Goals and Plans in Decision Making

Abstract

We propose a constructed-choice model for general decision making. The model departs from utility theory and prospect theory in its treatment of multiple goals and it suggests several different ways in which context can affect choice.
It is particularly instructive to apply this model to protective decisions, which are often puzzling. Among other anomalies, people insure against non-catastrophic events, underinsure against catastrophic risks, and allow extraneous factors to influence insurance purchases and other protective decisions. Neither expected-utility theory nor prospect theory can explain these anomalies satisfactorily. To apply this model to the above anomalies, we consider many different insurance-related goals, organized in a taxonomy, and we consider the effects of context on goals, resources, plans and decision rules.
The paper concludes by suggesting some prescriptions for improving individual decision making with respect to protective measures.

Full text

Judgment and Decision Making, Vol. 2, No. 3, June 2007, pp. 137–168
Goals and plans in decision making
David H. Krantz∗
Department of Psychology
Columbia University
Howard C. Kunreuther
The Wharton School
University of Pennsylvania
Abstract
We propose a constructed-choice model for general decision making. The model departs from utility theory and
prospect theory in its treatment of multiple goals and it suggests several different ways in which context can affect
choice.
It is particularly instructive to apply this model to protective decisions, which are often puzzling. Among other
anomalies, people insure against non-catastrophic events, underinsure against catastrophic risks, and allow extraneous
factors to influence insurance purchases and other protective decisions. Neither expected-utility theory nor prospect
theory can explain these anomalies satisfactorily. To apply this model to the above anomalies, we consider many different
insurance-related goals, organized in a taxonomy, and we consider the effects of context on goals, resources, plans and
decision rules.
The paper concludes by suggesting some prescriptions for improving individual decision making with respect to
protective measures.
Keywords: goals, plans, decision making, catastrophic risk, insurance, utility theory, prospect theory, protective behavior
1 Introduction
A concept of goal (or aim, motive, purpose, etc.) has
been long been taken for granted in many accounts of
human behavior. Behavior is attributed to aims or goals
in everyday discourse (“he wore his best suit in order
to make a good first impression”), in literature (epics,
novels, etc.), and in scientific discourse, for example, in
the widely differing accounts in Aristotle’s Ethics (circa
350 B.C.E.), William James’s Principles of Psychology
(1890/1950), and Kurt Lewin’s Dynamic Theory of Per-
sonality (1935). Despite the obviousness and wide use of
goal concepts, the dominant tradition in economics and
the decision sciences has emphasized utility or value as a
basis for choice. Particular goals are viewed as ways of
increasing utility (or happiness). Utility can be thought
of as a sort of common currency that integrates multi-
ple goals or multiple quantitative attributes of outcomes.
From this standpoint, tradeoffs among goals are attempts
to maximize utility.
∗We thank Xin Piao for her help in many phases of our thinking
about protective decisions, Jonathan Baron, Daniel Kahneman, Ralph
Keeney, Alex Leeds and Jay Russo for insightful comments on earlier
manuscript drafts, and Mark Pauly for much advice relating to insurance
anomalies. Financial support from NSF Grant # SES-0136872 and NSF
Cooperative Agreement # SES-0345840, to Columbia University and
from NSF Grant # CMS-0527598 to the Wharton Risk Management
and Decision Processes Center, University of Pennsylvania is gratefully
acknowledged.
The view that all goals contribute to a single common
currency was clearly enunciated by Plato, in The Sympo-
sium. He used the metaphor of weighing different plea-
sures and pains in a balance.
“And do you, like a skillful weigher, put into
the balance the pleasures and the pains, and
their nearness and distance, and weigh them,
and then say which outweighs the other... . ”
(Jowett translation)
This view is central to Plato’s thought, underlying his
theories of education and government. Studies in math-
ematics, science, and metaphysics are needed to educate
the “skillful weigher”, who must integrate across differ-
ent goals and across near and distant times. Similar views
dominated utilitarian thought in the 17th to 19th cen-
turies, and included integration of value across individu-
als in a society as well as different goals and times. (e.g.,
Bentham, 1789).
Aristotle’s Ethics, by contrast, partially disagreed, em-
phasizing multiple goods, and stating that the way in
which different goals fit together should vary with the
occasion. Aristotle can perhaps be read as advocating
situation-dependent integration of multiple goals, an idea
that we pursue and elaborate in this paper.
Despite this hint from Aristotle, Plato’s concept of a
single common currency that serves to integrate value
across myriad goals has largely held sway both in gen-
137

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 138
eral psychology and in decision science. Freud’s concept
of libido (1920), Beebe-Center’s hedonic tone (1932),
Hull’s concept of generalized drive (1951), work on re-
ward systems in the brain (Olds & Milner, 1954; Wise,
2004), and Diener’s and Seligman’s concepts of gen-
eral happiness (e.g., Diener & Seligman, 2002) all sug-
gest some general quality that is linked to many different
goals. An exception is Keeney (1992), who advocates
that decision analysis focus on separate goals and values
as a starting point, rather than on goal tradeoffs as repre-
sented by overall utility.
In decision science, the concept of maximization is
linked closely to a mapping onto a single dimension of
utility. A bounded set of real numbers has a limiting
maximum; but there is no natural total ordering of sets
of vectors in two or more dimensions, and therefore no
natural concept of maximization. In fact, total ordering
is fundamental to most foundational theories in decision
science (Savage, 1954; Krantz, Luce, Suppes & Tversky,
1971; Kahneman & Tversky, 1979; Tversky & Kahne-
man, 1992).
The idea that all human goods can be weighed in the
same balance is a fascinating scientific hypothesis that
has been worth pursuing, to determine the extent of its
applicability and the ways in which it fails. Translat-
ing many goods into a one-dimensional currency fits well
with human thought processes, especially analog mental
models (Attneave, 1974; Egan & Grimes-Farrow, 1982).
Unidimensionality opens the way to the application of
powerful mathematical methods for computing or for ap-
proximating maxima (Gregory & Lin, 1992; Nocedal &
Wright, 2006).
1.1 An alternative general theory
In this paper, we pursue a more Aristotelian theory of de-
cision making, where preferences are constructed based
on the decision context (Slovic, 1995), and a decision
maker focuses on goals, rather than on maximizing hap-
piness or utility. We attempt to show that this alternative
approach leads to new explanations of how people make
choices and raises novel questions with respect to de-
scriptive theories of behavior and prescriptive guidelines
for aiding the decision maker and improving choices.
As our title implies, the theory we present is about
goals and plans. Many plans — going out to a movie,
embarking on a shopping trip, purchasing insurance, or
the invasion of Iraq by the United States — are or were
selected with a view to achieving several different goals
at once. A shopping trip is paradigmatic, because it of-
ten involves several discrete stops, each with one or more
goals; but an evening at a movie may simultaneously sat-
isfy the goals of companionship, emotional and visual
stimulation, and keeping up with current culture. Insur-
ance plans are often aimed both at financial goals and at
emotional goals (“peace of mind”). We will discuss mul-
tiple goals of insurance and other protective plans in de-
tail below.
In accord with Miller, Galanter & Pribram (1960), we
regard the plan as a fundamental structural unit in deci-
sion theory. We do not consider each tiny muscle move-
ment or each phoneme in an utterance to be a choice.
Rather, decision theory offers explanations only at the
behavioral level where someone (consciously or uncon-
sciously) considers what goals will be accomplished by
various possible plans, or what plan can be designed that
will be likely to achieve several important goals.
Past research on plans can be unearthed from various
areas of cognitive science. The area of motor control
is particularly rich in evidence for (unconscious) higher-
level programs governing sequences of skilled move-
ments (Rosenbaum & Krist, 1996). Interest in movement
planning is very explicit in robotics (Patnaik & Karibas-
appa, 2005). Planning has sporadically been considered
in other theories of human and/or artificial intelligence.
One of the major virtues of considering goals and plans
together is that individual decision making is brought into
close analogy with organizational decision making. In the
latter, goals and plans are shared among the individuals or
groups that select and implement plans. For a new rail-
road bridge, or for an advertising campaign, the goals, the
plans, and their perceived relationship are made explicit
and often recorded. For an individual decision, goals,
plans, and their perceived relationship are hidden within
that person’s conscious and unconscious cognitions and
emotions, but one can try nonetheless to gather data that
provide some information about these processes.
This stance also requires us to distinguish between
goals and resources. Plans draw on resources to achieve
goals. Some resources, such as money or favors owed,
can be accumulated; sometimes, accumulation of a re-
source may itself become a goal. We nonetheless con-
tinue to distinguish, for example, between money consid-
ered as a resource and the same money considered as a
goal.
1.2 Protective decisions and insurance
A main concern of our paper is protective decision mak-
ing. In much of it, the focus is even narrower: we con-
sider insurance decisions at length. The theory we of-
fer is much more broadly applicable. However, we have
found the narrower focus to be useful for theory devel-
opment. There are many apparent anomalies in insur-
ance decisions by individuals and households. Classical
utility theory is supposed to account for decision mak-
ing under risk, and especially for insurance purchase. It
was not obvious to us initially that a goal/plan approach

