ArticlePDF Available

Abstract

We hypothesize that there is a general bias, based on both innate predispositions and experience, in animals and humans, to give greater weight to negative entities (e.g., events, objects, personal traits). This is manifested in 4 ways: (a) negative potency (negative entities are stronger than the equivalent positive entities), (b) steeper negative gradients (the negativity of negative events grows more rapidly with approach to them in space or time than does the positivity of positive events, (c) negativity dominance (combinations of negative and positive entities yield evaluations that are more negative than the algebraic sum of individual subjective valences would predict), and (d) negative differentiation (negative entities are more varied, yield more complex conceptual representations, and engage a wider response repertoire). We review evidence for this taxonomy, with emphasis on negativity dominance, including literary, historical, religious, and cultural sources, as well as the psychological literatures on learning, attention, impression formation, contagion, moral judgment, development, and memory. We then consider a variety of theoretical accounts for negativity bias. We suggest that I feature of negative events that make them dominant is that negative entities are more contagious than positive entities.
Negativity Bias, Negativity Dominance, and Contagion
Paul Rozin and Edward B. Royzman
Department of Psychology and Solomon Asch Center for Study of Ethnopolitical Conflict
University of Pennsylvania
We hypothesize that there is a general bias, based on both innate predispositions and
experience, in animals and humans, to give greater weight to negative entities (e.g.,
events, objects, personal traits). This is manifested in 4 ways: (a) negative potency
(negative entities are stronger than the equivalent positive entities), (b) steeper nega-
tive gradients (the negativity of negative events grows more rapidly with approach to
them in space or time than does the positivity of positive events, (c) negativity domi-
nance (combinations of negative and positive entities yield evaluations that are more
negative than the algebraic sum of individual subjective valences would predict), and
(d) negative differentiation (negative entities are more varied, yield more complex
conceptual representations, and engage a wider response repertoire). We review evi-
dence for this taxonomy, with emphasis on negativity dominance, including literary,
historical, religious, and cultural sources, as well as the psychological literatures on
learning, attention, impression formation, contagion, moral judgment, development,
andmemory.Wethenconsideravarietyoftheoreticalaccountsfornegativitybias.We
suggestthat1feature of negative events that make themdominantisthat negative enti-
ties are more contagious than positive entities.
Brief contact with a cockroach will usually render
a delicious meal inedible. The inverse phenome-
non—rendering a pile of cockroaches on a platter ed-
ible by contact with one’s favorite food—is unheard
of. More modestly, consider a dish of a food that you
are inclined to dislike: lima beans, fish, or whatever.
What could you touch to that food to make it desir-
able to eat—that is, what is the anticockroach? Noth-
ing! And the cockroach is far from unique: there is a
wide variety of animals (e.g., worms, caterpillars,
slugs, spiders) that share the cockroach potency,
along with a variety of microbially or toxin-contami-
nated objects. One of the best generic descriptions of
this relative power of negative contamination is em-
bedded in an age-old Russian adage: “A spoonful of
tar can spoil a barrel of honey, but a spoonful of
honey does nothing for a barrel of tar.” This apparent
dominance of negative over positive contamination is
played out on a vast human–social scale among the
large traditional segment of 800,000,000 living
Hindu Indians. People of higher castes are easily con-
taminated—that is, lowered in social status—by
contact with members of lower castes. The contami-
nation often occurs by eating food prepared by a
lower caste. On the other hand, when people of lower
castes consume foods prepared by higher castes,
there is no corresponding elevation in their status.
Stevenson (1954) summarized this feature of the
caste system with the phrase “pollution always over-
comes purity” (p. 50). The caste system pulls down-
ward; it is easy to pollute and hard to purify.
Similarly, in many Western and non-Western reli-
gious traditions, becoming possessed by a malevolent
demonic force is a relatively brief and easy affair
(Oesterreich, 1974), whereas the reversal of the pos-
session requires the painstaking, prolonged, and often
injurious ritual of exorcism. On the other hand, in
these same traditions, becoming “holy” or “saintly”
usually involves a long moral trajectory of positive
deeds, a state that can be compromised easily by one
or a few immoral acts. The general principle that, for
forgiveness to be achieved, the degree of acceptable
expiation must dramatically exceed that of the initial
fault, is dramatically played out in the Christian con-
cept of redemption from original sin. In a number of
traditions, there are some losses to one’s purity, such
as the sullying of female sexual honor through pre-
marital sex or illicit affairs, which cannot be reme-
died at all.
296
Personality and Social Psychology Review Copyright © 2001 by
2001, Vol. 5, No. 4, 296–320 Lawrence Erlbaum Associates, Inc.
PreparationofthispaperwassupportedbyNational Institute of Drug
Abuse Grant R21–DA10858–0, the Edmund J. and Louise W. Kahn
Chair at the University of Pennsylvania, and the Solomon Asch Center
for Study of Ethnopolitical Conflict at the University of Pennsylvania.
Requests for reprints should be sent to Paul Rozin, Department of
Psychology, University of Pennsylvania, 3815 Walnut Street, Phila-
delphia, PA 19104–6196. E-mail: rozin@psych.upenn.edu
The dominance of negative contamination affords a
particularly striking demonstration of what we take to
bea very general principle,a principle that holdsacross
awiderangeof domains,andinnonhumanaswell ashu-
man animals. The principle, which we call negativity
bias,is that in mostsituations, negative eventsare more
salient, potent, dominant in combinations, and gener-
ally efficacious than positive events. (There are excep-
tionstothisclaim, buttheyconstituteaminorityof cases
and often involve special circumstances).
