430
for describing relationships at multiple levels and scales. theories are ef-
forts to build understanding by making core assumptions about specific
working parts of frequently encountered phenomena and predicting general
outcomes. Models are very specific working examples of a theory – and
they are frequently confused with being theories themselves. as alchian
(1950) pointed out long ago, what is called “rational choice theory” is not
a broad theory of human behavior but rather a useful model to predict
behavior in a particular situation – a highly competitive market for private
goods. Predictions derived from the rational choice model are empirically
supported in open markets for private goods and other competitive environ-
ments (holt 2007; smith and Walker 1993; satz and ferejohn 1994). thus,
it is a useful model to retain for predicting outcomes in competitive settings
related to excludable and divisible outcomes.
While it is not possible yet to point to a single theory of human behavior
that has been successfully formulated and tested in a variety of settings,
scholars are currently positing and testing assumptions that are likely to be at
the core of future developments (smith 2003, 2010). these relate to (1) the
capability of boundedly rational individuals to learn fuller and more reliable
information in repeated situations when reliable feedback is present, (2)
the use of heuristics in making daily decisions, and (3) the preferences that
individuals have related to benefits for self as well as norms and preferences
related to benefits for others (see Poteete, Janssen, and ostrom 2010: chap.
9; e. ostrom 1998).
the assumption that individuals have complete information about all
actions available to them, the likely strategies that others will adopt, and the
probabilities of specific consequences that will result from their own choices,
must be rejected in any but the very simplest of repeated settings. When
boundedly rational individuals do interact over time, it is reasonable to
assume that they learn more accurate information about the actions they can
take and the likely actions of other individuals (selten 1990; simon 1955,
1999). some highly complex common-pool resource environments, however,
approach mathematical chaos (J. Wilson et al., 1994) in which resource users
cannot gain complete information about all likely combinations of future
events.
in many situations, individuals use rules of thumb – heuristics – that they
have learned over time that work relatively well in a particular setting. fishers
end up “fishing for knowledge” (J. Wilson 1990) where using heuristics over
time enables them to recognize diverse clues of environmental processes
that they need to take into account when making their own decisions. When
individuals do interact repeatedly, it is possible to learn heuristics that
approach “best-response” strategies and achieve close to local optima
(Gigerenzer and selten 2001). in eras of rapid change or sudden shocks,
however, heuristics may not enable individuals to achieve high payoffs.
individuals also learn norms – internal valuations that are negative or
positive related to specific actions such as lying or being brave in particular
situations (crawford and ostrom 2005). the strength of an internal commit-
431
ment (sen 1977) may be represented in the size of the internal weight that
an individual assigns to actions and outcomes in a particular setting. among
individual norms are those related to valuing outcomes achieved by others
(cox and deck 2005; cox, sadiraj, and sadiraj 2008; andreoni 1989; Bolton
and ockenfels 2000). fehr and schmidt (1999) propose that individuals
dislike unequal outcomes of interactions and thus have an internal norm
of “inequity aversion.” axelrod (1986) posits that individuals who adopt
meta norms related to whether others follow the norms that have evolved
in a group, increase the probability that norms will be followed. leibbrandt,
Gneezy, and list (2010) show that individuals who regularly work in teams
are more likely to adopt norms and trust each other more than individuals
working alone. frohlich and oppenheimer (1992) posit that many
individuals adopt norms of fairness and justice. not all individuals have the
same norms or perceptions of a situation (ones and Putterman 2007), and
they may differ substantially in whether they consider a way of sharing costs
to be fair (eckel and Grossman 1996).
simply assuming that humans adopt norms, however, is not sufficient to
predict behavior in a social dilemma, especially in very large groups with no
arrangements for communication. even with strong preferences to follow
norms, “observed behavior may vary by context because the perception of
the ‘right thing’ would change” (de oliveira, croson, and eckel 2009: 19).
various aspects of the context in which individuals interact affect how indi-
viduals learn about the situation they are in and about the others with whom
they are interacting. individual differences do make a difference, but the
context of interactions also affects behavior over time (Walker and ostrom
2009). Biologists recognize that an organism’s appearance and behavior are
affected by the environment in which it develops.
for example, some plants produce large, thin leaves (which enhance
photosynthetic photon harvest) in low light, and narrow, thicker leaves
(which conserve water) in high light; certain insects develop wings
only if they live in crowded conditions (and hence are likely to run
out of adequate food in their current location). such environmentally
contingent development is so commonplace that it can be regarded as
a universal property of living things. (Pfennig and ledón-rettig 2009:
268)
social scientists also need to recognize that individual behavior is strongly
affected by the context in which interactions take place rather than being
simply a result of individual differences.
B. The Central Role of Trust in Coping with Dilemmas
even though arrow (1974) long ago pointed to the crucial role of
trust among participants as the most efficient mechanism to enhance
transactional outcomes, collective-action theory has paid more attention to
payoff functions than to how individuals build trust where others are recip-