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 139
would do better than utility theory, or its popular alterna-
tive, prospect theory. Applying a multiple-goal theory in
this domain has been both challenging and enlightening,
compared to the more obvious examples such as shopping
trips and movie excursions.
In discussing insurance choices, we start from the idea
that insurance is designed principally as a device to share
financial risk, for situations where many are at risk, but
relatively few actually suffer a financial loss. In many
cases, it is easy to share financial risk, but difficult, if not
impossible, to share other risks. Having one’s home burn
down, for example, is extremely stressful, even if nobody
is hurt. The stress may persist for a considerable period
of time. Stress-induced suffering cannot readily be shared
by others not directly affected by the fire, although social
practices exist that can ease it. What can most easily be
shared is the financial risk. Each household at risk pays
a relatively small amount, the insurance premium, to par-
ticipate in this risk sharing. When a participant’s home
does burn down, the insurer pays the contracted amount
to cover some of the cost of rebuilding or purchasing a
replacement home. Instead of a few cases where victims
suffer catastrophic financial loss, everyone at risk bears a
much smaller financial payment.
Insurance premiums must be large enough in aggre-
gate to cover not only the total insured financial losses,
but also the costs to the insurer of the risk-sharing effort.1
Thus, premiums (unless subsidized) usually exceed av-
erage losses. In common-sense terms, and in standard
economic theory, people are willing to pay more than the
expected loss because they are risk-averse. Paying the
premium does not lower overall utility or happiness level
significantly, but suffering a large financial loss would re-
sult in a very large drop in utility or happiness. This ratio
of utility changes is much greater than the inverse prob-
ability of incurring the loss, thus suggesting that individ-
uals purchase insurance so as to maximize their expected
utility.2
From a financial standpoint, anomalies arise when con-
sumer choices with respect to insurance sometimes ap-
pear to be suboptimal given reasonable degrees of risk
aversion. For example, it is common for people to pay
added premiums that seem excessive to obtain automo-
bile collision insurance with a low rather than high de-
ductible that requires them to pay for the initial portion
1Operations include estimation of loss probabilities (on which pre-
miums depend), suitable investment of the pooled premiums, seeking
out and planning new opportunities for risk sharing that will attract in-
surance participants, direct dealing with potential participants to clarify
contract options and conclude contracts, ex-ante prevention of excessive
losses (e.g., by enforcing prudent protective measures), and ex-post ver-
ification of losses. All these require judgment and skill and the people
engaged in them are often well paid.
2Prospect theory offers a different view: people pay more than the
expected loss because the decision weight for a small loss probability is
disproportionately large. We consider this view in Section 5.
of the loss. To make matters worse, some decide not to
file a claim following a small accident whose cost could
largely be reimbursed via this low deductible3: they fear
that a claim would lead to increased premiums in the next
and succeeding years and/or they prefer to avoid incur-
ring the transaction costs involved in settling the claim
with their insurer. A higher deductible would have saved
these individuals money before the accident and avoided
the costs associated with deciding whether or not to file a
claim.
A rather different example emerged after the passage
of the National Flood Insurance Program (NFIP) in 1968.
Insurance coverage against water damage from flooding
was offered both to homeowners and to commercial en-
terprises in high hazard areas at subsidized low rates. Yet
there was limited interest in purchasing this coverage de-
spite the subsidy and despite the potential for catastrophic
losses (Kunreuther, 1978).
As a third example, many people are prepared to
pay considerably more to insure possessions that they
find very attractive than to insure possessions toward
which they feel neutral or negative (Hsee and Kunreuther,
2000). If such a possession is needed, then it must be
replaced after loss or damage. The replacement cost re-
mains the same, independent of one’s positive or negative
feelings toward the object. Since the insurance offers ex-
actly the claims payment, it is hard to justify financially
paying more for insuring the attractive than the less at-
tractive object.
The remainder of the paper is organized in the follow-
ing manner. We next (Section 2) discuss the types of in-
surance anomaly illustrated by the preceding three exam-
ples, and suggest psychological processes related to goals
that might account for each type. Section 3 presents the
elements of a quantitative theory of constructed choice,
based on goals, plans, and decision weights, and con-
trasts this theory with the standard theory of expected
utility maximization. Section 4 presents a taxonomy of
insurance-related goals. Section 5 draws on this taxon-
omy to explain the insurance anomalies in terms of our
constructed-choice theory. This section also probes the
failings of utility theory and prospect theory with respect
to characterizing insurance decisions. The paper con-
cludes (Section 6) with a set of prescriptive implications
concerning protective activities, based on our theory of
decision making.
3The term “pseudodeductible” was introduced by Braun, Fader,
Bradlow and Kunreuther, 2006, referring to a low deductible that is paid
for but not used when it could be.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 140
2 Types of anomaly
In each example presented above, people make choices
that seem anomalous, in the sense that they cannot be
explained on the basis of financial calculations with rea-
sonable risk aversion. One broad class of anomalous be-
havior is insuring against a non-catastrophic loss. A sec-
ond type is underinsuring against a truly catastrophic loss.
The third category is considering factors that have little or
nothing to do with magnitudes or probabilities of finan-
cial loss when making insurance-purchase decisions.
Insuring against non-catastrophic losses. Many in-
surance contracts have a deductible whereby only losses
in excess of that amount will be reimbursed. Thus, with
a $200 deductible, a $900 loss results in a reimbursement
of $700, a $300 loss yields only $100, and a $100 loss
will not be reimbursed at all. Usually, the additional pre-
mium for a low deductible is set high enough so that the
insurer has a positive expected value. On the average,
therefore, individuals lose money by purchasing insur-
ance with low rather than high deductibles. Nonethe-
less, low deductibles are popular, and a common strat-
egy is to purchase the lowest possible deductible (Kun-
reuther & Pauly, 2006). A number of years ago the Insur-
ance Commissioner of Pennsylvania, Herbert Denenberg,
mandated at least a $100 deductible (rather than a $50 de-
ductible) for automobile collision policies. Although the
plan purportedly saved consumers millions of dollars it
was opposed by the public and had to be rescinded (Cum-
mins, et al., 1974).
Apart from seeking low deductibles, people insure
against other non-catastrophic losses. A good example
is insuring mailed packages with only modest monetary
value. For people with considerable financial means,
loss of personal property (e.g., jewelry, expensive stereo
equipment) is often non-catastrophic, yet many purchase
floater insurance to cover such items. If one insures
against many non-catastrophic losses, one is nearly cer-
tain to come out behind financially, as compared with a
strategy of not insuring against any of those losses. This
follows from the law of large numbers, together with
fact that the insurance contracts have negative expected
value. As a purely financial norm, people ought to self-
insure against all manageable losses; we discuss below
why some do not.
Underinsuring against truly catastrophic losses.
Many people fail to purchase protection against low-
probability high-impact events unless they are required
to do so. Examples are the financial catastrophes that can
arise from a destructive earthquake or flood, from a pro-
longed major illness, or from a large adverse civil judg-
ment. For such events, the pool of individuals at risk is
often large. Therefore, although the probability facing
an individual is low for each of these events, some people
will inevitably be victims. For those who are affected, the
financial protection from insurance can make the differ-
ence between recovery of one’s life pattern versus very
deep and continuing difficulties. If the risk pool of in-
dividuals facing a potentially catastrophic event is large,
the financial risk can be spread widely. The cost of in-
surance for each person can be relatively low compared
to the loss should such an untoward event occur. We will
discuss below some of the reasons why individuals fail to
budget moderate amounts to protect themselves against
such financial catastrophe through insurance and why this
behavior seems imprudent.
Sensitivity to “extraneous” factors. We group to-
gether here behavioral phenomena in which insurance
purchase is influenced by factors that are irrelevant or ex-
traneous, in the sense that they affect neither the financial
costs and benefits of the insurance nor the probability of
an adverse event covered by it. Factors that are often ex-
traneous, in this sense, include: (i) the positive or nega-
tive affect attached to an object or event; (ii) recent expe-
rience of events such as flooding; (iii) what friends and
relatives have decided to do; and (iv) minor transaction
costs.
2.1 Psychological explanations for anoma-
lies
To illustrate our general approach, we introduce two
types of explanation at this point4: extra-financial goals
and context effects. First, individuals consider multiple
goals, not just financial ones in making insurance deci-
sions. As a result, people sometimes purchase insurance
that appears to be unattractive from a financial point of
view but achieves other goals. Second, the particular con-
text may increase or decrease the salience of some goals.
A person who puts heavy weight on salient goals may
decide to purchase insurance that is overpriced or not re-
ally needed, or, on the other hand, may neglect to adopt
protective measures that are attractive financially.
Consider the example of flight insurance, which typi-
cally costs $5 to $10 per $100,000 of coverage. A general
accidental death insurance policy that offers $500,000 for
death in any common-carrier accident (including com-
mercial airplane flights), plus many other benefits (injury,
automobile accident, etc.) can be obtained for about $12
per month for one person. Thus, coverage for $500,000
for a single airplane flight is much cheaper if one gets the
general insurance policy for a full month, rather than in-
surance for a specific flight. However, purchasing flight
4A more complete and systematic account is deferred to Section 5.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 141
insurance at the airport may, for some people, provide
“peace of mind” and for the purpose of such anxiety re-
duction may be preferred to drinking alcohol at an air-
port bar. A person might also purchase such coverage
to demonstrate that she cares for her loved ones. These
extra-financial goals might make flight insurance worth
the cost. Note, however, that flight insurance usually is
purchased because these extra-financial goals are espe-
cially salient in the airport context. Such coverage would
probably be much less popular if policies were sold at
grocery stores.
The presence of multiple goals and the influence of
context are important for prescriptive analyses and de-
cision aids as well as for understanding decision anoma-
lies. When important and stable extra-financial goals mo-
tivate insurance purchase, one might regard the decision
as anomalous from a financial perspective, but quite ra-
tional from a broader perspective of multiple goals. Such
extra-financial goals need to be taken into account in for-
mulating prescriptive aids to decision making. On the
other hand, goals that arise in a particular context might
be quite unstable. Someone who purchases flight insur-
ance might regret doing so after thinking (in a more neu-
tral context) about cheaper general accident coverage. In
this case, the anomalous purchase might be viewed as
a poor decision. From a prescriptive point of view one
might want to develop decision aids or instructional ma-
terials that help people avoid decisions that they will re-
gret in most contexts.
3 Goal-based models of choice
In this section we contrast a standard framework for
choice theory, based on subjective expected (multi-
attribute) utility, with a framework based on goals and
plans, inspired by the theory of constructed choice.
People often construct or select plans designed to
achieve multiple goals. Protective plans are no excep-
tion in this regard. For example, a plan to purchase fire
and theft insurance (on a home, say, or on the contents of
a rented apartment) may be designed to satisfy as many
as seven goals simultaneously: (i) reducing the chances
of a catastrophic loss, (ii) reducing anxiety about risks
of fire and theft, (iii) avoiding regret and/or providing
consolation in case a loss occurs, (iv) satisfying require-
ments stated by a bank or by a landlord, (v) presenting the
appearance of prudence to others who will learn about
the insurance purchase, (vi) maintaining ones relation-
ship with an insurance agent, and (vii) avoiding highly
burdensome insurance premium payments. The impor-
tance of these goals obviously varies with the decision
maker, but may also be affected temporarily by contex-
tual variables. When reflecting on paying monthly bills,
an insurance purchaser may think chiefly about the goals
of satisfying the requirements of the bank that holds the
mortgage loan (goal iv above), and minimizing the cost
of insurance (goal vii). When that same person reflects
on her valuable works of art, she may think chiefly about
reducing anxiety (goal ii) and avoiding regret (goal iv).
There is a subtle and important question about how
such multiple goals should be represented in theories
of human decision making and in prescriptive principles
aimed at better decisions. If goals are viewed as stable,
then the tradeoffs among different goals may well be rep-
resented at least approximately by a multi-attribute util-
ity function. In this framework one may include non-
financial goals in the utility function. If, on the other
hand, context strongly influences what goals are consid-
ered and how they are weighted, as illustrated by the
flight-insurance and fire-insurance examples mentioned
above, then a theory of constructed choice may be more
appropriate than a utility model.
The idea that preferences are constructed, rather than
revealed, emerged from many lines of research in the
late 1980s and early 1990s (e.g., Tversky, Sattath &
Slovic, 1988, Tversky, Slovic & Kahneman, 1990, Chap-
man & Johnson, 1995) and was well characterized by
Slovic (1995). This idea is more or less taken for granted
in current psychological work on decision making (e.g.,
Sedikides, Ariely & Olsen, 1999; Zhang & Markman,
2001). In order to apply the idea systematically to pro-
tective decision making with multiple goals, and in order
to consider its prescriptive implications, we outline and
discuss a theory of context-dependent choice parallel to
the theory of expected utility.
The remainder of Section 3 is structured as follows.
We first present matrix structures (Tables 1 and 2)
that partially represent utility and multiple-goal theories.
These two matrices have similar rows (representing al-
ternative possible strategies or plans) but interchange the
roles played by columns and matrix cells. In utility theory
(Table 1), the matrix columns represent uncertain events
and the cell entries represent outcomes (possibly involv-
ing multiple goals). In multiple-goal theory (Table 2),
columns represent goals, while cell entries represent un-
certainties or distances from achieving goals. Second, we
illustrate some similarities and differences between these
two matrix structures, using a simplified example of a de-
cision about whether to purchase flood insurance (Tables
3 and 4). Third, we discuss some relationships of the
partially formalized multiple-goal theory to constructed
choice. Fourth, we discuss possible decision rules for
plan selection in multiple-goal theory, and indicate how
the problem of quantitative measurement of goal values
and decision weights might be solved in the context of
this theory. Finally, we summarize the main advantages
of the multiple-goal/constructed-choice theory.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 142
3.1 Matrix structures for utility and
multiple-goal theories
In traditional decision theory, a choice among several al-
ternative plans, each with multiple goals, is usually cast
in the framework of multi-attribute utility theory (Debreu,
1960; Krantz, Luce, Suppes & Tversky, 1971; Keeney
& Raiffa, 1976). To incorporate uncertainty, a decision
problem is often represented as a matrix (Savage, 1954):
the rows represent possible actions or strategies for the
decision maker, the columns represent possible events
that could occur, and the entry in any cell of the ma-
trix (any given strategy-event combination) is a multi-
attribute outcome, composed of all goals that will be
achieved if that particular strategy is selected by the deci-
sion maker and that particular event happens to occur.
Table 1 depicts aspects of this model in an abstract ma-
trix form and also indicates how the model is used for
measurement and decision making. The outcome of the
ith strategy, given that the jth event occurs, is denoted
oij (cell entries). This simple notation does not explicitly
show the multiple attributes or multiple goals that com-
prised a single outcome oij. In the model, each such out-
come is assigned a multi-attribute utility uij (Keeney and
Raiffa 1976). Also each uncertain event is assigned a sub-
jective probability pj. Each strategy ithus has an expected
utility given by:
U(strategy i) =
n
X
j=1
pjuij
The central feature of the model is that people do or
should select the strategy that maximizes U. As shown
by Savage (1954), this expected-utility equation serves
as the basis for measurement of outcome utilities uij and
subjective probabilities pj. We refer to this model as sub-
jective expected multi-attribute utility (SEMAUT).
We contrast the matrix shown above with Table 2,
which depicts aspects of the plan/goal model in a similar
abstract form. Here, the ith plan yields a decision weight
wij (cell entries) for the jth goal, Gj. This notation does
not, however, explicitly show the contingencies leading
to each goal. Each Gjis assigned a value vj.
These two structures will be discussed more thor-
oughly below, following an example related to the pur-
chase of flood-insurance (depicted in Tables 3 and 4).
Here, we note a few important differences between the
strategy/event and plan/goal structure at a more abstract
level.
First, the SEMAUT matrix highlights a partition of un-
certain events, leaving multiple goals embedded in the
matrix cells (oij), while the plan/goal matrix highlights
multiple goals, leaving uncertainty embedded in the ma-
trix cells (wij). This difference is heuristic, rather than
Table 1: General strategy/event structure for decision
making under uncertainty. Matrix entries are multiat-
tribute outcomes for strategy/event combinations
Possible events
(mutually exclusive and exhaustive)
Possible strategies E1E2. . . En
strategy 1 o11 o12 . . . o1n
strategy 2 o21 o22 . . . o2n
... ... ... ... ...
strategy m om1om2. . . omn
Events have subjective probabilities pj=prob(Ej)
Outcomes have subjective values uij =value(oij).
uij may be integrated across multiple attributes of oij.
Multi-attribute utility is integrated across uncertain
events:
U(strategy i) =
n
X
j=1
pjuij
mathematical. Each of these representations may have
advantages for some decision settings.
Second, while the rows are analogous, the terminology
used is different: the SEMAUT rows are labeled strate-
gies, while the plan/goal rows are labeled plans. For im-
mediate purposes, one can treat these labels as synony-
mous. We provide reasons for our terminology later.
Finally, the SEMAUT model has a clear-cut decision
rule: select the strategy ifor which U(strategy i) is maxi-
mized. The values of U(strategy i) in turn depend on the
uij and the pj, with a clearly stated additive/multiplicative
formula (expected utility). In the plan/goal formulation,
we merely say that that each plan is evaluated in terms
of the vjand wij, without committing to any formula for
such evaluation. While in fact, an additive/multiplicative
formula might be useful here also, e.g.,
U(plan i) =
n
X
j=1
wij vj
Anecdotal evidence, at least, suggests that many different
decision rules may, in fact, be used. A focus on adoption
of a decision rule is an important aspect of constructed
choice theory. Below we discuss some possible decision
rules (and their relationship to measurement of vjand wij).
Note that even if an additive/multiplicative formula is
used to evaluate each plan, the resulting plan/goal model
might be quite different from SEMAUT because the val-
ues of the goals, indeed, the goals themselves, might be
strongly context-dependent.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 143
Table 2: General plan/goal structure for decision making.
Matrix entries are decision weights for different goals,
conditional on events
Active goals
Possible plans G1G2... Gn
plan 1 w11 w12 ... w1n
plan 2 w21 w22 ... w2n
... ... ... ... ...
plan m wm1wm2... wmn
Goals have subjective values: vj=v(Gj).
Plans have decision weights for each goal:
wij =w(Gj|plan i)
Plan i is evaluated in terms of the vjand wij.
The SEMAUT framework can require rather complex
utility measurements (uij) for the conjunctions of differ-
ent goals that are bundled together within one outcome
oij. Multi-attribute utility may be especially hard to mea-
sure for the mixture of goals that are involved in protec-
tive decision making. The plan/goal structure, by con-
trast, demands a separate (context-dependent) value vjfor
each goal. We illustrate this difference in Tables 3 and 4,
in which the matrices of Tables 1 and 2 are instantiated
for a simplified version of a decision concerning flood in-
surance.
3.2 Example: To purchase flood insurance,
or not
Tables 3 and 4 have only two rows. We achieve this
simplification by assuming that only one insurance pol-
icy is available, thus the only two strategies are to pur-
chase insurance or not. In Table 3, each of the two
rows is wide enough to encompass several different out-
come components (financial cost, hassles, various feel-
ings) listed as subheadings. The columns distinguish four
events: E1, no flood; E2, a flood in which one manages
to avoid major property damage; E3, a flood that leads
to major property damage; and E4, a destructive flood
that causes catastrophic property damage. We treat E1
through E4as mutually exclusive and exhaustive. For
concreteness assume that these four events have respec-
tive annual probabilities 85%, 9%, 4%, and 2%. Such
probabilities could be the output of an elaborate scien-
tific model, or might represent the decision maker’s best
(subjective) judgment. The utility assigned to a specific
outcome depends on the levels of the components, i.e.,
the level of financial cost, hassles, anxiety, etc.
By contrast, the rows in Table 4 are narrow, since they
need contain only decision weights — in this case, just
probabilities — but there are many columns, representing
multiple goals.
To use Table 3 as a decision aid, one needs to eval-
uate subjectively some rather complicated conjunctions
of consequences. For example, if one does not pur-
chase insurance and a damaging flood occurs, what is the
(dis)utility of the resulting combination of large financial
loss, acute anxiety at the time of the flood, chronic anx-
iety prior to the flood, major hassles, and major regret?
It might be difficult for a person to imagine this full sce-
nario and how unhappy he or she would be, and therefore
difficult to measure the utility of this complex outcome.
In Table 4, the flood-insurance decision is represented
by a two-row matrix with eight columns, many of them
corresponding to different aspiration levels5for avoid-
ance of financial loss or avoidance of negative emotions.
Each of these eight goals is assumed to have its own value
vj.Avoiding a catastrophic financial loss is one goal —
the minimal financial goal, since a major loss might still
occur. Avoiding any major loss is a higher aspiration,
and avoiding any loss, even a small one, is the highest
aspiration. The values assigned to such an ordered set of
goals are increments in value, as in cumulative prospect
theory; thus, someone who avoided any loss would attain
all three aspirations and would get all three vjvalues. In
Table 4, most of the outcome components shown in Table
3 have been transformed into goals, especially, losses or
emotions to be avoided. Not all eight goals may be active
or important for a given decision maker. Goals relating
to avoidance of hassles could also be included (they are
outcome components in Table 3), but we omit them here
because the hassles, minor or otherwise, are unavoidable
if a flood occurs, whether or not one purchases insurance.
Thus, they are irrelevant to the decision. In using Table
4, the value vjfor each active goal would be estimated
separately.
The rows are narrow in Table 4, because in this ex-
ample, the conditions under which a given plan achieves
each goal can mostly be stated as numeric probabilities,
by cumulating the event probabilities from Table 3 ap-
propriately. For example, the decision maker will avoid
catastrophic financial loss and vast regret, under the no-
insurance plan, provided that there is no destructive flood,
an event with probability .85 + .09 + .04 = .98. Note
that this cumulation of events relative to a particular goal
is akin to the formalism of cumulative prospect theory
5Much previous work has been devoted to goals that serve as ref-
erence points on a continuum, such as aspiration level (Shapira, 1995;
Fischer, Carmon, Ariely & Zauberman, 1999; Heath, Larrick & Wu,
1999; Fox & Hadar, 2006;). The present use of the goal concept is in-
tended to include such cases, as we discuss further in connection with
prospect theory below. Our goal concept is broader, however, and de-
rives from social psychology (Lewin, 1951) and from cognitive theory
(Newell, 1990).

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 144
Table 3: Strategy/event matrix for purchase of flood insurance
Events
E1 E2 E3 E4
event description no flood flood causes damaging destructive
little damage flood flood
event probability .85 .09 .04 .02
Strategy 1: purchase flood insurance
Outcome components
financial cost premium premium premium premium
hassles none minor major great
chronic flood-related anxiety none none none none
acute anxiety (at flood) none little little little
other feelings regret justification justification justification
Strategy 2: no insurance
Outcome components
financial cost none small large catastrophic
hassles none minor major great
chronic flood-related anxiety some some some some
acute anxiety (at flood) none much much much
other feelings justification relief major regret vast regret
(Tversky & Kahneman, 1992), and indeed, the model we
present here is motivated by that theory.
3.3 Goals, plans, and constructed choice
With additive utilities, the change from Table 3 to Table
4 could be viewed as just the rearrangement of displayed
information. For example, in Table 3, under strategy 1,
the additive utility of “feel justified” could be separated
out and multiplied by .09 + .04 + .02 = .15, correspond-
ing to the flood events E2,E3,E4; while in Table 4, un-
der plan 1, the value of attaining the goal “feel justified”
would again be multiplied by the weight factor .15. In
other words, we are just taking the components of multi-
attribute utility from Table 3 and turning them into goal
values for the columns of Table 4, and we are cumulating
the event probabilities from the columns of Table 3 and
inserting them appropriately as matrix entries in Table 4.
There are several reasons why the change from a strat-
egy/event matrix to a plan/goal matrix is deeper than just
a rearrangement. All of them relate to the theory of con-
structed choice.
Our first point is seen most easily by contrasting Tables
3 and 4 as guides to a decision about flood insurance. The
SEMAUT structure incorporates feelings, but at the price
of asking difficult questions about the utility of multi-
attribute outcomes compounded from heterogeneous el-
ements — financial outcomes combined with an assort-
ment of different feelings that may arise. To elicit such
utilities involves probing artificial decision situations that
are likely to puzzle respondents far more than the orig-
inal question about whether or not to purchase flood in-
surance. Preferences are constructed, not revealed, and
heterogeneous goals make their construction difficult.
When we put goals first, as in Tables 2 and 4, we drop
such artificial compounds. The goals whose values are
elicited are simple rather than compound. There does re-
main the question of how the goal values vjwill be used to
select a plan. The plan/goal approach moves the compli-
cations away from multi-attribute utility elicitation into
the problem of evaluating plans using specific decision
rules. Of course, there do exist natural compound goals
that can be reported as a single column in Table 4. An
example of a natural compound protective goal would be
insuring against both water damage and wind damage,
in a coastal zone subject to hurricanes.6The plan/goal
formalism can include natural compound goals as sin-
gle columns, but it does not force one to create extrane-
ous compounds. Difficult tradeoffs among heterogeneous
6Today homeowners are forced to purchase two separate policies
to protect themselves against wind (homeowners insurance) and water
(flood insurance) from hurricanes. (Kunreuther 2006).

