A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Experimental Psychology: Human Perception and Performance
This content is subject to copyright. Terms and conditions apply.
Journal
of
Experimental
Psychology:
Human
Perception
and
Performance
1986,
Vol.
12,No.4,53J-548
Copyright
1986
by
the
American
Psychological
Association,
Inc.
T»%-1523/86/J00.75
Reflection
in
Preferences
Under
Risk:
Who and
When
May
Suggest
Why
Sandra
L.
Schneider
and
Lola
L.
Lopes
University
of
Wisconsin—Madison
Conventional
economic
theory
assumes
that
people
are
uniformly
risk
averse.
Psychological
studies,
however,
have
shown
that
people
are
sometimes
risk
averse
for
gains
but
risk
seeking
for
losses,
a
phenomenon
termed
the
reflection
effect.
The
robustness
of the
reflection
effect
was
examined
both
within
subjects
and
across
subjects
differing
in
risk
style
for a set of
multi-outcome
lotteries.
Reflec-
tion
was
found
to be
weak
and
irregular
for all
choice
pairs
except
those
that
included
a
lottery
with
a
riskless
component.
The
latter
were
generally
preferred
for
gains
but not for
losses
by
both
risk-
averse
and
risk-seeking
subjects.
In all
other
choices,
risk-averse
and
risk-seeking
subjects
differed
systematically
from
one
another,
but in
ways
that
are
more
complex
than
pure
risk
aversion
or
risk
seeking
would
predict.
The findings
suggest
a
general
inability
of
weighted
value
theories
such
as
prospect
theory
(Kahneman
ATverskv,
1979)
to
adequately
describe
the
pattern
of
risk
preferences
over
individuals
and
over
the
full
range
of
lottery
types.
Such
inadequacy
suggests
the
need
for an
alternative
approach
to
risk
with
emphasis
on the
goals
and
strategies
that
individuals
bring
to the
risky
choice
process.
Since
the
time
of
Bernoulli, economists
have
noted that most
people
prefer
a
certain outcome
to a
gamble
of
equal expected
value.
This phenomenon
is
known
as risk
aversion.
Bernoulli
(1738/1967)
proposed
that such preferences arise because
peo-
ple
maximize
the
expected
utility
of
options.
He
suggested that
the
subjective
value,
or
utility,
of
money
is a
marginally
decreas-
ing
function
of
objective
value. Because such
a
function
is
con-
cave
everywhere,
a
person maximizing expected
utility
will
al-
ways
prefer
a
sure thing
to a
risky
option
of
equal expected
value.
Although
the
expected utility model
is the
cornerstone
of
many
current theories
of
risky decision making, recent evi-
dence
(Fishburn
&
Kochenberger,
1979; Kahneman
&
Tversky,
1979;
Laughhunn,
Payne,
&
Crum,
1980; Williams,
1966)
has
suggested
that
when
potential losses
are
involved, most people
prefer
a risky
option
to a
certain outcome
of
equal expected
value;
that
is,
they
are risk
seeking
in the
domain
of
losses.
Kahneman
and
Tversky
(1979)
labeled this switch
from
risk-
averse
preferences
for
gains
to
risk-seeking
preferences
for
losses
the
reflection
effect.
In
part because modern expected
utility
theory typically does
not
account
for
such
reflection,
Kahneman
and
Tversky
have
developed what
they
believe
to
be
a
more descriptive
and
comprehensive
model
of
preferences
under
risk. This model, embodied
in
what
Kahneman
and
Tversky
call
prospect
theory,
describes individual decision mak-
ing
under risk
as
consisting
of two
separate stages.
First,
pros-
pects
are
psychologically edited
in
order
to
simplify
their repre-
sentation,
and
second,
the
edited prospects
are
evaluated
in
terms
of
subjective
value
and
probability
weighting
functions.
The
present article
focuses
on the
reflection
effect
and its
the-
oretical interpretation
in the
prospect theory
framework.
An
outline
of the
essential
features
of
prospect theory
is
presented,
followed
by a
discussion
of the
conditions theoretically
neces-
sary
for
the
occurrence
of
reflection.
Next, what
little
empirical
evidence exists regarding
the
reflection
effect
is
considered.
Fi-
nally,
the
present
study
is
introduced
as an
expanded test
of the
reflection
effect.
Basics
of
Prospect
Theory
Prospect theory
is a
complex axiomatic description
of
deci-
sion
making under risk. Rather than discussing
it in
detail,
we
will
consider
only
those portions
of the
theory that
are
relevant
to the
present study. Kahneman
and
Tversky
(1979)
propose
that
before
prospects
are
evaluated,
they
are
psychologically
represented through
the
application
of
several editing opera-
tions. Of
primary importance
for the
reflection
effect
is
that
outcomes
are
coded
relative
to
some
reference
point,
usually,
but
not
always,
the
status quo.
Thus,
decision makers
are
seen
as
thinking
in
terms
of
gains
and
losses rather than
final
asset
positions.
Once
the
prospects
have
been edited,
they
are
evaluated.
First,
the
outcomes
and
associated probabilities
in
each
edited
prospect
are
interpreted according
to a
subjective value
func-
tion
and
a
probability
weighting
function.
These
subjective
in-
terpretations
are
then integrated quantitatively, using
a
format
similar
to
that employed
by
expected utility
theorists,
to
deter-
mine
the
overall worth
of
each
of the
prospects.
The
prospect
with
the
maximum worth
is
then
identified
and
chosen.
This
research
was
supported
in
part
by
Office
of
Naval
Research
Con-
tract
N00014-84-K-0065
to
Lola
L.
Lopes.
Correspondence
concerning
this
article
should
be
addressed
to
San-
dra L.
Schneider,
Department
of
Psychology,
University
of
Wisconsin,
Madison,
Wisconsin
53706.
Value
Function
An
illustration
of the
value
function
of
prospect theory
(which
we
will
call
the FT
value
function)
is
presented
in
Figure
1.
The
function
is
concave
for
gains
but
convex
for
losses, giving
535
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