The principle of negativity bias has not escaped the
attention of thinkers in many disciplines. The principle
hasbeennotedbyanumberofprominenthumanists,in-
cluding Shakespeare, Pushkin, and Schopenhauer (see
quoteslaterinthisarticle).Contemporarypsychologists
havepointedtoformsofnegativitybias in particulardo-
mains, and in general. These contributions will be dis-
cussed in more detail later, but to acknowledge our
predecessors, we note here that, in a political science
journal, Jordan (1965), summarizing results from dif-
ferentstudies,concludedthat“a positiveattitudeorpos-
itive affect does not have an effect on measured
behavior oppositely equivalent to the effect of a nega-
tive attitude or negative affect” (p. 315). Kanouse and
Hanson(1972) identified a negative bias effect with re-
spect to a well-defined range of phenomena. Guido
Peetersandhiscolleagueshaveproducedmanydemon-
strations of negative bias, particularly in the context of
attitude and impression formation (Lewick, Czapinski,
& Peeters, 1992; Peeters, 1971, 1989; Peeters &
Czapinski, 1990).
Negativity bias has been the focus of attention in a
few other lines of thought in psychology. The greater
generalpotencyofnegativeeventsisatthecoreofpros-
pecttheory,asdescribedintheprospectfunction and la-
beled as loss aversion (Kahneman & Tversky, 1979;
Tversky&Kahneman,1991).Taylor(1991) focused on
the negative–positive asymmetry, in many of its mani-
festations, in the framework of demonstrating adaptive
asymmetricalcopingprocessesinvolvedinneutralizing
the greater negative potency. More recently, in a series
of articles, John Cacioppo and his collaborators
(Cacioppo & Bernston, 1994; Cacioppo, Gardner, &
Bernston,1997,1999)notedanegativitybiaseffect in a
number of domains, including three of the four that we
document and organize in this article. Finally, inde-
pendentlyofourwork,and at the same time, a review of
articles emphasizing negativity bias, particularly in the
social interaction and impressions domain, has been
completed (Baumeister, Bratslavsky, Finkenauer, &
Vohs, in press).
Although negativity bias is often striking, it is far
from universal. On the contrary, there is sufficient evi-
dence for a positive bias that an entire book, The
Polyanna Principle (Matlin & Stang, 1978), has amply
documented the wide range of positive biases. These
appear in higher frequency of positive words, positive
experiences, and positive views of the world, and in
other domains. This puts us in the peculiar position of
describingwhatwe believe to be a basictendencyinthe
faceofdocumentedevidencefortheoppositetendency,
as well.
GuidoPeetersandhiscolleagues(e.g.,Lewick et al.,
1992; Peeters, 1971, 1989; Peeters & Czapinski, 1990)
directly addressed this apparent contradiction, which
they described as positive–negative asymmetry. They
treated the evidence for both positive and negative bi-
asesinasophisticatedandbalancedway.Theynotedthe
interesting fact that, because negative events are much
rarer than positive events, it is adaptive to assume the
positive (the most likely occurrence) while being
watchfulforthedangerousnegative.Thus,manyexam-
plesofpositivebiasresultfromthesamebasicfactabout
the world, the dominance of positive experiences, as
doesnegativitybias.Peetersandhiscolleagueshaveset
thestage,appropriately,foracarefulexaminationof the
contextsinwhicheachtypeofbiasappears,andattempt
to bring both phenomena under one conceptual um-
brella. They have pinpointed the issue in noting the
greater frequency, but lesser “urgency,” of positive
events. Organisms must be most efficient at dealing
with the most frequent occurrences, but also the most
important occurrences. Cacioppo and his colleagues
(Cacioppo & Bernston, 1994; Cacioppo, Gardner, &
Bernston, 1997, 1999) also recognized positive and
negative bias; they focused on a negativity bias as a
more rapid recruitment of negativity than positivity
withincreasingstrengthof elicitors and a positivity off-
set defined in terms of a bias to treat relatively neutral
entities as weakly positive.
Our contribution in this article to the prior work is
fivefold.
1. We extend the range of domains in which
negativity bias has been noted.
2. We present a taxonomy of negativity bias phe-
nomena.
3. We clearly distinguish the special and, we be-
lieve, most robust and informative subclass of
negativity bias phenomena: negativity dominance.
4. In particular, we highlight the previously un-
noted area of contagion and contamination, which we
believe is the most robust and informative subclass of
negativity dominance. We argue that contagion and
contamination matters both as a domain in which
negativity bias makes some of its most dramatic ap-
pearances and a possible mechanism that mediates
negativity bias effects in other spheres.
5. We review and extend, in light of the first four
points, the significance, extent, and theoretical ac-
297
NEGATIVITY BIAS
countsofthevarious negativity bias phenomena and at-
tempt to include many of the phenomena of positive
bias under the same theoretical accounts.