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 145
Table 4: Plan/goal matrix for purchase of flood insurance
Goals
feel avoid avoid avoid avoid avoid avoid avoid
justified small major catastrophic chronic acute major vast
Plans loss loss loss anxiety anxiety regret regret
Plan 1: Purchase flood
insurance
.15 0 1 1 1 mostly 1 1
Plan 2: No insurance .85 .85 .94 .98 0 .85 .94 .98
goals are still present in the plan/goal setup, but they are
taken into consideration at the level of decision rules for
selecting among competing plans, not at the level of util-
ity measurement.
Second, the goals considered by the decision maker are
context dependent, as are their subjective values v(G),
which function as contingent weights, in the sense of
Tversky, Sattath and Slovic (1988). Context elements,
including some of the plans available as options, can sug-
gest goals that might not otherwise be considered. This
is almost standard when one goes to a new restaurant,
reputed to be excellent: the menu offers plans, or com-
ponents that can be put together into plans; seeing what
is available often leads the diner to adopt a goal that is
entirely novel (e.g., ostrich livers in garlic butter) or to
change the value vjfor an existing goal (e.g., a low- calo-
rie diet). The same can be true for protective goals: one
can adopt a new goal or revise the valuation of an exist-
ing goal in light of a protective device offered for sale
in a catalog (e.g., a new type of car alarm) or in light of
a change in the provision of an insurance contract. (e.g.
raising the coverage limit could increase the weight given
to peace of mind).
An innovative plan may be selected because it seems
more likely that an important goal can be achieved, i.e.,
the decision weight wfor that goal is high, for the plan
in question. Here, however, we are making a different
point: it is not merely that the decision weight wis high,
rather, it is the value vfor the goal that is changed when
the particular plan (or any other context element) makes
that goal salient. In the extreme, the goal might be one
that was entirely unknown to the decision maker before
the context made it salient. We discuss this further below
in relation to Tversky-type intransitivities.
Third, the uncertain events on which outcomes depend
are relegated to the cells of the matrix. In Tables 2 and 4
we show probabilities or decision weights in the cells.
The thinking underlying the determination of decision
weights is hidden in the matrix representation, but it has
to be made explicit when one uses this model. For exam-
ple, the weights in Table 4 are cumulative sums of prob-
abilities for an ordered series of events. In other settings,
decision weights may arise from subjective support, i.e.,
the weighing of evidence (Tversky & Koehler, 1994). De-
cision weights may also be affected by poor timing in the
delivery of an outcome (e.g., untoward delay) or by in-
completeness. This is illustrated by the entry mostly in
Table 4 when referring to the impact of purchasing flood
insurance on avoiding acute anxiety. One may not en-
tirely escape acute anxiety over an impending flood by
purchasing insurance, but greatly reduce it.
Fourth, Table 2 allows, but does not commit to a sum-
of-products decision rule. In the next subsection we dis-
cuss three classes of rules that we think play important
roles in some types of decisions, all of them different
from a sum-of-products rule. Suppose a sum-of-products
rule is used, and is formally identical to that used in
calculating SEMAUT because the uij are additive multi-
attribute utilities and the wij are sums of atomic prob-
abilities. The plan/goal framework is still not equiva-
lent to SEMAUT because the vjcan change with context.
Among other things, such context-dependence allows in-
transitivity of pairwise choice.
Our final point relates to the substitution of plan for
strategy. In many cases, nothing is gained or lost by
this change: in non-technical contexts, the two words are
more or less synonyms. However, strategy already has
a technical meaning in game theory, where it refers to a
choice element for a game represented in normal (rather
than extensive) form. That is, a strategy specifies exactly
what the player will do in every circumstance that arises
in the course of playing a game. This technical meaning
is one that we emphatically do not wish to convey: we
view plans as hierarchically organized, containing new
decision nodes within them. For example, a plan to buy
insurance would not normally include a specification of
how to travel from one’s home or workplace to an insur-
ance agent’s office. If the latter trip becomes necessary,
a new decision process is set in motion to decide how to
get there. Similarly, chess players (including grandmas-
ters) select plans, with room for improvisation; they can-
not select strategies, because the number of branches in

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 146
the chess tree is much too large to allow even one game-
theoretic strategy to be specified fully.
Our use of plan is partly drawn from Miller, Galanter
and Pribram (1960), and from the similar uses of this
term in the psychology of motor performance and prob-
lem solving. Like Miller, Galanter and Pribram, we
are concerned with behavior structure, and particularly
with the problem of specificity level in the field of de-
cision making. What is actually decided (consciously
or unconsciously) and what is simply done pursuant to
a plan already adopted (with minor parametric adjust-
ments guided by external circumstances)? We confront
this problem by making explicit the assumption that de-
cision making is plan selection. Executing a plan usually
requires many actions, but may not require any new de-
cisions. Sometimes, a plan leaves open a choice of sub-
plans at some critical juncture, and in that case, there is
an additional decision that has to be made.
The distinction between plan selection and plan execu-
tion is related to that between categorical and continuous
perception. We perceive objects on continua (size, dis-
tance, weight, friendliness, etc.) in order to adjust exist-
ing plans to reality, but we categorize objects in order to
decide what new plan (or new subordinate plan) should
be chosen. For example, one may be concerned about
fire hazard from old papers in a storage room, and may
select a plan that involves spending an afternoon clearing
them out. One has categorized the papers as sufficiently
at risk to adopt a goal of eliminating this specific hazard
and one selects a plan that can achieve it. When it comes
to executing the plan, details will vary depending on a
more continuous perceptual response. If one is unexpect-
edly done after one hour’s work, one begins to modify
the plan so as to make good use of the time gained by
finishing early. Similarly, a plan to seek insurance may
be triggered by categorization of a financial risk as too
great to ignore, but the execution of the search plan will
be a function of the details of information sources about
insurance plans.
3.4 Measurement issues and decision rules
Two crucial questions about the plan/goal framework in
Table 2 have not yet been addressed adequately:
(i) measurement: How can values vjand decision
weights wij be measured in practice?
(ii) decision rule: How are these measured values com-
bined when selecting among plans?
These questions are closely related. In one approach to
measurement (discussed below), numeric values are in-
ferred from observed choices, and this must be based on
a quantitative decision rule relating observed choices to
underlying vjand wij.
These questions can be addressed from either a de-
scriptive, normative or prescriptive standpoint. In the de-
scriptive mode, one asks what plan-selection rule a de-
cision maker actually uses in a given setting, and relates
numeric values of vjand wij to that decision rule. In the
normative mode, one asks what decision rule(s) can be
considered rational in a given setting. In the prescriptive
mode one asks how one can improve the choice process
for a particular problem/context given an understanding
of how actual behavior (descriptive) differs from what
is considered to be rational. One might aid a decision-
maker by using methods to “elicit” numeric values of vj
and wij. These numeric values can then be used in con-
junction with a normative decision rule to recommend
which plan should be selected.
There remains a widespread view that SEMAUT repre-
sents rational decision making, while constructed choice
represents actual choice behavior with its deviations from
rationality. Based on this view, questions about decision
rules in the plan/goal framework can only be descriptive.
We believe, on the contrary, that the claim to rationality
of SEMAUT is flawed, because the supposed preferences
sometimes do not exist, but can only be constructed.
SEMAUT may be an excellent deliberative preference-
construction tool, for some situations, but it is no more
than that. In our view, therefore, questions about deci-
sion rules and measurement have both descriptive and
prescriptive aspects. Prescriptions should be based on
normative principles consistent with the view that choices
are constructed, rather than based on definite preferences.
There still is much thinking to be done in order to at-
tain “reflective equilibrium” between formal principles
of rationality and human intuitions about good decisions;
the latter are essential, albeit sometimes deeply flawed.
Goodman (1955) discussed the equilibrium between in-
ductive intuitions and inductive logic. Rawls (1972) dis-
cussed a similar equilibrium between intuitions and the-
ories of justice. A similar equilibrium must be sought
with respect to decision making, and should, in our view,
consider a variety of “rational” rules for plan selection.
Measurement. For quantitative models of decision
making, there are two different approaches to measure-
ment of desired quantities: values or utilities (vjor uij)
and decision weights or subjective probabilities (wij or
pj). We refer to these two approaches as behavioral and
psychophysical.
The behavioral approach mirrors the concept of re-
vealed preference in economics: estimates of the desired
quantities are inferred from an individual’s choices. This
approach has dominated theoretical research on measure-
ment in relation to SEMAUT. For details and discussion,
see Chapters 5 and 8 in Krantz, Luce, Suppes and Tver-
sky (1971). It has sometimes been used, with limited
success, to estimate utilities and subjective probabilities

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 147
Table 5: Goal conflict for three insurance plans
Goals
Insurance plans Avoid catastrophc loss Avoid regretting a modest loss Minimize up-front costs
A excellent (high limit) OK poor (expensive)
B OK poor (high deductible) excellent (cheap)
C poor (low limit) excellent (low deductible) OK
from laboratory or field observations. Tversky (1967)
provided a paradigmatic example using tradeoffs between
cigarettes and candy in risky and riskless situations. A
broad class of later examples related to marketing re-
search is found in the use of conjoint analysis (Green &
Srinivasan, 1978; Gustafsson, Huber & Hermann, 2001).
In the psychophysical approach the desired quantities
are obtained from an individual’s numerical judgments
or comparisons of intervals. These measurement theories
are discussed in Chapter 4 of Krantz, Luce, Suppes &
Tversky (1971). We mention a few of the many examples
of the use of psychophysical methods, mainly tied to the
SEMAUT framework: Galanter (1962), Anderson and
Shanteau (1970), Seaver, von Winterfeldt and Edwards
(1978), Breault (1983), and Edwards and von Winterfeldt
(1987).
The behavioral approach requires commitment to a de-
cision rule, a law linking observed choices to underlying
measured values. For SEMAUT, the decision rule is max-
imization of expected utility. The conditions under which
pjand uij can be consistently measured from observed
choices are given in various axiomatizations, e.g., Sav-
age (1954), Krantz, Luce, Suppes & Tversky (1971), or
Koebberling & Wakker (2004). Methods of measurement
are implicit in the various constructive proofs of represen-
tation and uniqueness theorems for the expected-utility
representation (Krantz, 1991).
A behavioral approach within the plan/goal framework
would likewise depend on commitment to a decision rule,
a specific linkage between the vjand wij and the choice of
a plan. Our discussion of question (ii) below suggests
that several different decision rules may be used in dif-
ferent contexts. It therefore seems premature (at least) to
axiomatize behavioral measurement procedures based on
a particular decision rule. Estimating vjand wij based on
psychophysical judgment, or, in the case of wij, based on
the relationship between decision weights and probabil-
ities previously established in the literature on prospect
theory, is less problematic because it does not assume a
decision rule. In fact, such numerical estimates could be
used to test which individuals use a particular decision
rule in a given context.
Decision rules. We now turn to question (ii), types of
plan-selection rule that might be considered within the
plan/goal framework. We have already mentioned sum-
of- products maximization, similar to SEMAUT and a
generalization of cumulative prospect theory. Almost any
descriptive choice model that differs from SEMAUT can
be considered a candidate plan-selection rule. We dis-
cuss four here: Tversky’s additive difference model, ma-
jority voting by goals, contingent weighting models, and
conjunctive- choice models with thresholds.
The additive difference and majority-vote models ap-
ply only to pairwise choices. We illustrate their applica-
tion to pairwise choice among insurance plans, showing
that they can generate opposite intransitive cycles. Con-
sider someone who is choosing among three plans, A, B,
C, with three goals in mind: avoiding catastrophic loss;
avoiding regret should a modest loss occur; and avoiding
large premium payments. The three plans are shown as
rows and the three goals are the columns in Table 5. The
cell entries (decision weights) in this case are just ver-
bal descriptions of how well a given plan satisfies a par-
ticular goal. Plan A guarantees protection against catas-
trophic loss through a high coverage limit but is expen-
sive. Plan B is the least costly of the three plans. Plan C
has lower coverage limits so that the insured will have to
cover a significant portion of the losses if a catastrophe
occurs; however, it has a lower deductible than the other
two plans, so modest losses will not lead to regret as little
will be paid out of pocket.
Table 5 is structured so that for each pair of plans, there
is a large, presumably salient difference with respect to
one of the goals, but smaller and opposite differences
with respect to the other two goals. For example, Plan B
is much cheaper than A, but B has a coverage limit some-
what lower than A (less guarantee against catastrophic
loss) and B has a deductible somewhat higher than A, so
there is more chance of a modest loss that will lead to re-
gret. Similarly, Plan C has a much lower deductible than
B but is somewhat worse than B for the other two goals,
and Plan A has a much higher coverage limit than C but
is not as attractive with respect to the remaining goals.
The first thing to note is that someone might decide be-
tween any two plans by counting the number of goals that

