Benjamin Franklin
remarked in a letter in 1789, "in this world nothing is certain but death and
taxes." The Center for Adaptive Behavior and Cognition (ABC) investigates reasoning and
decision making under uncertainty, at the levels of both individuals and social groups. The topic
is interdisciplinary, and this feature is mirrored in the composition of the research group. The
group consists of researchers with degrees in psychology, mathematics, computer science,
evolutionary biology, anthropology, economics, and other fields. The working modus of the
Center is to have researchers with different methodological abilities - such as experimental
methods, computer simulation, and mathematical analysis - cooperate on the same problem.
The spatial, organizational, and social set- ups of the Center help to facilitate and organize
active interdisciplinary collaboration.
The ABC program
combines a strong theoretical focus with diverse practical applications. The
theoretical focus can be (artificially) divided into three aspects.
1. Bounded rationality.
Models of bounded rationality try to answer the question of how real
people with limited time, knowledge, money, and other scarce resources make decisions. This
program is an alternative to the dominant optimization paradigm in cognitive science,
economics, and behavioral biology that poses the question of how Laplacean superintelligences
and other heavenly beings would behave. We study the proximal mechanisms of bounded
rationality, that is, the adaptive heuristics that enable decisions under uncertainty in a fast and
frugal way. The collection of heuristics and their building blocks is what we call the adaptive
toolbox.
2. Ecological
rationality. Models of ecological rationality describe the structure and
representation of information in environments and their match with mental strategies, such as
boundedly rational heuristics. To the degree such a match exists, heuristics need not trade
accuracy for being fast and frugal - they can have it both ways. The simultaneous focus on the
mind and its environment, past and present, puts research on decision making under uncertainty
into an evolutionary and ecological framework, a framework that is missing in most theories of
reasoning, both descriptive and normative. In short, we study the adaptation of mental and
social strategies to real-world environments rather than compare strategies to the laws of logic
and probability theory.
3. Social rationality.
Social rationality is a variant of ecological rationality - one in which the
environment is social rather than physical or technical. Models of social rationality describe the
structure of social environments and their match with boundedly rational strategies people use.
Social environments increase the variety of goals in decision making and those of the heuristics.
In addition to the goals that define ecological rationality - to make fast, frugal, and fairly
accurate decisions - social rationality is concerned with more specific goals, such as choosing
an option that one can defend by argument or morally justify, or that can create consensus.
Unlike the purely cognitive focus of most research on bounded rationality, socially adaptive
heuristics include emotions and social norms that can act as heuristic principles for decision
making.
These three notions
of rationality converge on the same central issue: to understand human
behavior and cognition as it is adapted to specific environments (ecological and social), and to
discover the heuristics that guide adaptive behavior. They diverge from the view of rationality as
coherence (e.g., internal consistency), as optimization or maximization (e.g., rational choice
theory), and as a domain-general calculus (e.g., Leibniz's dream of a universal characteristic).
This theoretical program functions
as a metatheory, and the research group works on the
metatheory, specific models, and applications. Given the interdisciplinary background of the
research group, the applications range from teaching statistical thinking to using heuristics to
invest in the stock market. For instance, in our studies on ecological rationality, we worked out
a particular representation of numerical information that improves statistical reasoning
(precisely, Bayesian inference); subsequently, we applied this result to help physicians, AIDS
counselors, and expert witnesses in law courts to improve their statistical intuitions. The
concept of ecological rationality is particularly apt for practical applications, because it can
provide models for representing information in the environment, enabling people to "see
through" a problem without elaborate training in statistics or decision theory.
A body of results
concerning fast and frugal heuristics has recently appeared in the book
"Simple heuristics that make us smart" (G. Gigerenzer, P. M. Todd, & the ABC Research
Group, Oxford University Press, 1999) collectively written by the group. The individual
chapters in this book are all multi-authored, reflecting the interdisciplinary collaboration among
the researchers. This work exemplifies the fruitful interaction among experimenters (who test
whether a heuristic is actually used by people and when), computer scientists (who simulate
how effective a heuristic is in real-world environments), and mathematicians (who prove which
environmental structures simple heuristics can exploit in order to be as accurate as complex
"rational" strategies are). Most of the new results in this book would not have been obtained
without the simultaneous input of (and a healthy competition between) different methodological
approaches.
The Dahlem Workshop
"Bounded Rationality: The Adaptive Toolbox" organized by Gerd
Gigerenzer and Reinhard Selten (Berlin, March 1319, 1999) brought the ABC researchers
together with colleagues from economics, artificial intelligence, anthropology, and animal
behavior. The goal was to develop an understanding of bounded rationality on the basis of
psychologically plausible abilities - cognitive, emotional, and social - rather than on the fiction
of
optimization. The results of this Workshop have recently appeared in G. Gigerenzer and R.
Selten: Bounded
rationality: The adaptive toolbox <dahlem_workshop.htm>(MIT Press,
Cambridge/MA).
In
developing new views of adaptive behavior and cognition, researchers in evolutionary
psychology can suggest alternative views to tool-laden theories that tend to picture the mind as
a statistician or econometrician equipped with the latest software. For instance, the notion of
adaptation reminds us to study the structure of environments, past and present, in order to
generate hypotheses about the structure of the mind. Furthermore, important adaptive
problems, such as avoiding predators and finding prey, also remind us that for these goals
neither the classical virtue of rationality - logical consistency - nor optimal accuracy in inference
is necessary or sufficient. In many situations the important thing is to react quickly, and to do so
on the basis of limited information.
Finally,
research on methodological, historical, and theoretical questions, in particular the
influence of methodological preferences - such as linear models - on theories of cognition,
constitutes a source of ideas that is of central importance for modeling visions of rationality. For
instance, by historical accident, the standard statistical methods institutionalized in psychology
do not include methods for search and stopping rules (the key processes of bounded
rationality); thus, the standard cognitive theories tend not to include search and stopping rules
either. By the tools-to-theories heuristic, ingredients of tools often turn into ingredients of
theories and, thus, what is missing in a tool may turn into a blind spot in theories. An historical
and contemporary analysis of the role of the "expert" was the topic of the Schloeßmann
Seminar "The Expert in Modern Societies: Historical and Contemporary Perspectives" (Berlin,
November 1998). The Seminar was co-sponsored by the Max Planck Institute (MPI) for
Human Development (Berlin), the MPI for Foreign and International Criminal Law (Freiburg
i.Br.), the MPI for the History of Science (Berlin), the MPI for the Study of Societies
(Cologne), the MPI for History (Göttingen), the MPI for Foreign and International Patent,
Copyright, and Competition Law (Munich), the MPI for Research into Economic Systems
(Jena), and the MPI for Demographic Research (Rostock). The proceedings will be published
in a book.
The
ABC program is an invitation to participate in a journey into largely unknown territory. The
journey ventures into a land of rationality that is different from the familiar one we know from
many stories in cognitive science and economics - tales in which humans live in a world with
unlimited time and knowledge, where the sun of enlightenment shines down in beams of logic
and probability. The new land of rationality we set out to explore is, in contrast, shrouded in a
mist of dim uncertainty. People in this world have only limited time, knowledge, and
computational capacities with which to make inferences about what happens in their world. The
notions of bounded, ecological, and social rationality are our guides to understanding how
humble humans survive without following the heavenly rules of rational choice theory.
1. Frames that Work?