We regard our work as being largely complemen-
tary to the simultaneous work of Baumeister et al. (in
press). The two articles differ both in respect to the
structure of their arguments and the range of examples
they proffer in support of the general proposition that
negativity bias is a pervasive and consequential feature
of human existence. However, the main difference be-
tween our works lies elsewhere. Bauimeister et al. ana-
lyzed the phenomenon at hand primarily in terms of
independent, orthogonal influences of bad and good
things on behavior and cognition. Our work, on the
other hand, emphasizes the tendency of the effects of
the negative to dominate (or even utterly overwhelm)
those of the positive when the two are blocked together
to form a single configuration. This echoes the Ges-
talt-like claim that, in a wide range of cases, the “cog-
nitive interaction” of two stimuli, such as those
denoting evaluatively positive or negative personal at-
tributes and brought together to form a novel compos-
ite, cannot be accurately anticipated from prior
knowledge of the values of the two stimuli taken apart
(Rokeach & Rothman, 1965; Royzman, 2000).
Four Aspects of Negativity Bias
In this taxonomic section we propose three or four
types of negativity bias: negative potency,greater
steepness of negative gradients,negativity dominance,
and negative differentiation.
Negative Potency
The principle of negative potency asserts that, given
inverse negative and positive events of equal objective
magnitude, the negative event is subjectively more po-
tent and of higher salience than its positive counterpart.
More generally, the claim is that negative events are
more potent with respect to their objective magnitude
than are positive events. This is described in the pros-
pect function and is at the core of the loss aversion phe-
nomenon (Kahneman & Tversky, 1979; Tversky &
Kahneman, 1991). For example, in the domain of poli-
tics, Bloom and Price (1975) showed that short-term
economic conditions, when they are downturns, reduce
the vote for the party of the incumbent in American
presidentialelections, whereas upturns have virtually no
effect. The endowment effect is perhaps the purest and
most robust instantiation of loss aversion (Kahneman,
Knetsch, & Thaler, 1990); people will demand much
more to give up something they possess (a loss) than
they will pay to obtain the same item (a gain).
Loss aversion has been demonstrated in a number of
domains,but it does not alwaysoccur. The demonstration
of negative potency is unfortunately limited, because it
requires a metric (usually money) to establish the objec-
tive equality of negative and positive events (e.g., losing
and gaining $100, or 10 lb of weight, or 5° in tempera-
ture), or comparison of two contexts for the same mate-
rial entity (as in the case of the endowment effect).
Greater Steepness of
Negative Gradients
Thereisminimalbutprovocativeevidencethatnega-
tive events grow more rapidly in negativity as they are
approached in space or time than do positive events.
There have been only a few empirical attempts to dem-
onstrate this, but we consider this principle potentially
important.Theclearestdemonstrationsanddiscussions
ofthisphenomenonappearintheanimallearning litera-
ture(Brown, 1948; N.E. Miller, 1944;discussed later).
Recently, Cacioppo and colleagues (Cacioppo &
Bernston,1994;Cacioppo, Gardner, & Bernston, 1997,
1999)discussedthisasymmetryandincorporated it into
their computational model of evaluative space. In par-
ticular, they posited steeper negative gradients (which
theydescribedasnegativitybias)alongwithatendency
for a net positive outcome with very weak negative and
positive inputs (which they described as positivity off-
set). In addition to the work of Brown and Miller, they
citedevidencefrom the impressions literature, suggest-
ing that negative events dominate positive events only
when both are strong.
It is possible that steeper negative gradients are sim-
ply a manifestation of negative potency because the
steeper gradient follows from the fact that additional
negativeunits(measuredasstimuli)willproducelarger
psychological effects than additional positive units.
However, in light of the phenomenon of positivity off-
set,thegradienteffectmaybe distinctfromnegativepo-
tencybecause at low levelsnegative inputs donot seem
to be more potent than “equivalent” positive inputs.
Negative potency might result from higher subjective
levels of negative stimuli at all stimulus levels, so that
thetwofunctionsmighthave thesameslopebuttheneg-
ative function might have a higher intercept. However,
the gradient results suggest that this is not the case;
rather,thenegativefunctionsseemtohavealower inter-
cept but a higher slope.
Negativity Dominance
According to the principle of negativity dominance,
the holistic perception and appraisal of integrated neg-
ative and positive events (or objects, individuals,
298
ROZIN & ROYZMAN
hedonic episodes, personality traits, etc.) is more nega-
tive than the algebraic sum of the subjective values of
those individual entities. The entities being summed
algebraically are not stimuli, but evaluations; hence,
negativity dominance occurs after we take into any
possible effect of negative potency and is, in principle,
independent of it. Negativity dominance does not re-
quire, operationally, the use of objectively equated or
objectively measured stimuli. We consider negativity
dominance the most robust and most common exem-
plification of negativity bias, and this article focuses
primarily on this principle. All of the examples offered
at the beginning of this article illustrate negativity
dominance. In the purest condition, negativity domi-
nance holds that the combination of events of equal but
opposite subjective valence will be negative. Thus, if
losing $100 is worse than winning $100 is good, we
have an instance of potency. But if we then find that
losing $100 is as bad as winning $150 is good, and that
losing $100 and winning $150 is negative, then we
have negativity dominance.