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 148
Figure 1(a) Condorcet cycle
A > B & B > C & C > A
each by 2 goals to 1
A
B
Ccost &
deductible
coverage &
deductible
cost &
coverage
Figure 1(b) Tversky cycle
A < B & B < C & C < A
based on most salient difference
A
B
C
cost
deductible
coverage
Figure 1: Two bases for intransitivity
are better satisfied by each of them. This is the majority-
vote rule for pairwise choice, applied here to individual
rather than social choice, with the individual’s goals as
“voters.” This method may seem perfectly reasonable
for any one pairwise choice, but in Table 5, it leads to
the classic Condorcet/Arrow intransitivity, as depicted in
Figure 1 (a). The figure shows that A > B, B > C, and C
> A, each by a “vote” of two goals to one.
Alternatively, one might decide between two plans by
evaluating the differences in the decision weights for each
goal using a function that expands large differences rel-
ative to small ones, and then integrating across the goals
by adding up the signed differences. This is a special case
of the additive-difference theory of Tversky (1969). If the
difference between excellent and OK is taken as one unit
and that between OK and poor is also one unit, and if
one cubes the differences (preserving sign), then the A,
B difference reverses: B > A, because the cost difference
is worth 23= 8 units, while the regret and coverage dif-
ferences are each only -1. In fact, the whole intransitive
cycle is reversed, as shown in Figure 1 (b).
There is strong evidence that, in multi-attribute situa-
tions, people tend to make within-attribute comparisons
early and often (Russo & Dosher, 1983; Payne, Bettman
& Johnson, 1993). This supports the idea that differences
with respect to particular goals are evaluated first, and
then integrated in making a choice. Russo and Dosher,
in fact, showed that some subjects do choose the alter-
native that has the majority of confirming dimensions:
a vote-count decision rule. One might guess that vote
counts would often be used to integrate multiple differ-
ences when decision making is deliberative and tradeoffs
among different goals are difficult, whereas something
like the Tversky mechanism would often be used where
time is short and the decision maker looks for the most
salient difference between two plans.
While there have been many models for pairwise
choice, fewer apply to larger sets of alternative plans.
Contingent weighting models (Tversky, Sattath and
Slovic, 1988) are an important class that apply to multi-
alternative choice. They also relate well to context-
dependent constructed choice. One could account for
the Tversky-type intransitivity (Figure 1b) by contingent
weighting by assuming that large differences with respect
to a particular goal lead to a high weight on that attribute.
In the context of the plans/goals model, a high attribute
weight increases the vjfor that goal.
Finally, a threshold model is one that requires that one
or more goals be satisfied to some minimum degree —
the threshold is set with respect to the decision weight.7
An insurance purchaser who considered plans such as
those depicted by Table 5 might first set a threshold of
“pretty likely” for the goal “avoid catastrophic loss.” This
decision-weight threshold for one goal would exclude
plan C because of its low coverage limit. By eliminating
C, the conflict between A and B might be resolved in fa-
vor of A (two goals to one) or in favor of B (minimize up-
front cost being given a very high weight). If, instead, the
7The model might be called conjunctive when thresholds are set for
two or more goals simultaneously.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 149
decision weight threshold for “avoid catastrophic loss”
were set at “nearly certain” only plan A could be chosen.
A decision-weight threshold could also be set for each of
two goals, e.g., “pretty likely” to avoid catastrophic loss
and also to avoid regret. This would eliminate plans B
and C, leaving A as the only choice. A decision-weight
threshold of “nearly certain” for both those goals would
create severe conflict, since none of the plans available
satisfy these criteria.
In summary, behavioral measurement of vjand wij,
within the plan/goal framework, would depend on a par-
ticular plan-selection rule, which would have to be used
both during the measurement process and in applica-
tion to subsequent decision problems. Design of behav-
ioral measurement would depend on the particular deci-
sion rule and would be facilitated by axiomatization of
measurement relative to that rule. The plan/goal frame-
work forces the decision analyst to think about what plan-
selection rules are actually used for a particular problem,
and about their appropriateness for prescriptive analysis.
A psychophysical approach to measurement of vand w
parameters would facilitate investigation of various plan-
selection rules.
3.5 Advantages of the plan/goal framework
The overriding difference between the plan/goal and the
SEMAUT framework lies in what questions are asked and
answered most naturally. The plan/goal framework leads
one to raise many different questions that are not usually
considered under SEMAUT.
A strategy/event structure, as in SEMAUT, focuses on
the uncertain events that determine what outcomes will
occur, given the choice of a particular strategy. Consider-
ation of the decision- maker’s goals is secondary. In fact,
consideration of multiple goals is inhibited, because it re-
quires evaluating complex multi- attribute outcomes for
each cell of the strategy/event matrix. A plan/goal layout
focuses on active goals and asks how likely or to what ex-
tent each plan can achieve that goal. The approach is thus
similar to the one taken by Keeney (1992) on the impor-
tance of values and goals as the driving force in decision
making.
Since it is easy to think about each goal, the plan/goal
schema is easy to apply. In analyzing a choice among
plans, one tries to encompass all the goals under consid-
eration, to ask which ones actually are considered and
whether any important goals are not yet included. One
also can ask probing questions about each goal. Is it a
product of the particular context? Would the decision
maker pursue the goal if the context made it less salient?
Is a particular goal underweighted because of the context?
The plan/goal representation appears to capture the in-
surance decision making process, as illustrated by the fol-
lowing example (already mentioned in Section 1).
People often purchase flood insurance after suffering
damage in a flood, but then many cancel their policies
when several consecutive years pass with no flood (Kun-
reuther, Sanderson & Vetschera, 1985). A simple expla-
nation, using Table 4, is that avoiding anxiety and feeling
justified are both important goals. Following flood dam-
age, anxiety is high, and reducing it is a salient goal; it is
also easy to justify buying the insurance, since a flood has
just occurred. Thus, plan 1 is selected, based strongly on
feel justified and avoid anxiety. After several years, many
people may find that the prospect of a flood no longer
troubles their “peace of mind” so anxiety avoidance now
has low value (v). Meanwhile, insured individuals do not
feel justified in continuing to pay premiums and not col-
lecting on their policy; the unfulfilled feel justified goal
becomes more salient. The differential weighting of these
two goals can lead to not purchasing insurance. Note that
this theory predicts that a decision maker who puts heavy
weight on the goals of avoid catastrophic loss and avoid
vast regret will likely continue to purchase flood insur-
ance year after year if the cost is modest. A decision advi-
sor, confronting someone who wants to cancel insurance,
might well ask the individual about the role played by
feeling that insurance is unjustified, about the importance
of having a good justification for purchasing insurance,
and about whether protection against catastrophic losses
might justify paying the premium.
As noted above, the plan/goal framework also raises
the useful question of what plan-selection rules are used
and should be used. It thus implicitly raises questions
about the effects of uncertainty (wij) on any given goal
and the effects of time delay on the importance of each
goal. One can also ask whether a goal is valued for itself
(vj) or valued as a resource in pursuit of other goals. The
discussion (above) of plan selection versus plan execution
also leads one to ask whether choosing a particular action
was based on a decision process (conscious or not) or on
a previously selected plan.
4 Taxonomy of insurance-related
goals
In this and the next section, we apply the general frame-
work of Section 3 to consumer insurance decisions. Our
general framework assumes that choice involves the se-
lection of a plan among several alternative options, that
most plans are designed to achieve multiple goals, and
that the set of goals to which a plan is directed may or
may not be fully integrated into a single coherent mental
representation with a clear evaluative component. Choice
context (which includes the particular set of plans made
explicitly available) affects which goals are considered,

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 150
what value is placed on each goal, and how each plan is
evaluated. Evaluation of a plan may take into account un-
certainty about achieving various goals, the time at which
different goals are likely to be achieved, who is likely to
benefit or lose in each case, and what resources are likely
to be needed. This section focuses on the goals under-
lying insurance decisions, while Section 5 considers how
these goals and the process of constructed choice account
for anomalies in consumer insurance decisions.
We discuss five main goal categories: sharing financial
risk, getting a return from an investment, emotion-related
goals, satisfying legal or other official requirements, and
satisfying social and/or cognitive norms.
(1) Financial protection by risk sharing. Individuals
can purchase insurance at relatively low cost and be finan-
cially protected against a catastrophic loss if the negative
event in question has low probability, there are many at
risk, and occurrences of the event are statistically inde-
pendent.
Some people may also hope for financial protection
against negative events that have relatively high proba-
bilities and relatively small financial impacts. An exam-
ple would be floater insurance that covers property that
is easily movable. In this case, one does expect to make
insurance claims and be reimbursed, perhaps many times
during a period of years. Such a goal might be predi-
cated on the belief — usually, but not always erroneous
— that total premium payments, over a period of years,
will be less than aggregate claims from the series of neg-
ative events. One believes, in effect, that the insurer will
lose money in the long run. One might also hold this
goal without such a belief, by simply neglecting proba-
bility considerations and focusing on the consequences
if the event occurs. Failing to think about probability and
believing that an insurer will lose in the long run, are cog-
nitive errors. There is nothing paradoxical or uncommon
about maintaining a goal on the basis of a wrong belief
and/or an error in reasoning.
(2) Benefits of investment. Life insurance and annu-
ity contracts sometimes combine financial protection for
beneficiaries in the event of the death of the insured per-
son with various investment benefits, such as capital ac-
cumulation and/or regularly received payments. These
contracts are attractive because people do have such in-
vestment goals. Health insurance policies usually do not
accumulate cash value or pay dividends, but often provide
other financial benefits, such as free or low-cost annual
checkups and discounts on prescription drugs.
The ability to be reimbursed, irregularly but frequently,
for small losses may be perceived as a dividend stream
coming from insurance contracts with low deductibles.
The goal of getting fairly regular returns can be distin-
guished from the goal of protection against small losses,
discussed above. For the investment goal, one does not
expect a net positive cash flow from the insurance; one
merely expects some kind of cash-back return. In this
sense, many individuals view one of the goals in pur-
chasing an insurance policy as getting a financial return
on their investment. Those who do not collect on their
policies for several years period feel that their premiums
have been wasted. It is hard to convince them that the
best return on an insurance policy is no return at all.
Consider the case of flood insurance. As pointed out
above, many individuals who are not required to have
insurance cancel their coverage if they have not made a
claim over the past several years (Kunreuther, Sanderson
& Vetschera, 1985). Such behavior would be understand-
able for people who revise the probability of a loss down-
ward in the light of experience. Most people respond that
the probability and the consequences of a similar event
remain about the same as before and they are generally
correct in this view.
Furthermore flood insurance in the United States has
been highly subsidized on existing homes by the federal
government so that the purchase of coverage has posi-
tive expected value in addition to protecting against other
catastrophic events. We hypothesize that it is unpopular
because it fails to provide cash-back returns.
(3) Emotion-related goals. There is a growing litera-
ture on how affect and emotional goals impact on in-
dividuals decisions under risk (Loewenstein et al 2001;
Finucane et al 2001). Three goals in this category with
respect to insurance are reduction of anxiety (i.e. peace
of mind), avoidance of anticipated regret and consolation.
Because emotions, even anticipations of anxiety or regret,
have considerable immediate presence, insurance expen-
ditures to satisfy these goals now may lead to a shortage
of funds to pursue goals more distant in time.8
We noted above that for low-probability, high-impact
events, one benefit from purchasing insurance is being
protected against the risk of a large financial loss. In addi-
tion individuals may buy coverage to reduce their anxiety
about experiencing such a financial loss. It is important
to separate these two goals, financial protection against
the loss and reduction of anxiety about the loss, because
people vary as to the importance of each goal, situations
vary in the degree to which they make financial losses
vivid and to which they provoke or relieve anxiety, and
the relative importance of these goals may change over
time. One may also anticipate anxiety, and take measures
to avoid it. For example, some people claim that they
85 We thank Jonathan Baron for pointing out this aspect of emotion-
related goals.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 151
refuse to fly, not because they fear a crash, but because
they anticipate and dislike being anxious about a crash.
However, if one cannot avoid anxiety about a loss, one
may still find opportunities to reduce the anxiety by tak-
ing protective measures, including insurance, where ap-
propriate.
Regret (Loomes & Sugden, 1982) is quite different
from anxiety, in that it is primarily experienced after a
loss occurs rather than before. Consider the example of
mailing a package worth $50. Insurance may be readily
available. If one does not purchase it, then if the package
is lost or badly damaged, one is quite likely to wish that
one had purchased the coverage. Sometimes, the emotion
of regret accompanying such a wish is quite unpleasant.
If, at the time of mailing, one anticipates unpleasant re-
gret if an uninsured loss occurs, then one may decide to
purchase insurance as a way of avoiding the possibility of
such regret.
Individuals may also purchase insurance as a form of
consolation should they suffer a loss. In particular, if one
has special affection for an item, such as a piece of art,
then the knowledge that one can make a claim should the
item by destroyed or stolen has special meaning to the
person. Hsee and Kunreuther (2000) attribute the need
for consolation as the reason why individuals are willing
to pay higher premiums for the same amount of coverage
for objects they love than for those where they don’t have
special feeling.
Usually, a strong positive attachment to an object either
has no effect on the probability of damage, theft, etc., or
may even reduce this probability, if extra care is taken.
Indeed, in a recent study of willingness to purchase war-
ranties (Piao & Kunreuther, 2006), subjects believed that
loving an object made it seem less likely that the object
would need repair than if one was neutral or disappointed
with the object. This was true whether or not statisti-
cal information about repair frequencies was given. This
same study also showed that love did not, on average, pro-
duce a significant change in the anticipated cost of repair.
If anything, anticipated cost decreases for objects that one
loves. People should thus be less willing to purchase war-
ranties for loved objects than for ones for which they have
no special affection, but, in fact, are more willing to do
so.
(4) Satisfying requirements. Insurance coverage is of-
ten mandatory. Automobile liability insurance is required
by most states, homeowners insurance is normally re-
quired by mortgage lenders, flood insurance must be pur-
chased as a condition for a federally insured mortgage in
special flood hazard areas, and malpractice insurance is
needed for several different professions.
In these cases, purchase of insurance may be viewed as
a subgoal for meeting endgoals such as owning a car or a
home or practicing one’s profession.
(5) Satisfying social and/or cognitive norms. Many
insurance decisions are based on what other people are
doing, or on what respected others think is an appropri-
ate action to take. For example, a new parent may pur-
chase life insurance mainly because his or her own par-
ent, partner, or financial advisor thinks that it is important
to provide protection for the spouse and child, and the
amount purchased might follow some standard guideline
(e.g., three times annual income). Once again, multiple
goals may come into play: the new parent may be try-
ing to achieve goal (1), financial protection of the fam-
ily against a low-probability high- impact event, but also
may be trying to do what others expect or wish.
There is also empirical evidence that purchase of insur-
ance, like adoption of new products, is based on knowl-
edge of what friends and neighbors have done (Kun-
reuther, et al., 1978). There is a vast literature on social
influence, some of it especially relevant to protective de-
cision making (e.g., Riad, Norris & Ruback, 1999). For
present purposes, however, we mainly want to distinguish
between “non-extraneous” social influence — those ac-
tions and opinions of other people that provide useful in-
formation to a decision maker about the probability of a
catastrophic event, about the likely consequences of such
an event, or about the nature of insurance plans that could
be advantageous — versus social influence that seems ex-
traneous, in the sense used here.
A clear-cut demonstration of extraneous social influ-
ence would show an associated change in the likelihood
of selecting a particular plan involving insurance that is
unaccompanied by changes in beliefs about the probabil-
ities or consequences of a loss event. An illustration of
this behavior came from a pretest interview of an earth-
quake questionnaire when a homeowner hearing that his
neighbor had purchased earthquake insurance indicated
that he would want to buy such coverage himself with-
out changing his beliefs about the risk he was facing or
knowing about the actual cost of coverage (Kunreuther,
1978).
Numerous other examples can be cited. In our discus-
sion of flood insurance using a plan/goal matrix (Table
4), we introduced feel justified as a possibly important
goal. Someone who purchases flood insurance soon after
suffering damage from such a disaster may do so in part
because it is easy to justify the expenditure by pointing
to the flood that just occurred. Cancellation of insurance
coverage after some years of coverage may occur by us-
ing the social norm that it is hard to justify an expenditure
that has not paid off.
In fact, people are concerned with justifying their deci-
sions to themselves and others (Shafir, Simonson & Tver-
sky 1993). In the process, people often use arguments

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 152
that have little to do with the tradeoffs between the cost
of insurance and the expected loss that forms the bases of
economic analyses of insurance or warranty transactions
(Hogarth & Kunreuther 1995).
5 Explaining insurance decisions
and anomalies
In this section we examine whether SEMAUT, Cumu-
lative Prospect Theory (CPT), and the present model of
constructed choice can explain what appear to be anoma-
lies or suboptimal insurance decisions. It will be obvious
that one of our reasons for preferring the constructed-
choice model to SEMAUT has to do with the context-
sensitivity of insurance decisions. We argue, however,
that quite apart from this criticism, neither SEMAUT nor
CPT comes close to describing the realities of insurance.
5.1 The expected-utility account and its
limitations
We begin by considering the simple classical account of
insurance decisions, based on a concave utility function
for total assets. Since insurance contracts normally have
negative expected value, it is natural to account for the
fact that people do purchase insurance by hypothesiz-
ing a risk-averse (concave) utility function U(A), relat-
ing utility Uto total assets A. Such a concave function
accords with the more qualitative concept of catastrophic
loss. Compare the reduction in assets by a loss Lor by
a loss one tenth as large, L/10. For a concave curve,
the decrease in utility for loss Lis more than 10 times
larger than the decrease for loss L/10. Thus, the decision
maker prefers to make a premium payment Q=L/10
rather than to accept a 10% chance of the loss L. More
generally, the decision maker prefers to pay premium Q
rather than to accept a probability pof loss Lif and only
if
U(A−Q)> pU(A−L) + (1 −p)U(A)(1)
We next give a concrete example, both to illustrate the
reasoning given above and as an introduction to the limi-
tations of this idea. Suppose that a decision maker has an
exponential utility function for total assets, specifically:
U(A) = 1 −e−A/A0(2)
This utility function approaches an asymptote of 1 for
very large A; the parameter A0is the asset level for which
utility difference from zero asset level is about 63.2% of
the difference between zero assets and the saturated max-
imum utility level.
Note that 1/A0is usually thought of as the Arrow-Pratt
measure of absolute risk aversion, i.e., it is the curvature
−U00/U 0of the utility function. The exponential function
is often used because it has the simple property that ab-
solute risk aversion is independent of asset level A. Here,
however, we find it more useful to interpret A0as a gauge
of the magnitude of a loss L. To do this, we view zero as-
sets as a natural reference point. An increment or decre-
ment in assets of A0, which spans over 60% of the util-
ity range between zero assets and the utility asymptote
can safely be classified as a large gain or loss. The usual
(Arrow-Pratt) interpretation of A0is local: its inverse is
the curvature of the utility function. By treating zero as-
set level as a natural reference point, we are able to give
it this more global interpretation.
An example: the “loading” factor of an insurance
contract. Consider a household with total assets of
$300,000, including a home worth $200,000, and sup-
pose that for this household, A0happens to be $100,000.
Suppose that the probability of a severe fire or natural
disaster that would destroy its home is about 1/400 per
year. By the preceding criterion, the loss of $200,000
would be viewed as a large loss for this household. Its
expected annual loss, however, is $200,000/400 = $500.
The household might be able to purchase insurance that
would fully reimburse a $200,000 loss for about $1000
annual premium: this would allow the insurer to pay
claims, cover administrative costs, and make a profit. Be-
cause of the sharp curvature of the exponential utility
curve, the household should be happy to pay $1000 an-
nually for this coverage; in fact, with this utility function,
the maximum value of Qthat satisfies inequality (1) is
about $1585.
This account of insurance seems plausible in the given
example. However, it does not fare well when it comes to
smaller losses. First, at the opposite extreme, it predicts
unwillingness to pay much more than an actuarially fair
price to eliminate deductibles.
Suppose that the household under discussion has a 1/20
chance per year of fire or damage producing a loss of
$1000 or less. Its utility is nearly linear with money over
a range of $1000, and so the household should be will-
ing to pay only a few cents more than the expected loss
to eliminate a $1000 deductible. In fact, the premium in-
creases substantially to eliminate a $1000 deductible, and
many people are willing to pay it.
Even if one is willing to treat behavior vis-à-vis de-
ductibles as an anomaly that the theory simply does not
address, there remains the problem of insuring against in-
termediate losses. Consider the household in the above
example with a home worth only $100,000 that could be
destroyed by fire or natural disaster. The expected annual
loss is cut in half, to $250, but because of the concavity of

