Part I
Historical Foundations
1
Decision Making by Experts
Influence of Five Key Psychologists
JAMES SHANTEAU AND WARD EDWARDS
Psychological studies involving experts date back to the earliest days of experimental psychology. Research on domain experts has also been a fundamental part of the history of judgment and decision making (JDM). The purpose of this chapter is to look at how domain experts have been viewed in the psychological literature. The focus will be on an unappreciated historical bias derived from a misinterpretation of the foundations of experimental psychology.
This chapter will first focus on the contributions of five historically significant psychologists: Wilhelm Wundt, James McKeen Cattell, Edward Titchener, Edwin Boring, and Ward Edwards. The discussion will turn on the impact of an overlooked assumptionâthe assumption of the generalized normal adult human mind (abbreviated as GNAHM)âthat arises from this history. The chapter concludes with an analysis, drawn from the ideas of Wundt and Edwards, about where research on experts should go in the future.
Much research done today investigates judgment-and-decision-making phenomena involving normal adult minds. This emphasis on normality reflects some unrecognized assumptions about the human mind, namely, that methods and theories are assumed to be generalizable to most people. However, the study of outliers, such as experts, has revealed some of the most intriguing aspects of human behavior. In this chapter, we review the work of five foundational thinkers whose research on expertise provides insight into the source of pivotal assumptions that guide contemporary research in neuroeconomics, judgment, and decision making.
James McKeen Cattell
The first study of the psychological processes of domain experts appears to have been conducted in 1886 by James McKeen Cattell as part of his doctoral degree under Wilhelm Wundt, the founder of experimental psychology. Cattell investigated individual differences using âassociation timesâ (reaction times) for two skilled observers with different domain expertise. One (B) was âa teacher of mathematicsâ and the other (C) âbusied himself more with literatureâ (Cattell, 1887, p. 31); thus they had expertise in contrasting domains. Cattell asked both observers to answer common questions, some oriented toward math, some oriented toward literature, and others oriented toward unrelated fields. The questions were designed to âcall to mind facts with which (both observers) were familiarâ (Cattell, 1887, p. 71).
The results revealed that mean reaction times over all questions were similar for the two observers: .42 seconds for B and .44 seconds for C. However, there was an interaction between question type and domain of expertise. The results from Cat-tellâs study show a 2 Ă 2 crossoverâexpertise in a domain leads to faster response times for questions in that domain. As Cattell put it, âC knows quite as well as B that 5 + 7 = 12, yet he needs 1/10 sec longer to call it to mind. B knows quite as well as C that Dante was a poet, but needs 1/10 sec longer to think of it.â In other words, expertise has an impact on the rate of thinking above and beyond simply getting the right answer.
In another research phase, Cattell assessed âthe time it takes to form a judgment or opinionâ (1887, p. 69), e.g., deciding which of two eminent men was thought to be greater. The results again showed a 2 Ă 2 crossover. Expertise in a domain leads to faster judgments or choices. âIn judging as to the relative greatness of eminent men, the times were shortest ⌠if the (observer) had already compared the men togetherâ (Cattell, 1887, p. 72). Cattell (1887, p. 73) saw this research as just the beginning and concludes: âI think it ⌠desirable to still further increase the number of experiments.â
Despite this promising beginning, there was little effort to follow up this line of research. Important questions about the effects of expertise on the speed of decision making remain unexplored. Other aspects of decision making by experts also received little attention. Moreover, few theories were advanced and few unique methodologies were developed to study the behavior of experts. One purpose of this chapter is to explore a previously unappreciated reason for the paucity of research. To understand this reason, however, it is first necessary to look at the efforts of three major historical figures in the development of experimental psychology.1
Wilhelm Wundt
There is little dispute that Wilhelm Wundt is the founder of experimental psychology. âWundt is the founder because ⌠he brought the empirical methods of physiology to the questions of philosophy and also created a new, identifiable roleâthat of psychologistâ (Leahey, 1987, p. 182). Wundt made his vision clear in his seminal book, Principles of Physiological Psychology (1873), noting that it was his intention to establish psychology as a new domain of science.2
Wundt argued that the goal for psychology should be âthe construction of explanations of experience and the development of techniques for objectifying experienceâ (Blumenthal, 1975, p. 1081). Experiences to be analyzed ranged from memory and perception to social psychology and language. In his research, Wundt advocated the use of âinternal perception.â Although this methodology has often been labelled introspection, âWundt did not use the kind of qualitative introspection in which subjects describe in detail their inner experiencesâŚ. The majority of Wundtâs studies involved objective measurements that used a variety of sophisticated laboratory equipmentâŚ. Wundt then inferred information about conscious processes and elements from these objective measuresâ (Schultz & Schultz, 1987, pp. 65â66).