Kanouse and Hanson (1972) recognized the partic-
ular importance of negativity dominance. They framed
the power of negative properties in terms of their abil-
ity to interfere with enjoyment of positive aspects, as
when a rancid taste completely ruins the good taste of a
soup. They suggested that “negative components of a
complex object are overweighted only when the good
and the bad are found together in one object, when they
are inseparable” (p. 58). Moreover, as one of us noted
elsewhere (Royzman, 2000), it is precisely when nega-
tiveand positive stimuli are “blended together” to form
anovel gestalt thatone sees oneof the mostremarkable
manifestations of the dominance principle—“negative
overassimilation,”in which aproperty that isevaluated
negatively in its own right “may be judged even more
negatively when … lodged in a positive subject”
(Rokeach & Rothman, 1965, p.130), so that “irrespon-
sible father” could be judged more negatively than “ir-
responsible” (Rokeach & Rothman, 1965) and “loyal
martinet” more negatively than “martinet” per se.
We find it useful to draw a further distinction be-
tween the synchronic (simultaneous) and the
diachronic (successive) manifestations of negativity
dominance.The former concerns the appraisal of nega-
tive and positive components as co-occurring constitu-
ents of a single whole (as in the case of forming a
holistic impression of a person on the basis of a list of
adjectives that describe that person’s negative and pos-
itive traits). Under these conditions, the negative com-
ponent would be disproportionately more influential in
determining the overall appraisal than the positive
components of comparable magnitude. The diachronic
subtype, on the other hand, is revealed in the cancella-
tion of positive by negative events, and vice versa (as
in the case of determining how many lives a murderer
has to save to neutralize one act of murder). The cock-
roach and the purification rites examples cited at the
beginning of this article embody distinctly the
diachronic subtype of negativity dominance.
We believe that instances of negativity dominance
afford us more dramatic and effectual means of
showing that the core of the positive–negative asym-
metry in both physical and moral domains may be the
fact that the corrupting or “devaluing” power of bad
things is greater than the redeeming power of good
things (Royzman & Kumar, 2001). The image that
comes to mind from the physical domain is that of a
single cancerous growth or germ that radiates itself
through and ultimately consumes a perfectly healthy
body. The image that comes to mind from the moral
domain is that of a single vice corrupting and pervert-
ing and bringing the moral downfall of an otherwise
perfectly good person.
Greater Negative Differentiation
Negativity bias manifests itself in the fact that nega-
tive stimuli are generally construed as more elaborate
and differentiated than the corresponding positive
stimuli. This phenomenon of greater negative differen-
tiation represents, in our view, yet another facet of the
general negativity bias principle. The most reliable
finding consistent with this phenomenon is that the vo-
cabulary used to describe the qualities of evaluatively
negative phenomena is far richer and more varied than
that employed to depict those associated with
evaluatively positive stimuli (Peeters, 1971), suggest-
ing that our cognition is perhaps more complex, elabo-
rated, and fine-tuned when it comes to the occurrences
of the former (e.g., Czapinski, 1985). Negative differ-
entiation is distinguished by Peeters and his colleagues
from the other negativity bias effects under the term in-
formational negativity effect, in contrast to the affec-
tive negativity effect, which includes what we call
potency and dominance. Another example of greater
negative differentiation, described later, is the gener-
allygreater number ofnegative than positive emotions.
A Possible Additional Aspect of
Negativity Bias: Less Adaptation
to Negative Events
There are suggestions in the literature, from studies
of pain and stress (see, e.g., Taylor, 1991) and adjust-
ment to major good and bad life-events (Brickman,
Coates,&Janoff-Bulman,1978),thatthereislessadap-
tation to negative than positive situations. We will not
consider this possibility further for two reasons: (a)
thereisnotmuchevidenceto support this claim; and (b)
299
NEGATIVITY BIAS
evenifthere was evidence, it wouldbe subject to the in-
terpretationthatitwasabyproduct of negative potency.
Carefulmatchingofnegative and positive inputs would
havetobecarriedouttoestablishlessadaptationtoneg-
ative events as independent of negative potency.
A Note on Method
The logic of argument for negativity bias is com-
plex,largely because ofthe difficulty ofequating nega-
tive and positive events. At one level, one can compare
subjective reactions to objectively equal negative and
positive events; this depends on some sort of accept-
able scale for events, such as money (losing vs. win-
ning $100) or temperature. A second approach is to
compare combinations of stimuli equated for subjec-
tive intensity, or more generally, showing that the out-
come evaluation of combined stimuli is more negative
than an algebraic sum of subjective (or objective) in-
tensities. A third possibility, depending on the claim, is
to show an interaction effect, such that, for example,
mixed negative and positive stimuli become more neg-
ative the closer one is to them. A fourth way of making
a meaningful comparison is to show that a negative
event pushes some output into negativity, but a posi-
tive event that corresponds in some way to it has no ef-
fect in the positive direction. This avoids the scaling
problem. A fifth strategy is to show that there is no ex-
act positive equivalent of a given negative construct.