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 153
0.00
0.05
0.10
0.15
0.0 0.5 1.0 1.5 2.0
loss (relative to 63.2%−saturation asset level)
probability of loss
Figure 2:
maximum acceptable loading factor for insurance
(exponential utility)
Figure 2: Maximal acceptable loading factor for insurance (exponential utility)
U, willingness to pay for insurance drops by more than a
factor of 3 to less than $430. If the household can insure
a $200,000 home for $1000, the premium for a $100,000
home will be at least $500; and so the theory predicts that
it would prefer to self-insure for $100,000!
This is not a minor anomaly. Many types of insurance
contracts have loading factors of 2.0 or more: the insurer
charges at least twice the expected loss, in order to cover
claims with a safety factor, cover administrative costs and
make a profit. Homeowners insurance, even with a large
deductible, generally has a loading factor of more than 2.
For example, Sydnor, 2006 reports a loading factor of 2.6
in one western state for policies with $1000 deductible,
and much larger loading factors for decrements in the de-
ductible. The first example shows that for a loss of 2A0,
with probability 1/400, the maximum acceptable load-
ing factor for the household in question is $1585/$500 =
3.17, a quite reasonable figure. But if the loss in question
is only A0=$100,000, the maximum acceptable loading
factor is $430/$250 = 1.72, a value that might be diffi-
cult or impossible to find in the market for homeowner’s
insurance.
Maximum acceptable loading factor for a discrete loss
L.One can solve inequality (1) and equation (2) for the
maximum premium, Q, that a household is willing to pay,
as a function of the loss Land the probability p, and ask
when the ratio Q/pL will be at least 2. Figure 2 shows
a contour plot of the maximum acceptable loading fac-
tor, Q/pL. Here L, the horizontal axis, is scaled in units
of A0. In the above example, 1 unit is $100,000. The
vertical axis is just probability, confined to the domain of
reasonably low-probability events. What the figure shows
dramatically is that acceptance of loading factors of 2 or
more requires losses that are at least 1.4 A0(increasing
to 1.6 for higher loss probabilities). All the maximum
acceptable loading factors exceed 1; thus, as we already
knew, exponential utility leads everywhere to insurance

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 154
seeking; but under this theory, realistic prices will be paid
only for insuring losses that are large, on the scale of the
63.2% saturation constant. Since even wealthy people
routinely insure possessions worth a few tens of thou-
sands of dollars, accepting loading factors of 2 or more,
the theory is contradicted widely by actual consumer in-
surance behavior.
Since Figure 2 is scaled by A0, it is hard to argue that
the conclusions would be changed by any plausible de-
gree of risk aversion. To explain why a millionaire would
insure a $50,000 possession with an insurance loading
factor around 2, one would have to assume that $50,000
> 1.4 A0, or A0< $35,000. But for the millionaire, a loss
leading to total assets of only $35,000 would be a finan-
cial disaster.
Similar calculations could be made for more realistic
scenarios, in which there is a continuous probability dis-
tribution of losses rather than a discrete probability for
a single known loss, and with alternative concave utility
functions. Calculations with hyperbolic and power utility
functions (for which the zero asset level is a very clear
reference point) suggest that results similar to those in
Figure 2 are fairly general.
Note that Figure 2 makes it easy to see why the expo-
nential utility fails to account for decisions to have low or
zero deductibles. In the extreme case, for a loading factor
of 2 to be acceptable for eliminating a $100 deductible,
so that one has full coverage, the individual must satisfy
the equation $100 ≈1.4 A0, thus, A0≈$70. This means
that the decision maker in question would prefer a sure
$70 over a 60% chance to win billions of dollars9.
In light of the fact that people do insure against small
and intermediate losses, and in light of the discussion
of insurance-related goals in Section 4, one may wish
to modify the preceding simple theory by using the full
strength of SEMAUT to incorporate multiple goals for
insurance decisions, while retaining the core idea of a
concave utility function for total assets. In Section 3, we
suggested one reason why this might be difficult, or might
lead to an unsatisfactory theory. Permanent goals with
stable tradeoffs might indeed lend themselves to such a
theory. Goals that are context dependent and hence influ-
enced by what appears salient and/or goals that are sug-
gested by the very alternative plans among which the de-
cision maker must choose can not be incorporated in a
SEMAUT model.
In addition to the above argument based on context-
dependence, there is another reason why SEMAUT can-
not explain behavior. This is the phenomenon of refer-
ence dependence, which provides a segue to discussing
Prospect Theory. Many financial goals are perceived as
deviations from the status quo, rather than as changes
9This argument is similar to that of Rabin (2000), which is set en-
tirely in a frame of lotteries rather than insurance.
in total assets. The exponential utility function partially
deals with this perception, since its prescribed choices
among lotteries are independent of wealth and can thus be
interpreted in terms of increments and decrements from
the current asset level. However, even with an exponen-
tial utility curve, expected utility theory does not account
for the difference between loss framing and gain fram-
ing for the same increments and decrements. For exam-
ple, many owners of relatively new automobiles would be
happy to pay a $40 additional premium per year to have
a small deductible, say $100, on their collision insurance,
rather than a $500 deductible on an otherwise identical in-
surance contract. Yet the same owners would balk at pay-
ing $40 for a lottery ticket that wins $400 just in case their
automobile happens to suffer appreciable collision dam-
age during the coming 12 months. One would view such
a lottery ticket as unlikely to win; moreover, the prize is
not very large, even if one does win. So $40 seems much
too high a price. From the standpoint of utility theory,
however, the lottery ticket is at least as valuable as the
lower deductible on the insurance contract. The lottery
ticket nets $400 if there is appreciable collision damage.
The insurance contract nets the full $400 only if the dam-
age exceeds $500.
5.2 Prospect theory and its limitations
Characteristic features of prospect theory. Prospect
theory (Kahneman & Tversky, 1979) was developed as a
descriptive account of risky choice. One important type
of context effect — the effect of reference point — is ac-
cepted as an empirical fact and used as a starting point for
the theory. Current assets serve as a reference point, rela-
tive to which losses are differentiated sharply from gains,
and are valued quite differently. The theory replaces the
concave utility function for total assets with a value func-
tion that has one branch for changes perceived as gains
and another for perceived losses. Investigations of lotter-
ies with known probabilities and gains or losses as out-
comes led to a specification of the general form of the
value function — concave for gains, but convex for losses
— and to a general form for decision weights as a func-
tion of specified probability. Decision weights are ap-
plied as multipliers to the values that arise from gains and
losses. For present purposes, the most important feature
of the decision weight is that low probabilities, the ones
most relevant in insurance contexts, are overweighted.
Cumulative prospect theory (CPT) was developed by
Tversky and Kahneman (1992) for multi-outcome lotter-
ies. It deals with the phenomenon of rank-dependence of
the decision weights that are applied to particular gains or
losses. This has become the standard form for the theory.
For our simplified examples, where the insurance deci-
sion compares a sure loss of Q(i.e., the premium paid)

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 155
with a chance pto lose L(otherwise nothing), the distinc-
tion is unimportant, but we continue to use the acronym
CPT.
As a sidelight on CPT, one interpretation of rank de-
pendence might be that multiple gains or losses are con-
strued by the decision maker as defining a set of goals.
For example, consider a lottery in which one wins $1000
with probability 1%, $50 with probability 9%, and other-
wise (90%) nothing is won or lost. The CPT value for this
lottery can be written in the following form, where V(x)
is the value for gain xand W(p) is the decision-weight for
probability p.
V= [V(50)−V(0)]W(.10)+[V(1000)−V(50)]W(.01)
(3)
One can interpret this in terms of two goals: getting
at least $50, and getting $1000 rather than $50. The first
term multiplies the value of the first goal [V(50) - V(0)]
by a decision weight based on the cumulative probabil-
ity (9% + 1%) of achieving it, the second term similarly
multiplies the incremental value of the second goal (given
that the first has been achieved) by the decision weight for
1%. This makes sense when goals are ordered, as they
often are (e.g., getting at least a cost-of-living salary in-
crease, then perhaps getting a large increase; at least not
losing a chess game, then perhaps winning the game).
It is this form of the equation that suggested the general
sum-of-products decision rule for combining goal values
with decision weights in Table 2, i.e., Pvjwij , and thus
led to the more general discussion of possible decision
rules in a plan/goal framework.
CPT and insurance. Explanations of insurance deci-
sions based on CPT differ in a fundamental way from
ones based on expected utility. The CPT value function
is convex, not concave, in the loss domain — the opposite
of what might be thought appropriate for explanation of
insurance purchases with negative expected value for the
purchaser.
This shape is strongly supported by robust and oft-
replicated laboratory findings that decision makers are
risk-seeking, not risk-averse, in the domain of losses. The
question arises as to how to reconcile two basic facts:
people are risk-seeking with respect to losses, yet will-
ing to insure against losses.
The usual CPT explanation of insurance purchase is
based entirely on decision weights, rather than gains
and losses: people exhibit high decision weights to low-
probability events. This means that they are willing to
pay more than the expected loss for insurance.
This explanation has some intuitive psychological
plausibility: people worry (sometimes excessively, some-
times not) about low-probability high-negative-impact
events, and give them high weights in decision making.
Within the standard parameterization of CPT (Tversky &
Kahneman, 1992), however, this explanation is not vi-
able quantitatively, as we show next. Corresponding to
Inequality 1, the CPT condition for paying premium Qto
insure against loss Lwith probability pis simply this:
V(−Q)> W (p)V(−L)(4)
Define the maximum acceptable loading factor as λ=
Q/pL, where Qis the maximum willingness to pay for the
insurance. Combining this with Inequality (4) gives
V(-λp L) =W(p) V(-L). In the standard parameteriza-
tion of CPT, however, Vis a power function (using abso-
lute values) with exponent α. This means that the factor
Lαdrops out of both sides of the above equation, giving
a simple formula for λas a function of p:
λ=W(p)1/α
p(5)
Therefore the maximum acceptable loading factor is a
function of loss probability and is independent of L. In
Figure 2, depicting the exponential utility, the loading-
factor contours are fairly vertical, i.e., they depend mostly
on Land only weakly on p, especially for low or interme-
diate losses relative to the saturation asset level, and for
p > 1/100 . By contrast, in CPT the corresponding con-
tours would be perfectly horizontal, depending only on
pand not at all on L. Consequently, rather than plotting
loading-factor contours in (L, p) coordinates, we simply
plot maximum loading factor as a function of p.
The solid (lower) curve in Figure 3 shows the plot of
equation (5), using the standard CPT parameterization of
W(p) for losses. The maximum acceptable loading factor
is already below 2 for p> 2.6%, and falls below 1 for p>
25%. People with maximum loading factor below 1 will
not purchase insurance even when the transaction has a
positive expected value.
The loading factors below 1 are perhaps not fatal for
CPT: negative events with probability over 25% are un-
likely to have high impact (or we would all be in sorry
shape), and it may be both difficult and undesirable to
insure against them. However, loss probabilities in the
range between 3% and 10% are staples of insurance sales,
mostly at loading factors above 2. Thus, CPT, in its usual
form, simply cannot account for the market for insurance.
Figure 3 carries the calculation of loading factor down
to a loss probability of 1/100; but in fact, another prob-
lem for CPT is that the probability weighting function
derived from laboratory experiments cannot be extrap-
olated to very low probabilities, because at some point
many people in effect round a low probability down to 0,
saying “this means it won’t happen to me.” Thus, CPT
does not really deal with insurance against events with
very low probability but very high negative impact.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 156
0.0 0.2 0.4 0.6 0.8 1.0
0
1
2
3
4
5
6
7
probability of loss
maximum acceptable loading factor
Figure 3: loading factors in cumulative prospect theory
(standard parameterization)
NLIB interpretation
standard interpretation
Figure 3: Loading factors in Cumulative Prospect Theory (standard parameterization).
An alternative interpretation. Sydnor (2006) exam-
ined loading factors for deductibles in homeowner’s in-
surance. He reached a similar conclusion: neither ex-
pected utility (where he focuses chiefly on the power,
rather than the exponential form for the function U) nor
CPT in its standard form can account for people’s pur-
chase of low deductibles. However, Sydnor suggested an
important variant of CPT, according to which payments
made for purchase (including premiums for purchase of
insurance) are treated as negative gains, rather than as
losses. This is the “NLIB” (no loss in buying) interpre-
tation (Novemsky & Kahneman, 2005). This assumption
changes inequality (4): V(−Q)is replaced by −V(Q),
i.e., losing Qis replaced by losing the value that corre-
sponds to a gain of Q. The inequality becomes:
−V(Q)> W (p)V(−L) (40)
In the standard parameterization of CPT, the ratio
−V(Q)/V (−Q)is 1/2.25, the inverse of the loss-
aversion coefficient. The NLIB assumption therefore
leads to the loading-factor equation
λ=[2.25W(p)]1/α
p(50)
This is shown as the dashed curve in Figure 3. Since α=
0.88 in the standard parameterization, the dashed curve is
about 2.5 times higher than the solid curve. The loading
factors given by this curve do not reject prospect theory
out of hand as an account of insurance purchase. This
is consistent with Sydnor’s conclusion for deductibles on
homeowner’s insurance policies.
In short, Sydnor interprets willingness to purchase in-
surance as principally an effect of framing: though people
are risk-seeking in the domain of losses, reframing a sure
loss as payment of an insurance premium eliminates the
loss- aversion factor for the premium, and then the over-
weighting of low probability for potential losses covered
by the policy makes the insurance policy attractive. As

