Take The First: Option-generation and resulting choices

https://doi.org/10.1016/S0749-5978(03)00027-XGet rights and content

Abstract

Experimental decision-making research often uses a task in which participants are presented with alternatives from which they must choose. Although tasks of this type may be useful in determining measures (e.g., preference) related to explicitly stated alternatives, they neglect an important aspect of many real-world decision-making environments—namely, the option-generation process. The goal of the present research is to extend previous literature that fills this void by presenting a model that attempts to describe the link between the use of different strategies and the subsequent option-generation process, as well as the resulting choice characteristics. Specifically, we examine the relationship between strategy use, number and order of generated options, choice quality, and dynamic inconsistency. “Take The First” is presented as a heuristic that operates in ill-defined tasks, based on our model assumptions. An experiment involving a realistic (sports) situation was conducted on suitable participants (athletes) to test the predictions of the model. Initial results support the model’s key predictions: strategies producing fewer generated options result in better and more consistent decisions.

Introduction

How do people choose what to choose from? That is, how do people generate possible solutions to a task when they are not restricted to selecting from among a set of alternatives given to them? Unfortunately, this question has received relatively little attention in the judgment and decision-making literature, compared with the study of people’s choices among given alternatives. Although there exist research streams such as the work of Gettys and colleagues (e.g., Engelmann & Gettys, 1985; Gettys, Mehle, & Fisher, 1986; Gettys, Pliske, Manning, & Casey, 1987) and Klein and colleagues (e.g., Klein & Wolf, 1998; Klein, Wolf, Militello, & Zsambok, 1995), the topic of option generation seems to be underrepresented, when considering its vital importance. Descriptively, examining the option-generation process can help us to better understand human decision behavior and develop more precise models than can be achieved solely through, e.g., process-tracing techniques of choices among given gambles. Practically, we can use this information to assist decision makers in some settings (e.g., business) to be more aware of their “predecisional” behavior, and perhaps we can develop prescriptive tools to help them in systematic analysis. Although contemporary decision-making models make assumptions about how people search through a set of given options—such as by using “normative” optimization methods (e.g., Luce, 2000) or “fast and frugal” heuristics (e.g., Gigerenzer & Todd, 1999)—the question remains of where these options come from, if they are not readily available. For example, Bayesian models require that the hypotheses be precisely formulated, and thus they could not be applied to option generation. Although some research makes the distinction between external search for information in the environment and internal search for information in memory (e.g., Hastie & Pennington, 1995), it is often assumed that the options are there, and one must simply discover a way to get to them. For example, subjective utility theories describe how the attributes for various options are weighted and integrated, without mentioning from where the options under consideration come.

In contrast, the current research will examine the option-generation process—how alternatives are generated “from scratch,” when they are not “out there” in the environment. To do so, the current research differs from the majority of decision-making studies in the use of a divergent-thinking task. That is, we employ a procedure that presents an ill-defined problem to which participants must develop possible solutions and select among them, rather than presenting information such that participants need only to integrate it and choose. Our theoretical model does draw on the topics mentioned above, such as memory retrieval and decision strategies (fast and frugal heuristics), as well as the existing research on generation. A brief review of the previous literature, therefore, will first be presented and related to our approach. This is followed by an introduction to the relevant concepts (memory and search). Then, we will integrate these into “Take The First,” an option-generation and choice heuristic for use in divergent-thinking situations. This will lead to predictions regarding the resultant choice behavior, and the presentation of an experiment to test these predictions. In conclusion, we will assess the validity of the model, relate our model predictions and findings to previous work, and propose directions for future research.

Section snippets

Review of literature on option generation

A great deal of relevant work has been done by Gettys and colleagues, who examined the processes that precede active choices, such as the generation of possible actions, the potential outcomes of these actions, and the assessments of each outcome’s plausibility. They propose many concepts that will be incorporated in our model, in both the generation of acts (Gettys et al., 1987) and hypotheses (Gettys et al., 1986). Gettys et al. (1986) assume that hypotheses are generated by searching memory

Take The First: An option-generation heuristic

It is imperative that we use some framework for making predictions as to how people generate options—what are the possible processes responsible, and how can these be formulated in terms of definable strategies? The model we propose and the resulting heuristic, “Take The First,” utilizes the principles of associative memory networks in conjunction with the rules of fast and frugal heuristics, discussed above.

Effects of option generation on decision making

Hypothesis 1 (H1). There will be differences in the number of options generated, depending on the strategy employed. Our definition of the strategies necessarily implies that different types of options will be generated, depending on the strategy used. However, it seems likely that there will also be different numbers of options generated depending on the strategy. For example, if there are more spatially connected nodes in a sports player’s memory (due to experience, training, perceptual bias,

Discussion

We presented a model of option generation in ill-defined tasks and a resulting heuristic for these situations. Our primary hypothesis concerning strategy use, that different strategies would result in generation of different types of options, was operationalized and confirmed. Differences were also found in the number of options generated for each strategy (H1). Subsequently, these generated options resulted in differences in choice quality, per the “less-is-more” effect (H2): the serial

References (52)

  • E.U. Weber et al.

    And let us not forget memory: The role of memory processes and techniques in the study of judgment and choice

  • M. Amelang et al.

    Differentielle Psychologic and Persönlichkeitsforschung [Differential psychology and personality research]

    (1996)
  • J.R. Anderson et al.

    Knowledge representation

  • J.R. Anderson et al.

    Performance

  • J.R. Anderson et al.

    Reflections of the environment in memory

    Psychological Science

    (1991)
  • R.M. Berger et al.

    Plot titles

    (1969)
  • J.R. Busemeyer et al.

    Dynamic and consequential consistency of choices between paths of decision trees

    Journal of Experimental Psychology: General

    (2000)
  • P.A. Christensen et al.

    Alternate uses

    (1960)
  • P.A. Christensen et al.

    Consequences

    (1958)
  • A.M. Collins et al.

    A spreading activation theory of semantic processing

    Psychological Review

    (1975)
  • T. Connolly

    Action as a fast and frugal heuristic

    Minds and Machines

    (1999)
  • K.A. Ericsson et al.

    Long-term working memory

    Psychological Review

    (1995)
  • C. Facaoaru

    Kreativität in Wissenschaft und Technik [Creativity in science and technics]

    (1985)
  • K. Gerard

    Wing play in team handball

    Team Handball News

    (1978)
  • G. Gigerenzer et al.

    Reasoning the fast and frugal way: Models of bounded rationality

    Psychological Review

    (1996)
  • Gigerenzer, G., Todd, P.M. (1999). Fast and frugal heuristics: The adaptive toolbox. In P. M. Gigerenzer, G. Todd, &...
  • Cited by (0)

    PROBRAL (DAAD, 1997–1999) funded this study. The experiment is part of a series of studies carried out in cooperation with the University of Heidelberg, Germany, and University of Minas Gerais, Brazil. The authors thank Anita Todd for proofreading and corrections, and Terry Connolly, Gerd Gigerenzer, Gary Klein, Torsten Reimer, Lael Schooler, and an anonymous reviewer for helpful suggestions and comments on an earlier version of this paper. In addition we thank our cooperation partners Klaus Roth, Pablo Greco, and Jörg Schorer for help in this project.

    View full text