For example, Baumeister et al. (in press) argued that
there is no apparent positive equivalent to the event of
psychic trauma and the associated condition of Post
Traumatic Stress Disorder. A sixth approach is to show
that, given two logically related but oppositely
valenced constructs (e.g., pessimism vs. optimism, bad
parenting vs. good parenting), it is the absence or pres-
ence of the negative construct that is the principal de-
terminant of an outcome of interest (e.g., recovery),
with the positive counterpart making little or no mar-
ginal contribution. An example of this strategy at work
is afforded by a recent study by Schulz, Bookwala,
Knapp, Scheier, and Williamson (1996). Drawing on
the prior research, indicating that pessimism and opti-
mism are better viewed as separate factors rather than
bipolar opposites, these authors examined separately
the effects of dispositional pessimism and
dispositional optimism (assessed as responses to four
negatively phrased and four positively phrased
subscales of Scheier & Carver’s, 1985, Life Orienta-
tion Test) on survival among 238 patients with ad-
vanced cancer. Pessimism was a significant inverse
predictor of survival at the 8th month’s follow-up for
the younger age group (30–59). On the other hand, op-
timism was not a significant predictor of survival at
any age. Baumeister et al., who also cited this and re-
lated evidence, made excellent use of this strategy in
their article, illustrating extensively how the absence
of the negative matters more than the presence of the
positive across a variety of domains, including health,
parenting, and relationships. Seventh, less logically
sound but often convincing, are findings of a large dis-
parityin effect betweena negative anda positive event,
as when there is obviously nothing to match the po-
tency on the positive side of a cockroach touching
food. All seven of these lines of evidence can be sup-
ported by experiments, careful observations, and/or
anecdotal reports.
Evidence for Negative Bias
in Different Domains
This article is organized in terms of the domains in
which negativity bias may occur: sensory, memory,
contamination, impressions of persons, moral judg-
ments, and so forth. Superimposed on these domains is
our organization of the ways in which negativity bias is
manifested.Thistaxonomy can be fit withinthePeeters
affective and informational distinction. Some of the ar-
eas that we review have received thorough recent re-
views, in the context of a negative bias, particularly by
Peeters and his colleagues (Lewick, Czapinski, &
Peeters, 1992; Peeters, 1971, 1989; Peeters &
Czapinski,1990), Taylor (1991), Cacioppoand his col-
leagues (Cacioppo & Bernston, 1994; Cacioppo,
Gardner,& Bernston, 1997, 1999), andBaumeister and
his colleagues (Baumeister et al., in press). In those do-
mains, our review will be less thorough and will cite
principally the conclusion of prior reviews.
Physiological Arousal
Taylor (1991) recently reviewed this literature and
concluded that there is generally more physiological
arousal to negative events and that arousal alone is
generally interpreted negatively. She also pointed out
that negative stressors (changes) seem to have more of
an effect on health than “equivalent” positive “stress-
ors.” As she admitted, the evidence on all of these
points is not convincing, because there have been few
direct comparisons, and because it is difficult to equate
positive and negative events for purposes of compari-
son. All of the evidence on physiological arousal is di-
rected at the demonstration of negative potency (as
opposed to negativity dominance, which involves
combinations of negative and positive events). How-
ever, Cacioppo et al. (1999) reviewed evidence from
evoked potentials in humans suggesting disproportion-
ately negative outputs from combinations of negative
and positive inputs (negativity dominance).
300
ROZIN & ROYZMAN
Sensation and Perception
As Schopenhauer (1844/1995) noted more than 100
years ago, the absence of pain, unlike pain, has no dis-
tinct phenomenological presence:
We feel pain, but not painlessness. … We feel the de-
sire as we feel hunger and thirst; but as soon as it has
beensatisfied, it is like the mouthfulof food which has
been taken, and which ceases to exist for our feelings
the moment it is swallowed. (p. 575)
Schopenhauer’s (1844/1995) claim is correct for the
bodyinterior. With the exception ofpositive sensations
arising in muscles (as in massage), the inside of the
body is basically a source of evaluatively negative in-
put. No news is good news, from the point of view of
the body interior. The sensations that arise from the
body interior are essentially painful indications that all
is not well (Rozin, 1999; Troland, 1928). Thus, from
the inside point of view, the “positive” state of normal
functionistheessentiallyneutraldefault.Not only does
one not go to the doctor when ones’ organs feel good,
but one does not even notice it.
However, the body surface, and especially its aper-
tures, represent both pleasure and pain (Rozin, 1999;
Troland,1928). But even on the body surface, thereis a
wider distribution of pain. Pain can be produced any-
where on the body surface, whereas the loci for pleas-
ant sensations are far more circumscribed, even named
for one set of pleasant sensations: the erogenous
zones—there are no labels for “torturogenous zones.”
The“wefeelpain,but not painlessness”dictumfinds
its reformulation in Scitovsky’s (1974) idea of the
phenomenologicalpaleness of comforts, goods that are
dedicatedto preserving a minimal levelof painlessness
andthat,sotospeak,keeppainatbay(e.g.,aircondition-
ers).Scitovskypointedoutthat we gain little from com-
forts, because we adapt to them quickly (see also
Frederick & Loewenstein, 1999). People generally
don’tgetpleasurefromtheir airconditioning,butwould
experienceimmediatediscomfortifitceasedtooperate.
Negativepotency (higher psychological intensity of
negative as opposed to positive events) in sensory sys-
tems is also exemplified and organized by Troland’s
(1928) analysis of sensory systems into nociceptive
(indicating harm), beneceptive (indicating benefit),
and neutroceptive (informational, but innately
affectively neutral). The principal neutroceptive sys-
tems are vision and audition, but proprioception and
pressure and touch are also included. In all of these
systems, it is generally true that high levels of stimula-
tion are aversive, but lower levels are often neutral.