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 157
Figure 3 shows, Sydnor’s theory makes strong quantita-
tive predictions about willingness to pay for insurance:
the maximum loading factor should be independent of
loss magnitude, and people should purchase insurance at
loading factors of around 2.0 even for loss probabilities
between 1/2 and 1. These predictions, which seem un-
likely to be confirmed, could be tested more extensively
in laboratory and field studies of acceptable insurance
contracts.
Overweighting, underweighting and neglecting prob-
abilities. There is a more fundamental difficulty with
the CPT account of insurance purchase and Sydnor’s vari-
ation of the theory. People considering insurance con-
tracts rarely, if ever, have available explicit loss probabili-
ties. Often, loss probability does not seem to play a role in
their decisions (Camerer & Kunreuther, 1989; Hogarth &
Kunreuther, 1995; Huber, Wider & Huber, 1997). When
loss probability is in fact considered, it is derived from
experience, not from actuarial tables. However, Hertwig
et al. (2004) showed that when the probabilities are based
on experience, rather than on statistical summaries, peo-
ple underweight low probabilities in making risky deci-
sions except when there has been a very recent occur-
rence of the event class in question. Such underweighting
could, of course, be one important factor in phenomena
such as the cancellation of flood insurance policies that
was noted above; it may not be easy to separate under-
weighting of the probability from the difficulty in justi-
fying a decision to renew an insurance policy and from
fading anxiety. In any case, the overweighting postulated
in CPT may not be relevant to most insurance decisions.
It also seems implausible that people’s willingness to
pay for insurance is linked tightly to probability of loss
and not at all to magnitude of loss. On the one hand, peo-
ple often neglect probability in their thinking about insur-
ance; on the other hand, people undoubtedly pay some
attention to the affordability of losses. Insurance protec-
tion against very small losses, e.g., breakage of glassware
in one’s kitchen, would be viewed as absurd. The power-
function formulation of the CPT value function leads to
willingness to pay that is scale free, as shown in Equation
5 and 50above. This is one of the great conveniences in
applying CPT — it does not matter whether the monetary
amounts are dollars, Euros, or yen. But from the stand-
point of insurance, it is a weakness. Losing 100 yen is
much less serious than losing 100 dollars, and it is im-
plausible that the loading factor for insurance will be ex-
actly the same. In the domain of insurance, utility theory
makes more sense psychologically, since it postulates that
magnitude of loss does matter considerably. It would, of
course, be possible to replace the CPT value function by a
family of functions that do reflect the underlying scale of
gains and losses. The family of functions could be con-
structed to be linear over some range of sufficiently small
losses and gains, to be concave for gains and convex for
intermediate losses, and finally, to have another concave
region corresponding to catastrophic losses for a given
decision maker. Even such a function, however, would
fail to account for insurance decisions that are based on
multiple goals. In our view, a satisfactory understand-
ing of insurance behavior needs to take into account the
behavioral evidence that a number of different goals are
sometimes considered in connection with insurance and
other protective decisions. It is to this multiple-goal per-
spective that we now turn.
5.3 Explanations based on constructed
choice
In the introduction we noted three classes of anomaly:
(1) insuring against non-catastrophic losses, (2) underin-
suring against truly catastrophic losses, and (3) exhibit-
ing sensitivity to “extraneous” factors. In exemplifying
these anomalies, we have already indicated several ways
in which the theory partially explains behavior.
Our classification of anomalies assumes that the main
goal for insurance purchase is protection against catas-
trophic financial loss. Failure to pursue this goal, when
it is readily attainable, as well as pursuit of other goals
are viewed as anomalous. It is natural to explain these
anomalies in terms of the alternative goals that people do
in fact pursue, such as avoiding regret, reducing anxiety,
getting a return on investment, or satisfying social norms,
as suggested in our taxonomy in Section 4.
According to the plan/goal account, choices depend
not only on goals, but on the decision rules used in plan
selection. For example, the discussion of possible in-
transitivity with multiple goals (Figure 1) indicated how
particular decision rules might increase the tendency to
choose a low deductible and/or decrease the tendency to
insure fully against a catastrophic loss.
5.3.1 A process-oriented schema for constructed
choice
In the remainder of this section, we focus on some of
the main cognitive processes involved in choice accord-
ing to the plan/goal theory. To this end, Figure 4 offers
a process-oriented schema for context-dependent con-
structed choice. The four dark arrows indicate some of
the psychological functions essential to plan construction
and plan selection, while the five dotted arrows indicate
mechanisms of human perception or memory through
which context influences plans, goals, resources and de-
cision rules.
Central to the schema, as discussed in Section 3, are
goals and plans. The heavy downward arrow from goals

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 158
Figure 4: A schema for constructed choice. (Dashed arrows show context effects.
to plans indicates the main way in which goals bear on
plans: via decision weights wij. Psychologically, this ar-
row corresponds to the decision maker’s beliefs about the
likelihood of attaining goal j, or the degree of approxima-
tion to goal j, if plan iis selected.
There is also an upward, dashed arrow from plans back
to goals. This represents a feedback process, whereby
a plan that is considered suggests additional goals that
could be achieved through that plan, or, more generally,
alters the importance values vjfor the various goals. This
feature was discussed at length with respect to Table 2.
As examples in which a goal is suggested or emphasized
by a plan, a person may consider installing a protective
device after seeing one offered for sale in a catalog, or
may opt for an innovative provision available as an option
in an insurance contract. We emphasize that this is not
merely a question of a plan being attractive because it
gives a high decision weight wto an existing important
goal. Rather, it is the value vof the goal that is changed
when it is made salient by the plan. In the extreme, the
goal may have been unknown, in effect, v=0, prior to
the decision maker’s considering a particular plan.
The upward feedback arrow is one of five dashed ar-
rows, each suggesting a different route by which context
can affect constructed choice. We regard this feedback as
a context effect because the plans initially presented for
consideration constitute an important part of the choice
context. In a fluid situation, where novel plans can be
sought and constructed, the two arrows reciprocally link-
ing goals and plans could be activated repeatedly.
The dashed arrow directly from context to goals repre-
sents another type of contextual influence. This arrow
is labeled memory activation (Cantor & Engle, 1993),
because the context serves as a memory cue to activate
concepts and thoughts related to particular goals. For ex-
ample, if one recently regretted one’s failure to purchase
insurance, then thoughts about the possibility of a similar
loss will remind one of that recent regret, and might thus
strongly activate the goal of avoiding regret. Similarly, if
one experienced anxiety because of lack of insurance, a
contextual reminder of that anxiety might strongly acti-
vate the goal of avoiding future anxiety. More generally,
most effects of context on emotion-related goals are tied
to this arrow.
Context can also affect what plans are available and
what decision rule is used to select a plan. These two
dashed arrows in Figure 4 are labeled with another pro-
cess in human memory: matching (Seifert, et al., 2002).
Generally speaking, matching is linked to recognition: a
present stimulus matches something stored in memory.
Here, the matching process establishes a correspondence,
or partial analogy, between the current choice context and
some familiar situation. Finding such an analogy often
leads the decision maker to recall the plan selected pre-
viously, and to select a partially analogous plan. Context
thus adds to the set of plans under consideration. For ex-
ample, if one has decided in one situation to self-insure
(or not) against small losses, one may in a different but

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 159
partly analogous situation seek an insurance plan with a
high (or a low) deductible.
Alternatively or additionally, if there is severe conflict
concerning the available plans, the decision maker may
recall the method used previously to resolve the anal-
ogous conflict. Thus, context can make an additional
decision rule available. This is most obvious in cases
where analogy is used to support a complex decision rule,
such as estimation and maximization of subjective ex-
pected utility, but context can also suggest simpler de-
cision heuristics. For example, if one has previously
simplified a decision problem by eliminating some of
the available plans (e.g., by setting a threshold decision
weight for an important goal), one may be led to attempt
a similar simplification in the current choice situation.
The final dashed arrow in Figure 4 goes from context
to resources. One psychological process that can mediate
this influence is mental accounting (Kahneman & Tver-
sky, 1984; Thaler, 1999); we use it to label this arrow.
Changing the arrangement of mental accounts may either
increase or decrease the resources available to solve a de-
cision problem, and may thus suggest new plans or elim-
inate plans that draw on untapped resources. Here is an
illustrative anecdote. A couple planning a temporary par-
tial move to a not-too-distant city was considering how
to manage their two owned vehicles through the duration
of this move. Each plan considered had severe disadvan-
tages. In the context of a recent car rental, however, it oc-
curred to them that in addition to their existing vehicles,
money could be used as a resource to solve transporta-
tion problems. They could simply lease an additional
vehicle in the new city. Although this solution involved
additional expense, it seemed superior to previously con-
sidered plans. As long as their mental accounting only
concerned the management of existing vehicles, the bet-
ter solution was unavailable; their insight was that money
could be added to the resources already in play.
Insurance decisions are often influenced by consider-
ations related to resources. A person might choose not
to pay a $150 annual premium for general accident in-
surance (paying $500,000 for loss of life in any acci-
dent involving a common carrier), but might pay $25
for $500,000 insurance on a single flight. The $150 an-
nual premium could be reframed temporally (less than
$3/week), or it could be framed as an expenditure within
an insurance budget (find the needed $150 by increasing
the deductible on homeowners and/or automobile colli-
sion insurance), or yet alternatively, it could be framed
as an expenditure within a general household budget
(save $150 by bring lunch from home every day for two
months). Context (including advertising) determines how
such an expenditure is framed.
Without going into great detail, we remark that under
the plan/goal theory, mental accounting is not always “ir-
rational.” A decision rule that treats each of several goals
as extremely important might well lead to selection of
a plan that sets aside or otherwise designates resources
devoted to each goal. As noted in Section 3, such deci-
sion rules cannot be considered irrational. Indeed, group
decisions (by governments and other organizations) use
explicit budgets as a resolution of goal conflict. For indi-
vidual decisions, similarly, resolution of large intra- indi-
vidual goal conflict may only be feasible through such ac-
counting, whether done mentally or through written bud-
gets, dedicated savings, etc. Yet excessive rigidity in ac-
counting may indeed be irrational: this is shown most
clearly when small shifts in context lead to redrawing the
boundaries of such accounts.
Figure 4 provides a rich set of possible mechanisms
that should be considered in a general theory of deci-
sion making and in particular as they relate to insurance
anomalies. The first two classes of anomaly, insuring
against non-catastrophic losses and underinsuring against
truly catastrophic losses, relate directly to the multiple
goals involved in selecting an insurance plan. Since goals
can be made salient by plans, and salient goals may dom-
inate the decision rule that is used to choose a plan,
it is easy to see how marketing of insurance plans af-
fects choices, and how the default settings for parameters
within insurance plans also affect choices. Flight insur-
ance is sold at airports, package insurance at post offices,
credit-balance life and disability insurance is offered to
credit-card applicants and bill-payers, dormitory theft in-
surance is offered to parents of college students, etc. In
all these cases, the available insurance plans make salient
one or more goals that might get little or no weight in
a different context. Insurance plans also define temporal
frames for planning (e.g., month, year, or lifetime), and
thereby affect mental accounting and alter the perceived
resources available.
The mechanisms shown in Figure 4 also help explicate
some of the effects of “extraneous” factors on insurance
decisions. Positive or negative affect attached to an object
or event activates emotion-related goals. Thus, thinking
about the possible loss of a loved object activates a goal
of avoiding regret, and insurance may be perceived as re-
ducing regret, even though the financial benefit of the in-
surance is no greater than in the case of an unloved (but
needed) object. We have already discussed how recent
experience of events, such as flooding, can also activate
emotion-related goals.
What friends and relatives advise or themselves decide
to do is often extraneous, i.e., it does not change the set of
plans available or expectations about probability or mag-
nitude of loss. Adherence to quasi-legal or legal norms
and to social norms is itself an important goal, however,
and may strongly influence the selection of a plan. Some-
times the decision rule itself may be no more complicated

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 160
than adherence to a clear norm. One may purchase insur-
ance (or not) simply because that is the expectation of
one’s boss or one’s spouse. One may start wearing a seat
belt because it is the law to do so. There are also im-
portant cases of formally mandated protection: examples
include homeowners insurance, automobile liability in-
surance, and professional malpractice insurance, as well
as fire-safety inspections, required fire drills, and seatbelt
requirements. Large employers in the United States pro-
vide group health and life insurance. We also note the
important role of defaults (which in many cases can be
viewed as social norms) in affecting the details of chosen
insurance plans. Being able to justify decisions is itself
an important social norm; protective actions that other
people select themselves define a norm, and in addition,
doing the same is unlikely to be challenged or will be
easy to justify.
5.4 Novel prescriptive questions in con-
structed choice
The goal/plan approach to decision making raises pre-
scriptive questions of a sort not addressed in previ-
ous work on constructed choice. Consider the original
loss/gain framing demonstration of Kahneman and Tver-
sky (1979). The intent was purely descriptive: gain fram-
ing produces risk aversion, while loss framing, for logi-
cally equivalent outcomes, leads to selection of the risky
plan. There is no guidance about which frame, if either,
should be used. Advice about the right choice would pre-
sumably be given on an entirely different basis, maxi-
mization of expected utility. In the constructed choice
framework, however, the “true” preferences that lead
to maximization are mythical: all choices are context-
dependent. In the absence of other prescriptive guidance,
one is forced to consider whether a given problem should
be framed in terms of a loss or gain.
In what follows, we discuss in greater detail the two
features of our theory that are especially important for
understanding insurance anomalies. That plans available
in the decision context can suggest new goals leads to in-
sights about insuring against non-catastrophic losses and
about sensitivity to extraneous factors. We mention sev-
eral additional examples. The idea of decision mecha-
nisms involving decision-weight thresholds for important
goals helps us to understand failures to insure against
truly catastrophic losses.
5.4.1 Goals suggested by plans
The very idea of risk sharing through insurance was once
a novel family of plans. Prior to its invention, some severe
financial risks (especially in shipping and trading over
long distances) would have seemed unavoidable. The ad-
vent of insurance plans would have led many to adopt the
goal of reducing financial risk, just as the availability of
seat belts, and later of air bags led many to install them.
The availability of these innovations led to people adopt-
ing the goal of reducing their risk of death from automo-
bile accidents. At a later time, both became mandatory.
Indeed, the fact that people adopt new goals in response
to novel plans is probably one of the principal forces un-
derlying technological and social change.
In the realm of health insurance deductibles, a rather
important goal comes into play: one might wish to avoid
being placed in a future position where one can reduce
out-of-pocket costs by accepting increased health risks
for oneself or one’s family, e.g., by declining preventive
care or other advisable but optional treatments. Avoid-
ing such difficult tradeoffs is made salient by the avail-
ability of health-insurance plans that pay for routine and
preventive care. If the available health plans included
only hospitalization and major medical expenses, then
people would continually make health-care/money trade-
offs. They might very well wish they didn’t have to do
so; but if people were accustomed to making these trade-
offs, many would be reluctant to pay large up-front pre-
miums to gain coverage for routine and preventive care.
This also illustrates how plans affect temporal framing:
full-coverage plans place health-care decisions in a fairly
long-term frame, envisaging repeated use of the health-
care system, and thus make more salient the goal of
avoiding repeated difficult tradeoffs.
Though one can give many additional positive exam-
ples of important goals that are adopted in response to
novel plans, it is also clear that marketers can take advan-
tage of human goal malleability to generate plans that are
deceptively attractive to consumers and highly profitable
for the marketers. One broad area where this can occur is
in setting default values for plans.
A dramatic example of the effect of defaults was un-
covered with the introduction of no-fault automobile in-
surance in New Jersey and Pennsylvania. Both states in-
troduced the option of a reduced right to sue accompanied
by lower insurance rates. In New Jersey individuals had
to acquire the right to sue and in Pennsylvania the default
was the full right to sue. When offered the choice, only
about 20 percent of New Jersey drivers chose to acquire
the full right to sue while approximately 75% of Pennsyl-
vania drivers retained this right. In other words residents
in both states maintained the default option. This finding
was confirmed in a laboratory experiment where subjects
were given a similar choice, but were randomly assigned
to one of three groups: current automobile insurance with
full right to sue, with limited right to sue, or with no in-
formation about current right to sue. (Johnson, Hershey,
Meszaros and Kunreuther, 1993)