The âsubjectsâ in Wundtâs laboratory were far from the naive students commonly used now in research. Instead, highly trained observers (frequently Wundt himself or his graduate students) âwere presented with carefully controlled, sensory events and required to describe their mental experiencesâ (Schultz & Schultz, 1987, p. 65). To yield valid observations, Wundt insisted that certain rules be enforced: the observer had to be âmaster of the situation ⌠all observations were to be repeated many times; and finally experimental conditions were to be varied systematicallyâ (Hothersall, 1984, p. 88). As Wundt argued, âwe learn little about our minds from casual, haphazard self-observation ⌠It is essential that observations be made by trained observers under carefully specified conditions for the purpose of answering well-defined questionâ (Wundt, 1904, p. 7). Wundt dismissed self-observation as âcontemplative meditation that led only to fruitless debates and the grossest self-deceptionsâ (1904, p. 7). Therefore, psychological insights came from skilled observers, not naive subjects. Indeed, Wundt would likely disagree with the reliance on student subjects in research today.3
Wundt chose the term voluntarismus (voluntarism) for his school of psychology. To understand causality, Wundt used the concepts of purpose, value, and future anticipation. His âcentral mechanism of psychological causality was apperception, which in modern terms translates roughly to selective attentionâ (Blumenthal, 1980, p. 30). Attention was more than just selective: âthe mind is a creative, dynamic, volitional force ⌠It must be understood through an analysis of its activityâits processesâ (Hothersall, 1984, p. 89). Moreover, âthe act of choice constituted ⌠a special case of âvolitional activity.â ⌠This is the basis on which the later forms of volition, i.e. choice and decision, developâ (Danzinger, 1980b, p. 96). In short, Wundt developed a dynamic process theory of choice based on the concept of will operating selectively on goals, values, and future prospects.
Wundtâs view of psychology is surprisingly modern. As Fancher (1979, p. 148) concluded, current psychologists will âfind that their studies of such cognitive phenomena as information processing, selective inattention, and perceptual masking are much in the tradition of Wundtâs studies.â According to Wundt (1892, p. 495), âthe fundamental character of mental life ⌠in all its phases it is process; an active, not passive, existence.â Thus, it can be argued that Wundt would have been comfortable with the focus of much of modern cognitive research.4
The numbers associated with Wundt are truly phenomenal: he taught over 24,000 students, supervised 186 doctoral students, including 16 from North America. Many of the early psychology laboratories in the USA were founded by Wundtâs students. âWundt so influenced the first generation of American psychologists that the majority of American students of psychology [today] can probably trace their historical lineage back to himâ (Boring & Boring, 1948). Ward Edwards received his degree under Boring, who was a student of Titchener, who studied with Wundt. In other words, with Edwards you could shake the hand that shook the hand that shook the hand that shook the hand of Wundt.
Edward Titchener
Wundtâs most vocal advocate in the USA was Edward Titchener. After spending two years with Wundt, Titchener considered himself to be a true Wundtianâhe âaccepted Wundtâs psychology without reservationâ (Hothersall, 1984 p. 103). âTitchener pronounced himself to be Wundtâs loyal follower and true interpreter. He proclaimed Wundt as the source of his psychology and the precursor who validated his credentialsâ (Anderson, 1980, p. 95). American psychologists seldom felt the need to read Wundt in German because all they needed to know was provided by Titchener. (Of course, few Americans can read academic German.) Furthermore, what little was available in English often contained serious translation errors. It is noteworthy that âmany more of Wundtâs books were translated into Russianâeven into Spanishâthan were ever translated into Englishâ (Blumenthal, 1980, p. 28).
It was believed that Titchenerâs approach to psychology was essentially a mirror image of that of Wundtâif you knew Titchener, you knew Wundt. However,
recent research on Wundtâs writings casts doubt on this conclusion ⌠Evidence suggests that [Titchener] altered Wundtâs positions to make them appear compatible with his own, to lend credibility to his own views by asserting that they were consistent with those of psychologyâs founder. Titchener ⌠elected to translate only those portions of Wundtâs publications that supported his own approach to psychologyâŚ. For 100 years, texts in the history of psychology, and teachers of the history courses, have been compounding and reinforcing the error under the imprimatur of [Titchenerâs] alleged expertise.
(Schultz & Schultz, 1987, pp. 58â59)
Titchener named his system of psychology structuralism because of his emphasis on âdiscovering the elemental structure of consciousnessâ (Benjamin, 1988, p. 209). âAlthough Wundt recognized elements or contents of consciousness, his overriding concern was their organization into higher-level cognitive processes through the principle of apperceptionâ (Schultz & Schultz, 1987, p. 85). While Wundt emphasized the whole, Titchener focused on the partsâa view he adapted from James Mill (Danziger, 1980a). Wundt never described his psychology as structuralism. Indeed, after Wundtâs death, his students named his movement Ganzheit Psychology or âholistic psychology.â
Titchener considered subjects âto be a recording instrument, objectively noting the characteristics of the object being observed. Human subjects were nothing more than impartial and detached machinesâ (Schultz & Schultz, 1987, p. 91). Titchener talked about mechanized observation: âthe method of psychology is the same as the method of physics ⌠strictly impartial and unprejudicedâ (Titchener, 1910, p. 23). He often referred to his subjects as reagents: âa reagent is generally a passive agent, one that is applied to something to elicit certain reactionsâ (Schultz & Schultz, 1987, p. 91).
In contrast, Wundtâs subjects played a more active role. One of Wundtâs rules was that highly trained observers had to be âmaster of the situationâ (Schultz & Schultz, 1987, p. 91). To yield valid introspections, Wundtâs subjects controlled the situation, whereas Titchenerâs subjects were passive observers.5
Titchenerâs (1923) view of the appropriate content of behavioral research was narrow. He was interested in âpure psychologyâ with no room for comparative (animal) psychology, mental testing, educational psychology, industrial psychology, or social psychology. These fields were considered impure because subjects in these settings could not use introspection. When discussing why he opposed animal research, Titchener (1916, p. 267) argued, âif animals thought, they could undoubtedly use their voca...