Troland suggested that the nociceptive and
beneceptive systems are reporting on the state of the
organism, whereas the neutroceptive systems are re-
porting on the state of the environment. Pain is the
principal nociceptive system, but others include the
sensations resulting from empty lungs, a full bladder,
or certain types of gastrointestinal upset (producing
nausea). In each of these cases, there is no obvious
beneceptive input that results from normal function.
Troland identified the erotic system as the basic posi-
tive system. Some systems have nociceptive and
beneceptive components, such as taste and smell. In
taste, there are more negative (bitter, sour) than posi-
tive(sweet) subsystems. Trolandnoted that decrease in
rate of increase of activity in a nociceptive system, or
more clearly, cessation of its activity, may lead to posi-
tive sensations. Examples he offered include the plea-
sure of release of discomfort from emptying the
bladder or bowels, or from breathing following un-
pleasant symptoms resulting from oxygen deprivation
from the lungs. Indeed, hunger and thirst can be
thought of as nociceptive systems; the reduction of ei-
thersurely produces positiveaffect, but thesystems are
basically negative: we only appreciate satiation as a
positive sensation at the time that it directly follows a
period of hunger or thirst.
The response to stimuli leading to negative evalua-
tions are generally more distinct and intense than those
leadingto positive evaluations.This is clearin compar-
ing negative and positive facial expressions to tastants
in both humans (Steiner, 1979) and rats (Grill &
Norgren, 1978).
Although there is substantial adaptation to contin-
ued stimulation in most sensory systems, the pain sys-
tem stands as a notable exception. Pain, as an indicator
of something awry, remains an attention-getting input.
Furthermore, there is evidence that people adapt more
to fragrant than pungent smells (Cometto-Muniz &
Cain, 1992).
Attention and Salience
Generally, negative information seems to com-
mand more attention. The intellectual issue raised by
this small literature is the extent to which negative
bias in salience is completely a byproduct of negative
potency (greater psychological impact of negative
events than equivalent positive events), or whether
there are some special negative bias features that op-
erate particularly in the domain of attention. The last
study we review (Pratto & John, 1991) controls for
potency and still reveals a negative bias effect. The
small number of articles that address negative bias in
attention focus on the greater ease of identifying neg-
ative stimuli, the higher speed of locating negative
stimuli in a search task, or the greater masking power
of negative stimuli. The stimuli involved are faces,
words, and social information.
301
NEGATIVITY BIAS
Masking. Ohnesorg (1999) demonstrated that
negative words are more effective as backward masks
than positive words. The results also suggested that on
repetition as masks, the attention capturing capacity
dropsfasterforpositiveasopposedtonegativewords.
Identification. Steiner (1979) studied the judg-
ments by adults of the emotions expressed by infants,
presented on videotape, while the infant was experienc-
ing negative (sour or bitter) and positive (sweet) tastes.
He reported that adults are better at judging the negative
faces. Of course, this suggestive study does not distin-
guish between the salience of the expressions in the in-
fants or the recognition ability of the participants. Simi-
larly, and with the same proviso, H. J. Grill (personal
communication, 1990) reported a greater sense of “ur-
gency” in the Norway rat’s response to negative taste as
opposed to positive taste stimuli.
Search. There is one thorough and excellent
study in this area, dealing with the search for negative
or positive faces in a field (crowd) of other faces
(Hansen & Hansen, 1988). The authors reported an
“angersuperiorityeffect.”Thetask is to scan a “crowd”
of black-and-white faces of the same person and iden-
tify the one face that is discrepant from the others. The
discrepant face can be happy or angry, in a background
of neutral or opposite valence (angry or happy, respec-
tively) faces. Reaction times are much faster for angry
than for happy face targets. Further analysis demon-
stratesthestriking fact that identification ofanger faces
occursat about the sametime for crowds of four or nine
faces, whereas the search for a happy face takes longer
withthelargercrowd.Theseresultssuggestthatthereis
a parallel search for the angry face, a “pop-out” effect,
but not for the happy face. The authors proposed that
there is a possible preattentive parallel search for sig-
nals of direct threat.
Pratto & John (1991) measured reaction times for
naming the colors of words in a Stroop test and found
that times were longer for undesirable than desirable
trait words, suggesting an attention-grabbing power
for negative social information. This effect did not ap-
pear for negative versus positive nontrait words, oc-
curred when the diagnostic base rate for the negative
and positive trait words was controlled, and occurred
when the trait words were balanced for extremity
(eliminating a simple potency interpretation). The au-
thors reasoned that if attention is selectively diverted
by negative traits, then more should be learned about
them in an incidental learning situation. On repeating
the Stroop study, they found that there is greater free
recall of the negative as opposed to positive words af-
ter the trial.