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 161
The over acceptance of default values has become a
well-known and much studied phenomenon, and proba-
bly is caused by several different underlying mechanisms.
Among these, we believe, is the influence of the default
on people’s actual goals. Many more New Jersey than
Pennsylvania motorists may have thought that they would
actually like to sue, since that option was endorsed as a
norm by the default plan they encountered. More gener-
ally, defaults may be viewed by consumers as goal-setting
norms, when in fact, they are often the most profitable op-
tion for a vendor and thus possibly the least favorable for
the consumer.
Insurance plans that bring in high profits to insurers,
in part by suggesting goals that might otherwise not be
adopted, include low deductibles, so-called disappearing
deductibles, rebate plans, credit-card insurance, flight in-
surance, and insurance on packages sent through the mail.
Low deductibles were discussed above. Disappearing
deductibles are plans for which the deductible applies in
the case of small and moderate losses, but not for large
losses. This may suggest the goal of avoiding an out-of-
pocket payout in a situation where one is already dealing
with the hassles of a major loss. This sounds attractive;
but, in fact, loading factors are high: the probability of
a major loss is low, the additional amount reimbursed by
the insurer with a major loss is not very high (only the
deductible), and so the average return to the insurance
purchasers is much less than the additional premium paid.
Rebate plans return a portion of the insurance pre-
mium, provided that no claim has been made in a speci-
fied period of time. These plans may lead consumers to
adopt the goal of getting an “investment return” on their
insurance premiums. A number of years ago, an insur-
ance company introduced a disability insurance plan for
which policyholders received at age 65 a full rebate of
all the premiums paid to that date, provided that no claim
had been made. This promised an attractive return, but
it created an economic incentive for those approaching
the age of 65 to avoid making a claim, if they had never
made one up to that point. People who were likely to need
disability payments found themselves not wanting to use
their policies. This reverse moral hazard led regulators
to request that the company withdraw this plan from the
market, since it undermined the main goal of insurance:
those who suffer an insured loss should be able to collect
the amounts for which they are covered.
Credit-card insurance and flight insurance are egre-
gious examples of plans that suggest goals in a particu-
lar context and exhibit very high loading factors. In each
case, the suggested goals can be achieved much less ex-
pensively. Flight insurance was discussed in our intro-
duction. The cost of this insurance is 4 to 8 times higher
than for the same level of coverage with general accident
insurance, and the latter covers many additional risks for
a much longer period of time. Credit-card insurance is es-
sentially life-and-disability insurance, covering only the
amount of one’s credit card debt. Its selling point is that
the current credit-card balance would be paid off in case
of death or disability. For people whose credit-card bal-
ances are burdensome, this would have been an attractive
goal. Unlike the usual life and disability policies, the pre-
miums for such insurance plans do not take age into ac-
count. For a person at low risk, or one who is already ad-
equately protected by life insurance and disability insur-
ance, adopting the goal of paying off a credit-card balance
in this way seems unwise: the risk is not catastrophic, and
with premiums in the neighborhood of $2 per $100 cov-
erage, the loading factor is high. We do not know who
actually purchases this insurance. If it is mostly people at
high risk, then the average loading factor may be reason-
able.
As a final example, consider insurance on packages
sent through the mail. Within the U.S., current rates are
slightly over $1 per $100 of insured value (more for in-
sured value less than $200, and up to a maximum of only
$500). While we do not know the probability of loss or
damage, experience suggests that it is far less than 1%.
The loading factor is high, and the losses involved are
seldom catastrophic. However, people generally do what
they can to protect objects sent through the mail; for ex-
ample, they may take care in preparing the package. An
insurance plan may be viewed as offering an additional
opportunity to take care.
5.4.2 Thresholds for decision weights
Above, we discussed plan selection mechanisms involv-
ing thresholds for the decision weights pertaining to a set
of important goals. We suspect that this is one of the most
frequently used mechanisms. It is observable when group
decisions are made in public: several goals are discussed,
and the plan finally adopted is one that offers a good
chance to achieve each of them. A plan for a new bridge
is selected because the decision makers believe that the
construction cost will be within the budget limit, that the
bridge will handle the expected volume of traffic, that it
will be safe, and that it will have a long lifetime with ac-
ceptable maintenance costs. Other goals (e.g., aesthetic
quality) may also be considered and traded off against
additional costs, but the four goals just listed are essen-
tial, and the likelihood of achieving each must be high.
One may ask, how high? Certainty is not achievable for
any of the goals. Most likely the threshold for acceptance
will be set very high for safety and lower for the other
goals.
When it comes to protective decisions, low probabili-
ties of negative events may fall within acceptance thresh-
olds. Table 4 provides a simple example. Someone may

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 162
regard the avoidance of a catastrophic loss as an essential
goal, yet the probability of 0.98 of achieving that goal
may be high enough to satisfy this person’s acceptance
threshold — in which case plan 2, no insurance, may be
selected.
In fact, failures to purchase flood or earthquake insur-
ance are very common. Individuals know that floods and
earthquakes occur, but they “do not think it will happen to
them.” This can be viewed as underweighting low prob-
abilities, consistent with Hertwig, et al. (2004). We take
the slightly different view that the mechanism is one of
acceptance thresholds for plan selection. Support for a
threshold view of decision making comes from an insur-
ance experiment by McClelland, Schulze and Coursey
(1993). For the case where there was a .01 chance of
losing $10 they found a bimodal distribution of willing-
ness to pay for insurance: over 25 percent would not pay
even a penny while approximately 20 percent were will-
ing to pay premiums that were four or more times the
expected loss. This suggests a decision weight for the
outcome no loss that exceeded the acceptance threshold
for those who would not even pay a penny, but was below
this threshold for many others who decided to purchase
insurance at premiums that implied extraordinarily high
loading factors.
Theories that postulate over- or underweighting of low
probabilities must commit to one or the other, for any
given probability level. Our view, by contrast, is that the
threshold level in a decision rule depends on the impor-
tance of the particular goal, on the importance of other
competing goals, and quite possibly on other factors, es-
pecially on social norms concerning acceptable levels of
risk. Thus, a given risk of failure to achieve the goal may
appear to be underestimated (as in the example above,
where a 2/100 chance of severe flood is neglected) or the
same probability, based on the same evidence, may ap-
pear to be overestimated (e.g., where a social norm de-
mands near certainty for a particular goal).
In the case of natural hazards, these acceptance thresh-
olds may well be too low prior to the occurrence of a dis-
aster, since the events can be catastrophic for uninsured
victims. These cases are complex; however, because for
some hazards, private insurers are reluctant to enter the
market or charge very high premiums due to high spatial
correlation of losses from disaster. Two things are needed
in such cases. One is a clear reinsurance mechanism, gov-
ernment and/or private, to spread risk more widely and
thereby lower premiums. The other is a norm that leads
people to purchase the insurance at these reduced premi-
ums. This could take the form of a social norm within a
group or the form of a requirement to insure as banks cur-
rently do with respect to homeowners insurance as a con-
dition for a mortgage. Mandates, sanctions, and social
norms all play a role in getting people to protect them-
selves adequately.
6 Prescriptive implications
In standard economic analysis, consumer tastes go un-
challenged. It is not irrational to pay $500 for an ex-
traordinary bottle of wine that cannot be purchased for
less. Paying $80 for a feeling of “peace of mind” like-
wise should go unchallenged. In a constructed choice
framework, however, these are not literal tastes, compa-
rable to liking raspberries or disliking prunes; they are
constructed choices. It is the construction process that
can be examined critically.
Construction of choice is influenced by education and
culture. Probability offers a powerful example. Four
hundred years ago, people undoubtedly had some strong
probabilistic intuitions, but these were not systematized,
and the word probable had a meaning quite different from
most of its modern senses.10 Today, people sometimes
use numerical probabilities in their reasoning and deci-
sion making, and those educated in probability or statis-
tics do so much more than others (Nisbett, Krantz, Jep-
son & Kunda, 1983). Another example derives from
the statistical associations among women’s education,
contraceptive use, and fertility rate. It may be hard
to infer causality unambiguously, but there seems little
doubt that in this domain, some combination of education
and cultural change has powerful effects on reproductive
decision-making.
From this viewpoint, consumer education could lead
to changes in protective decision making that are finan-
cially advantageous to most people and also leave people
more confident and more satisfied with their decisions.
Suppose, for example, that “peace of mind” is obtained
by paying $80 to reduce the deductible on homeowner’s
insurance from $500 to $250, and that the purchaser also
believes (rather accurately) that she, and others in compa-
rable situations average only about 1 claim every 25 years
that exceeds $250. If she thinks about a commitment to
10 years of insurance-with-peace-of-mind, this will cost
$800 extra over that time frame. The chance that she will
collect the extra $250 at least 3 times in the 10 years (thus
breaking about even) is about 1 in 100,000.
Alternatively, she can think about the commitment in
terms of cost per week: less than $2 for peace of mind.
Her choice may very well change, depending on whether
she thinks about the cost for peace of mind in a weekly,
annual, or ten-year time frame. In the weekly frame, $2
seems trivial and $250 looms large (despite its low prob-
ability). In the ten-year frame, $800 looms large and the
10The old meaning is still somewhat preserved in the legal term
’probable cause’.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 163
prospect of a $250 loss will then not have much effect on
peace of mind.
Educating the consumer to consider temporal framing
ranging from weeks to decades and to translate costs and
probabilities across such time frames will probably lead
to decisions that are more financially advantageous for
most consumers.
The constructed-choice framework in fact raises a host
of difficult prescriptive questions. We provide a sampler
here but note that most of the answers are beyond the
scope of this paper and should be addressed by future re-
search.
•What is the “right” temporal framing?
•Assuming that long-term temporal framing is usu-
ally advantageous, compared to short-term horizons,
what education programs are needed to overcome
myopia?
•What about framing outcomes as gains versus
losses?
•What is the best way to keep mental accounts, i.e.,
what resources should be committed to important
goals and what should remain fungible?
•What decision-weight thresholds should be set on
various goals?
•Are there intrinsically maladaptive configurations of
goals? When should one use social norms as a
guide?
•When should one pursue emotion-related goals, ver-
sus attempting to manipulate and control one’s own
emotions?
Rather than tackling these questions, we limit our pre-
scriptive remarks to three short subsections. The first
concerns financial benchmarks for insurance, the second
discusses emotion-related goals, and the third offers a few
specific ideas about consumer education.
6.1 Financial benchmarks
Financial benchmarks are useful starting points for crit-
ical analysis of constructed choices. For example, the
preceding example, in which someone pays $80 per year
extra to decrease a homeowner’s deductible from $500 to
$250 should raise a warning flag, because we think that
in the world of insurance contracts, higher deductibles are
almost always better. More generally, violation of any of
the benchmarks we propose here deserves careful analy-
sis.
The three simple benchmarks we suggest provide ad-
vice that runs counter to popular choice. The reason
is simple: insurance plans that satisfy prevalent non-
financial goals and also have high loading factors are both
profitable and easily marketable, therefore are frequently
offered and chosen. Less popular plans, when available,
sometimes offer excellent protection from financial risk
at reasonable cost.
1. Consider the maximum loss and its probability.
What is the maximum loss one could suffer from a partic-
ular type of negative event, how likely is such a loss, and
how much would it cost to insure fully against it? A per-
son earning $60,000 per year probably does not need $1
million life insurance; a $300,000 policy would give that
person’s beneficiaries five years to adjust to the loss of in-
come. The probability of death in the next year, from all
causes, is never negligible, and death can be financially
catastrophic for the survivors, so insurance equivalent to
a few years annual income should be purchased, if one
can find the money for the premium. Similarly, home-
owner’s insurance should, if at all possible, cover the full
replacement value of one’s house at least in case of fire,
since the probability of total destruction by fire is not neg-
ligible.
2. Look for the highest deductible. To opt for a high
deductible means that one is self-insuring against losses
smaller than that amount. One is thereby saving money
on insurance. Of course, the deductible must be set low
enough so that one can handle the loss if it occurs. Again,
a person earning $60,000 per year may not be able to
manage a $10,000 loss; however, a $2000 loss, although
very painful, might be manageable.
To illustrate the operation of these two benchmarks,
consider the following three homeowner’s insurance poli-
cies:
annual coverage
premium deductible limit
policy A $600 $1000 $100,000
policy B $750 $ 500 $100,000
policy C $750 $1000 $150,000
Assuming that the individual can manage a loss of
$1000, policy A is superior to B: $150 is much more than
the expected financial gain from reducing the deductible
by $500. Assuming that the home would cost $150,000
to replace, policy C achieves much more important goals
than B. Someone who purchased B and whose home was
destroyed by fire would be $50,000 short of the funds
needed to rebuild. The comparison of A with C is a bit
less clear. If the loading factor for the extra coverage is
about 2.0, then the extra $150 premium implies a proba-
bility of $75/$50000 = .0015 per year for a total loss by

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 164
fire. If more than 1 in 1000 homes of this type in this re-
gion are lost to fire each year, then $150 extra would be
well spent on maintaining the needed coverage.
3. Avoid policies with rebates or other return of pre-
mium. Consider the following two policies for disabil-
ity insurance:
annual
premium deductible rebate
policy A $1000 $600 $0
policy B $1600 $0 $600
Policy A dominates policy B financially: it costs $600
less, and B only provides the rebate in case the individual
makes no claim during the given period. In fact, by pur-
chasing policy A rather than B, the decision maker can
earn interest on the $600 that he or she has saved. Yet,
when the choice between these two policies was given
to a set of subjects in an experiment, a majority chose
B over A. (Johnson, Hershey, Meszaros and Kunreuther,
1993) Presumably many of these subjects adopted a goal
of maximizing the chance that they would get a return on
their insurance investment.
As noted in Section 1, there is another reason for avoid-
ing Policy B: the insured person may decide not to make
a justified claim in order to obtain the rebate. This is a
form of reverse moral hazard for insurance. It defeats the
main goal of the insurance.
6.2 Emotion-related goals
Emotion-related goals may lead to purchasing financially
unattractive insurance policies. Hsee and Kunreuther
(2000) tested the following scenario:
You are shipping two vases you purchased for
$200 each to your home. Suppose that the two
vases will be packed in the same box so that if
one vase is damaged, the other is also damaged,
and if one is not damaged, the other is also not
damaged. Of the two vases, you love one much
more than the other. You feel that the vase you
love is worth $800 to you and the other one is
worth only $200 to you. Suppose you have the
opportunity to purchase shipping insurance and
that you have enough money to insure only one
vase. Which one of the two polices will you
purchase?
Policy A: The insurance premium for the vase
you love is $12.
Policy B: The insurance premium for the vase
you don’t love as much is $10.
The decision maker should choose Policy B, because it
costs only $10, yet offers exactly the same financial ben-
efit as Policy A. When subjects were asked to make the
choices between the two policies 63.5% of the respon-
dents chose Policy A. Presumably policy A evoked a goal
of being consoled somewhat for the loss of the vase that
was loved, or the goal of showing how much one cares
about the loved vase.
Of course, this is a contrived situation, in which there is
no real benefit, emotional or otherwise, to paying $2 extra
for the insurance. The more general point is that attach-
ment to objects, and other s emotional goals, should raise
a flag for consideration as to how important the goals
are and what one is really getting from the insurance, in
the same way that violation of the financial benchmarks
raises a flag.
There are certainly situations where an individual
should be willing to violate financial benchmarks to sat-
isfy important emotional goals. Consider the following
example. A couple is renting a car for a vacation trip
and is asked by the rental company whether they want
to pay $2 extra per day to avoid paying up to $1000 if
they should have an accident. The couple can manage the
payment of $1000 if necessary, and the chance of an ac-
cident happening on any one day is much less than 1/500,
so paying the $2 has negative expected value. According
to the second financial benchmark, they should take the
$1000 deductible and save $2 per day. Suppose, however,
that the person arranging for the contract is unable to con-
vince her spouse of the benefits of following this rule he
will worry much less if they pay the $2 per day for full
coverage. She will probably conclude that $2 per day is
a small price to maintain peace of mind for her husband
and harmony on the vacation.
6.3 Consumer education
What should be the goals and methods of a program of
consumer education for insurance? Based on the model
of constructed choice that we have developed in this pa-
per, the most appropriate instructional strategy would fo-
cus on the goals that underlie consumers’ choices and
would highlight features of desirable insurance plans that
can achieve those goals that matter most to each person.
6.3.1 Extend the frame
It seems useful to consider competing plans in multiple
frames. The most obvious framing is temporal: a small
deductible that generates peace of mind at what seems
like a low price for one year looks much worse, as noted
above, when the expense is added up over ten years.
Probability of loss is also easier to understand in a longer
time frame: a probability of 4% per year translates into

