more knowledgeable individuals are likely to correctly predict the consequences of particular actions), (b) the tendency to seek advice from the right people when personal knowledge is lacking, (c) the tendency to pursue adaptive goals that are likely to promote physical health, emotional health, or financial well-being, (d) the tendency to prefer options that satisfy multiple goals as opposed to options that satisfy only a single goal (e.g., find a car that is attractive, relatively inexpensive, and reliable), (e) the tendency to learn from decision-making successes and failures, (f) the tendency to engage in effortful and thorough examination of options and consequences for important decisions, but not expend considerable effort considering options for unimportant decisions, and (g) the ability to regulate one's emotional and impulsive tendencies in ways that keep these tendencies from interfering with appropriate consideration of options and consequences.

Although, as of the early 2000s, the evidence was still emerging regarding age differences in such characteristics, some studies suggested that older adolescents and adults are more likely than younger adolescents and children to (a) understand the difference between options likely to satisfy multiple goals and options likely to satisfy only a single goal, (b) anticipate a wider array of consequences of their actions, (c) evaluate their options in systematic ways and apply effortful strategies only for important decisions, and (d) learn from their decision-making successes and failures. Unfortunately, adolescents have also been found to seek advice less often from knowledgeable individuals than children, and they are more likely to pursue goals that could negatively affect their physical health, emotional health, or financial well-being (e.g., smoking cigarettes; drinking and driving).

MODERATING FACTORS

Earlier in this entry, it was noted that emotionality and impulsivity could have negative effects on decision-making. Some scholars argue that these tendencies should be considered moderating factors because people tend to make better decisions when they are calm and reflective than when they are emotionally aroused and impulsive. Generally speaking, these tendencies cause problems whenever they keep decision makers from fully considering the consequences of their actions and discovering better options than those that they implemented. Scholars such as Lawrence Steinberg and Ronald Dahl argue that adolescents are particularly vulnerable to the influences of emotions and impulsivity, and they argue that this vulnerability has a neural basis in their brain. As the brain continues to mature into adulthood, Steinberg and Dahl argue that decision makers gain the capacity to regulate their emotions and impulsivity. As with most other developmental claims regarding decision-making competence, however, the evidence supporting this view is relatively sparse and open to multiple interpretations. More research is needed to substantiate all of the age trends reported here.

INSTRUCTIONAL APPROACHES TO IMPROVE DECISION MAKING

In light of the rise in problem behaviors during adolescence (e.g., cigarette smoking, alcohol use, etc.), university researchers and policy makers in school systems have implemented a number of interventions to improve the quality of decision making in adolescents. Some of these interventions focus on general characteristics of good decision-making (e.g., seek advice from knowledgeable individuals when you do not know what to do) while others specifically target particular problem behaviors (e.g., illicit drug use). Those who implement the general approaches hope that students will apply the principles to all decisions, including those related to problem behaviors. The standard by which a program should be judged to be effective is that it both alters the manner in which decisions are made and reduces the incidence of problem behaviors. Proving the latter requires the application of rigorous methodological approaches such as randomly assigning teens to intervention and control conditions, measuring decision making and problem behaviors before and after the intervention, and using valid measures of decision making and problem behaviors. Reviews of the literature reveal that few studies meet these standards of quality. Many studies show that teens can learn the content of a program, but few show that their actual decision-making changed as a result. The studies that do meet the standards have been found to be more effective when (a) they are more comprehensive (i.e., they target multiple causes of the problem behavior), (b) the teens themselves learn the information in an active rather than passive manner, and (c) the participants engage in peer-to-peer instruction.