Learning
Inthedomainoflearning,wehavetheopportunityto
introducefindingsfromtheanimalaswellasthehuman
literature. The basic claim is the existence of negative
potencyand is that negativeevents, serving as reinforc-
ers, produce learning that is more rapid and more resis-
tant to extinction than learning based on comparable
positivereinforcers.Thelatterclaim,aboutresistanceto
extinction,hasnotbeentested,toourknowledge.These
predictionsamounttotheclaimthatlearningaboutneg-
ative USs is “prepared,” in the sense defined by
Seligman(1970).Athirdclaimisthatitshouldbe easier
toreverseinnatepreferencesthan innate aversions.This
entire analysis is subject to the problems raised about
comparingnegativeandpositiveintheprevioussection
on methodological issues. The problem is particularly
strong in this domain because much of the data come
fromanimalresearch,wherethereisnoacceptedwayto
equate subjective intensity.
Students of animal learning are generally aware that
learning with negative events (e.g., escape in the oper-
ant framework) is more rapid than learning with posi-
tive reinforcers. Perhaps the most striking case is
traumatic avoidance learning, which occurs in a single
trial, motivated by a single strong electric shock (Solo-
mon & Wynne, 1954). It seems to be generally true, al-
though there is no rigorous test nor systematic
comparison in the literature, that escape and punish-
ment are more effective in producing acquisition and
resistance to extinction than their positive equivalents,
and negative contrast effects may be stronger, on aver-
age, than positive contrast effects.
Taste aversions in animals. Conditioned taste
aversions, in both animals (reviewed in Garcia,
Hankins, & Rusiniak, 1974; Rozin & Kalat, 1971) and
humans (Garb & Stunkard, 1974; Logue, Ophir, &
Strauss, 1981), typically occur in a single trial. Indeed,
conditionedtaste aversions are acquired sorapidly, and
with such a robust effect, that it has been necessary to
reduce the magnitude of the US (often by reducing the
potency of the nausea producing procedure) to demon-
stratea learning curve.Positive learning in the food do-
main rarely occurs with such rapidity; Sclafani &
Nissenbaum(1988) demonstrated themost rapid learn-
ing, sometimes in a single trial, using a particular type
of carbohydrate (polycose), and also fat reinforcers.
However, overall, there is little doubt in the animal lit-
eratureaboutthegreater speed and robustness of condi-
tioned taste aversions, as opposed to preferences.
Zahorik(1979) attempted a direct comparison and con-
firmed this relation, although her study was not able to
accomplish a convincing demonstration that the nega-
302
ROZIN & ROYZMAN
tiveand positive reinforcerswere equated, insome rea-
sonable sense.
The negative bias hypothesis is strongly supported
in studies that attempt to reverse innate preferences or
aversions. The standard one-trial-effective taste aver-
sion paradigm uses sugar or saccharine solutions, for
which there is an innate preference. On the contrary,
extended efforts to reverse innate aversions to bitter or
irritant sensations in animals, over many trials and
months, have generally failed almost completely
(Rozin, Gruss, & Berk, 1979; Warren & Pfaffman,
1959). Exceptions are one study with rats using social
mediation (conspecific consuming an irritant diet),
which led to a modest preference for a mildly irritating
diet (Galef, 1989); one study showing a very gradual
development of a preference for piquant crackers by
captive chimpanzees, in a situation of extended social
interaction with humans (Rozin & Kennel, 1983); and
onecase of a dog that gradually developed a preference
for piquant foods, in a social and home environment
(Rozin & Kennel, 1983). Although humans regularly
and gradually develop preferences for many innately
unpalatablefoods such as chili pepper, ginger, raw gar-
lic, coffee, and alcohol, there are practically no cases
on record of spontaneous development of such prefer-
ences in animals. To the contrary, animals that regu-
larly consume spicy Mexican food (as garbage) in a
rural Mexican setting do not develop a preference for
chili pepper, whereas all the surrounding adults over
the age of about 5 do (Rozin & Kennel, 1983; Rozin &
Schiller, 1980).
Tasteaversions,phobias,andfetishesor passions
in humans. Theusuallyone-trial,robust,taste-aver-
sion phenomenon has been well documented in hu-
mans, primarily by retrospective questionnaire (Garb
&Stunkard, 1974; Logue etal., 1981; Pelchat& Rozin,
1982),and it appears thatnausea is the criticaluncondi-
tioned stimulus that produces the effect (Pelchat &
Rozin, 1982). Conditioned taste aversions have also
been produced in humans under controlled conditions
(e.g., Bernstein, 1978).
Phobias represent yet another area in which there is
strong, retrospective evidence for rapid one-trial nega-
tive learning. Single traumatic incidents with dogs and
other animals, in particular, seem sufficient to produce
strong negative responses (Solomon & Wynne, 1954).
Seligman (1970, 1971) used conditioned taste aver-
sions and phobias as prime examples of what he called
“prepared” learning. However, there is no well-docu-
mented opposite effect of very rapid and robust posi-
tive learning in humans. Humans come to develop
strong likes for many things, including foods, music,
and pets. In all of these, so far as we know, the acquisi-
tion process is much slower than for taste aversions or
phobias. (However, there is no evidence that these
stronglikes are less resistant to extinction than are pho-
bias or conditioned taste aversions.)
Humans may be unique, among animals, in the de-
velopment of strong and enduring likes (Rozin, 1982).
These are abundant and include the robust reversal of
innate aversions, as in the development of strong pref-
erences for chili pepper, coffee, horror movies, tragic
drama, and scary rides at amusement parks
(McCauley, 1998; Rozin, 1990). It is possible that this
perhaps unique human feature is an adaptation to cul-
ture, which requires adherence to, and preferably a de-
sire for, a whole set of cultural values (Rozin, 1982).