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 165
about a 1 in 3 chance of having one or more losses in 10
years, i.e., a 2/3 chance of no loss at all. Extended tem-
poral framing also has the effect of putting the loss in a
long-run perspective. Paying $1000 out of pocket is very
painful for most people, but may seem more manageable
if it happens only once in 10 or 20 years.
Apart from temporal framing, it may make sense to
broaden the frame of any single insurance decision to in-
clude other protective needs. A financially unfavorable
choice in one domain, multiplied by 3 or 4 different do-
mains, can seem obviously wrong.
The goal of obtaining a return on one’s investment may
be more resistant to broadening the frame temporally,
since the return is also multiplied by the extended time
period. Thus, insurance needs to be framed alongside
other investments. Paying $600 extra for a rebate pol-
icy, which may or may not bring a return of $600, seems
obviously inferior to investing the $600 in a certificate of
deposit or in mutual funds. Such reframing is of course
only meaningful for those who perceive other investment
opportunities as accessible.
6.3.2 Provide economic incentives
Our suggestion here is predicated on the constructive in-
volvement of insurers and agents in consumer education.
The most obvious economic incentive that could be pro-
vided to an individual is a rebate check: imagine that the
person is contacted by his insurance agent, after having
purchased a policy with a low deductible, and urged to in-
crease it to a higher one. The agent could indicate to him
that he would save $100 by converting his policy from a
$500 deductible to $1000, and would be mailed a check
for $100 if he decided to do so.
6.3.3 Address emotional concerns
Individuals are likely to buy insurance in order to meet
emotion-related goals, such as achieving peace of mind,
avoiding regret, or gaining consolation in case of a loss.
Consumer education could point out paths by which they
can still meet some of these goals through insurance plans
that are more attractive financially. In the example given
above under emotion-related goals, it may not be difficult
to convince a person who was willing to pay $12 for an
insurance policy on a vase he loved that he would still
be able to gain consolation by only paying $10 to insure
the vase he didn’t love, since the claim would be based
on both vases being destroyed simultaneously. The fact
that the person didn’t purchase a policy on the vase he
loved would be irrelevant for obtaining compensation and
hence consolation.
It may also be necessary to challenge emotion-related
goals. Wanting to avoid regret is understandable; but
once the individual sees clearly that this is what he is do-
ing, he may think it is not worth the extra money.
6.3.4 Examples of instructional material
Understanding and use of probabilistic concepts is a mod-
ern development in human reasoning, and is attained
only through some combination of formal instruction, ca-
sual learning, and apprenticeship experiences. Nisbett,
Krantz, Jepson and Kunda (1983) introduced the concept
of statistical heuristics: short-cut guides to reasoning that
incorporate some probabilistic concepts and lead to their
use in everyday reasoning. Fong, Krantz and Nisbett
(1986) showed substantial temporary improvements in
everyday probabilistic reasoning from suitably designed
instructional modules. Protective decision making could
similarly benefit from instruction. In fact, individuals
can be aided in their understanding of the functions of
insurance by instructional material (Piao, Kunreuther &
Krantz, 2006).
More specifically, one way to convince people of the
importance of financial benchmark 1 is to provide graphic
examples of the importance of having protection against
a catastrophic loss rather than a small loss. A homeowner
would be more likely to purchase a policy with coverage
of $150,000 and a $1000 deductible rather than $100,000
policy with a $500 deductible if she was presented with a
clear explanation of the types of tradeoffs that would have
to be made in choosing between the two policies and the
impact on her resources if a fire destroyed her $150,000
house and she only had $100,000 worth of insurance.
6.3.5 The importance of social settings
There are vast advantages to conducting consumer edu-
cation in groups, rather than in the normal one-on-one re-
lationship between seller (insurance agent) and prospec-
tive buyer (policyholder). Social facilitation increases en-
gagement in learning. People learn even more by teach-
ing one another. Misunderstandings are given voice and
can more easily be corrected. Finally, the group conclu-
sions are more readily adopted as social norms.
6.3.6 Regulation, education, and paternalism
In the preceding, we have emphasized consumer educa-
tion partly in the spirit of libertarian paternalism (Sun-
stein and Thaler 2003) and partly in the pragmatic be-
lief that the issues involved are too diverse and change-
able to permit effective regulation. Insurance is already
highly regulated, and the regulations do prevent some
attractive but financially disadvantageous products from
entering the consumer market, but the prevalence of low
deductibles shows how difficult it is to regulate products
that people strongly desire. Unlike nicotine, however,

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 166
low deductibles are not physiologically addictive, and ed-
ucation may lead to the construction of quite different
choices.
Insurance providers are perhaps in the best position to
provide good instruction, and many independent insur-
ance agents pride themselves on doing so. Mandating and
monitoring consumer education by insurance providers
may be a valuable regulatory function.
References
Anderson, N. H., & Shanteau, J. C. (1970). Information
integration in risky decision making. Journal of Exper-
imental Psychology, 84, 441–451.
Attneave, F. (1974). How do you know? American Psy-
chologist, 29, 493–499.
Arrow, K. J. (1951). Social choice and individual values.
Yale University: Cowles Foundation for Research in
Economics, Monograph 12.
Bentham, J. (1789). An introduction to the principles of
morals and legislation. New York: Kegan Paul.
Beebe-Center, J. G. (1932). The psychology of pleasant-
ness and unpleasantness. Oxford: Van Nostrand.
Braun, M., Fader, P. S., Bradlow, E. T. & Kunreuther, H.
(2006). Modeling the “pseudodeductible” in insurance
claims decisions. Management Science, 52. 1258–
1272.
Breault, K. D. (1983). Psychophysical measurement and
the validity of the modern economic approach: A pre-
sentation of methods and preliminary experiments. So-
cial Science Research, 12, 187–203.
Camerer, C. & Kunreuther, H. (1989). Decision pro-
cesses for low-probability events: Policy implications.
Journal of Policy Analysis and Management, 8, 565–
592.
Cantor, J. & Engle, R. W. (1993). Working memory ca-
pacity as long-term memory activation: An individual-
differences approach. Journal of Experimental Psy-
chology: Learning, Memory and Cognition, 19, 1101–
1114.
Chapman, G. B. & Johnson, E. J. (1995). Preference re-
versals in monetary and life expectancy evaluations.
Organizational Behavior and Human Decision Pro-
cesses, 62, 300–317.
Condorcet, J.-A.-N. M. d. (1785). Essai sur l’application
de l’analyse a la probabilité des décisions rendues a la
pluralité des voix. Paris: Imprimerie Royale.
Cummins, J. D., McGill, D. M., Winklevoss, H. E. &
Zelten, R. A. (1974). Consumer attitudes toward auto
and homeowners insurance. Philadelphia: Wharton
School, University of Pennsylvania.
Debreu, G. (1960). Topological methods in cardinal util-
ity theory. In K. J. Arrow, S. Karlin and P. Suppes
(Eds.), Mathematical methods in the social sciences,
pp. 16–26. Stanford: Stanford University Press.
Diener, E. & Seligman, M. E. P. (2002). Very happy peo-
ple. Psychological Science, 13, 80–83.
Egan, D. E., & Grimes-Farrow, D. D. (1982). Differ-
ences in mental representations spontaneously adopted
for reasoning. Memory & Cognition, 10, 297–307.
Edwards, W. & von Winterfeldt, D. (1987). Public values
in risk debates. Risk Analysis, 7, 141–158.
Finucane, M. L., Alhakami, A., Slovic, P., & Johnson, S.
M. (2001). The affect heuristic in judgments of risks
and benefits. Journal of Behavioral Decision Making,
13, 1–17.
Fischer, G. W., Carmon, Z., Ariely, D., & Zauberman, G.
(1999). Goal-based construction of preferences: Task
goals and the prominence effect. Management Science,
45, 1057–1075.
Fong, G. T., Krantz, D. H., & Nisbett, R. E. (1986). The
effects of statistical training on thinking about every-
day problems. Cognitive Psychology, 18, 253–292.
Fox, C. R. & Hadar, L. (2006). Decisions from experi-
ence =sampling error +prospect theory: Reconsider-
ing Hertwig, Barron, Weber & Erev (2004). Judgment
and Decision Making, 1, 159–161.
Freud, S. (1920). A general introduction to psychoanaly-
sis. New York: Horace Liveright.
Galanter, E. (1962). The direct measurement of utility
and subjective probability. American Journal of Psy-
chology, 75, 208–220.
Goodman, N. (1955). Fact, fiction, and forecast. Cam-
bridge: Harvard University Press
Green, P. E. & Srinivasan, V. (1978). Conjoint analysis
in consumer research: Issues and outlook. Journal of
Consumer Research, 5, 103–123.
Gregory, J. & Lin, C. (1992). Constrained optimization in
the calculus of variations and optimal control theory.
New York: Van Nostrand.
Gustaffsson, A., Hermann, A. & Huber, F. (Eds.). (2000).
Conjoint measurement: Methods and applications.
Berlin; New York: Springer-Verlag.
Heath, C., Larrick, R. P., & Wu, G. (1999). Goals as
reference points. Cognitive Psychology, 38, 79–109.
Hertwig, R., Barron, G., Weber, E. U. & Erev, I.
(2004). Decisions from experience and the effect of
rare events. Psychological Science, 15, 534–539.
Hogarth, R. & Kunreuther, H. (1995). Decision making
under-ignorance: Arguing with yourself. Journal of
Risk and Uncertainty, 10(1), 15–36.
Hsee, C. K. & Kunreuther, H. (2000). The affection ef-
fect in insurance decisions. Journal of Risk and Uncer-
tainty, 20, 149–159.
Huber, O., Wider, R., & Huber, O. (1997). Active infor-
mation search and complete information presentation

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 167
in naturalistic risky decision tasks. Acta Psychologica,
95, 15–29.
Hull, C. L. (1951). Essentials of behavior. New Haven:
Yale Univ. Press
Johnson, E., Hershey, J., Meszaros, J. & Kunreuther, H.
(1993). Framing, probability distortions, and insurance
decisions. Journal of Risk and Uncertainty, 7, 35–51.
Kahneman, D., & Tversky, A. (1979). Prospect theory:
An analysis of decision under risk. Econometrica, 47,
263–292.
Kahneman, D., & Tversky, A. (1984). Choices, values
and frames. American Psychologist, 39, 341–350.
Keeney, R. L. (1992). Value-focused Thinking. Cam-
bridge: Harvard University Press.
Keeney, R. L. & Raiffa, H. (1976). *Decisions with mul-
tiple objectives: Preferences and value tradeoffs. New
York: Wiley.
Koebberling, V. & Wakker, P. P. (2004). A simple
tool for qualitatively testing, quantitatively measur-
ing, and normatively justifying savage’s subjective ex-
pected utility. Journal of Risk and Uncertainty, 28,
135–145.
Krantz, D. H. (1991). From indices to mappings: The
representational approach to measurement. In D. R.
Brown & J. E. K. Smith (Eds.), Frontiers of mathemat-
ical psychology: Essays in honor of Clyde Coombs pp.
1–52. New York: Springer-Verlag.
Krantz, D. H., Luce, R. D., Suppes, P. & Tversky, A.
(1971). Foundations of measurement Vol. I, Additive
and polynomial representations. New York: Academic
Press.
Kunreuther, H. (2006). Comprehensive disaster insur-
ance: Has its time come? In R. Daniels, D. Kettl &
H. Kunreuther (Eds.), Risk and disaster: Lessons from
hurricane Katrina pp. 175–201. Philadelphia: Univer-
sity of Pennsylvania Press.
Kunreuther, H. (1978). Disaster insurance protection:
Public policy lessons. New York: Wiley & Sons.
Kunreuther, H. & Pauly, M. (2006). Insurance decision-
making and market behavior. Boston, MA: Now Pub-
lishers, Inc.
Kunreuther, H., Sanderson, W. & Vetschera, R. (1985). A
behavioral model of the adoption of protective activi-
ties. Journal of Economic Behavior and Organization,
6, 1–15.
Lewin, K. (1951). Field theory in social science. Oxford:
Harpers.
Loewenstein, G. F., Weber, E. U., Hsee, C. K. & Welch,
N. (2001). Risk as feelings. Psychological Bulletin,
127, 267–286.
Loomes, G. & Sugden, R. (1982). Regret Theory: An
Alternative Theory of Rational Choice Under Uncer-
tainty. The Economic Journal, 92 (368), 805–824.
McClelland, G. H., Schulze, W. D., & Coursey, D. L.
(1993). Insurance for low-probability hazards: a bi-
modal response to unlikely events. Journal of Risk and
Uncertainty, 7, 95–116
Miller, G., Galanter, E. & Pribram, K. (1960). Plans and
the structure of behavior. New York: Holt, Rinehart
and Winston.
Newell, A. (1990). Unified theories of cognition. Cam-
bridge: Harvard Univ. Press
Nisbett, R. E., Krantz, D. H., Jepson, C., & Kunda, Z.
(1983). The use of statistical heuristics in everyday
inductive reasoning. Psychological Review, 90, 339–
363.
Nocedal, J. & Wright, S. J. (2006) Numerical optimiza-
tion. (2nd Ed.) New York: Springer.
Novemsky, N. & Kahneman, D. (2005). The boundaries
of loss aversion. Journal of Marketing Research, 42,
139–140.
Olds, J. & Milner, P. (1954). Positive reinforcement pro-
duced by electrical stimulation of septal area and other
regions of the rat brain. Journal of Comparative and
Physiological Psychology, 47, 419–427.
Patnaik, S. & Karibasappa, K. (2005). Motion planning
of an intelligent robot using GA motivated temporal as-
sociative memory. Applied Artificial Intelligence, 19,
515–534.
Payne, J. W., Bettman, J. R. & Johnson, E. J. (1993). The
adaptive decision maker. New York: Cambridge Uni-
versity Press.
Piao, X. & Kunreuther, H. (2006). Object-oriented affect
in warranty decisions. Manuscript: Columbia Univer-
sity.
Piao, X., Kunreuther, H & Krantz, D. (2006). Im-
proving rationality of insurance choices. Manuscript:
Columbia University.
Rabin, M. (2000). Diminishing marginal utility of wealth
cannot explain risk aversion. In D. Kahneman & A.
Tversky (Eds.), Choices, values and frames pp. 202–
208. New York: Cambridge University Press.
Rawls, J. (1972). A theory of justice. Oxford: Clarendon
Press.
Riad, J. K., Norris, F. H. & Ruback, R. B. (1999). Pre-
dicting evacuation in two major disasters: Risk percep-
tion, social influence and access to resources. Journal
of Applied Social Psychology, 29, 918–934.
Rosenbaum, D. R. & Krist, H. (1996). Antecedents of
action. In H. Heuer & S. Keele (Eds.), Handbook of
Perception and Action, Volume 2: Motor Skills, pp. 3–
69. San Diego: Academic Press.
Russo, J. E., & Dosher, B. A. (1983). Strategies for mul-
tiattribute binary choice. Journal of Experimental Psy-
chology: Learning, Memory and Cognition, 9, 676–
696.

Judgment and Decision Making, Vol. 2, No. 3, June 2007 Goals and plans 168
Savage, L. J. (1954). The foundation of statistics. New
York: Wiley.
Seaver, D. A., von Winterfeldt, D. & Edwards, W. (1978).
Eliciting subjective probability distributions on contin-
uous variables. Organizational Behavior and Human
Performance, 21, 379–391.
Sedikides, C., Ariely, D. & Olsen, N. (1999). Contextual
and procedural determinants of partner selection: Of
asymmetric dominance and prominence. Social Cog-
nition, 17, 118–139.
See, K., Heath, C., & Fox, C. (2006). Motivating individ-
ual performance with challenging goals: Is it better to
stretch a little or a lot? Manuscript: New York Univer-
sity
Seifert, C. M., Hammond, K. J., Johnson, H. M., Con-
verse, T. M., McDougal, T. F., & Vanderstoep, S. W.
(2002). Case-based learning: predictive features in in-
dexing. In T. A. Polk, & C. M. Seifert (Eds.), Cognitive
Modeling, pp. 987–1007. Cambridge: MIT Press.
Shafir, E., Simonson, I. & Tversky, A. (1993). Reason-
based choice. Cognition, 49, 11–36.
Shapira, Z. (1995) Risk taking: A managerial perspec-
tive. New York: Russell Sage Foundation.
Slovic, P. (1995). The construction of preference. Amer-
ican Psychologist, 50, 364–371.
Sunstein, C. and Thaler, R. (2003) “Libertarian paternal-
ism” is not an oxymoron. University of Chicago Law
Review, 70, 1159–1202.
Sydnor, J. (2006). Sweating the small stuff: The de-
mand for low deductibles in homeowners insurance.
Manuscript: University of California, Berkeley.
Thaler, R. (1999). Mental accounting matters. Journal of
Behavioral Decision Making, 12, 183–206.
Tversky, A. (1967). Utility theory and additivity analysis
of risky choices. Journal of Experimental Psychology,
75, 27–36.
Tversky, A. (1969). Intransitivity of preferences. Psycho-
logical Review, 76, 31–48.
Tversky, A. & Kahneman, D. (1992). Advances in
prospect theory: Cumulative representation of uncer-
tainty. Journal of Risk and Uncertainty 5, 297–323.
Tversky, A. & Koehler, D. J. (1994). Support theory: A
nonextensional representation of subjective probabil-
ity. Psychological Review, 101, 547–567.
Tversky, A., Sattath, S. & Slovic, P. (1988). Contingent
weighting and judgment and choice. Psychological
Review, 95, 371–384.
Tversky, A., Slovic, P. and Kahneman, D. (1990). The
causes of preference reversal. American Economic Re-
view, 80, 204–217.
Wise, R. A. (2004). Dopamine, learning, and motivation.
Nature Reviews Neuroscience, 5, 483–494.
Zhang, S. & Markman, A. B. (2001). Processing product
unique features: Alignability and involvement in pref-
erence construction. Journal of Consumer Psychology,
11, 13–27.