However, whatever the reason for this, these strong
positive attachments develop gradually.
The opposite of a taste aversion or a phobia would
be what we call a fetish, or more generally, a passion.
Passionsexist on a grand scalein humans, especially in
the first world, where there is leisure time that would
allow for such activities. Passions, which we will de-
fine as strong liking for things or activities that have no
obvious biological function, become a major part of
one’s life and a major source of pleasure. Passions in-
clude activities such as collecting stamps or bottle or
hub caps, horse or automobile racing, and sports
fanship (Wrzesniewski, Rozin, & Bennett, in press).
These passions challenge any straightforward adaptive
account and are, in a sense, a challenge as well to the
principle of negativity bias. Although there has been
no experimental research in this area (except for sexual
fetishes, see Rachman & Hodgson, 1968), there is no
reason to believe that passions are rapidly acquired, al-
though they often are highly resistant to extinction.
Evaluative conditioning. Otherthan mere expo-
sure,evaluative conditioning is theonly account for the
acquisition of likes and dislikes with both a theoretical
basis and abundant supporting laboratory experiments.
Inevaluativeconditioninginhumans,a positive or neg-
ative US (e.g., an unpleasant or pleasant picture, face,
or taste) is contingently paired with a relatively neutral
CS. After a number of trials, the participant’s evalua-
tion of the CS moves in the direction of the US (Martin
& Levey, 1978). The initial investigators of this type of
conditioning, Levey and Martin (1975), concluded that
“The effect of negative evaluation was clearly stronger
than that for positive evaluation, and this is consistent
withour knowledge of aversive conditioning”(p. 224).
Subsequent studies of evaluative conditioning, most
prominently by a group of Belgian investigators led by
Frank Baeyens (e.g., Baeyens, Crombez, Van den
Bergh, & Eelen, 1988), are consistent with this result,
although no direct tests of the negative bias hypothesis
have been made. The Baeyens group initially used pos-
itive and negative faces as unconditioned stimuli,
303
NEGATIVITY BIAS
roughlybalancedforvalenceby subject ratings. The re-
sults supported slightly larger conditioned effects for
the negative stimuli. More recently, the Baeyens group
has shifted to a different paradigm, based on pairing of
flavors with pleasant or unpleasant tastes (Baeyens,
Eelen, Van den Bergh, & Crombez, 1990). This para-
digm was based on a positive finding for pairing of
sweet tastes with neutral flavors (Zellner, Rozin, Aron,
& Kulish, 1983). However, the Baeyens group found
the positive pairing to produce only marginal effects,
whereas pairing with a negative taste produced more
robust effects (Baeyens et al., 1990).
Both conditioned taste aversions and acquired pho-
bias fit within the paradigm of evaluative conditioning.
Both, as previously mentioned, show a substantial
negativity bias.
Motivation—Gradient Steepness
This minimally investigated area provides the only
direct evidence for one of the four manifestations of
negativity bias, the steeper approach gradients for neg-
ative as opposed to positive stimuli. The basic finding
is that as one approaches a negative entity, in either
time or space, the aversion for that entity or experience
increases more steeply than the increase in attraction
does for approach to a positive entity (N. E. Miller,
1944). Note that this relation does not depend as much
on careful balancing of the negative and positive enti-
ties as do the potency studies because the finding of in-
terest is an interaction effect.
The evidence for this potentially basic feature of an-
imal and human motivation is summarized by N. E.
Miller (1944). The most convincing study was carried
out by Brown (1948). Brown measured the rat’s ten-
dencyto pull toward a reward (food) at the end of an al-
ley, at different points in the alley; he did the same for
the tendency to pull away from a negative event
(shock) at different points in the alley. He reported
steeper negative than positive gradients in terms of dis-
tance from site of feeding or shock. This was con-
firmed at two different levels of motivation (amount of
food deprivation or intensity of shock). In Brown’s de-
sign, temporal and spatial closeness are confounded.
Although it is likely that the gradient phenomenon
holds for time or space alone, this has not yet been
demonstrated.
Anothertype of demonstrationof the gradienteffect
would be to show that for events (or roughly simulta-
neous combinations of events) of mixed positive and
negative characteristics, the net response to these
eventsbecomes more negative the closer one is to them
(in space or time). We have recently (Rozin, Kurzer, &
Royzman, 2000) demonstrated this with human partic-
ipants,who rated theirnet evaluative responseto a neg-
ativeand positive event scheduled to occur on the same
day, from the vantage point of 1 month ahead versus
tomorrow. For tomorrow, the event combination was
rated as more negative.
N. E. Miller (1944) offered an account of the
steeper negative gradient in terms of the source of mo-
tivation. He pointed out that electric shock is an exter-
nal event, such that closeness to its source should
reasonably increase intensity of response. For food,
however, there is an underlying motivation of hunger,
which presumably does not vary much as one ap-
proaches the goal object. In other words, insofar as
negative motivations are more externalized, closeness
should be a more powerful dimension in the negative
case. Miller also noted that the gradient steepness is
separable from the gradient height and referred to stud-
ies (including the work of Brown) that indicated that
overall strength of motivation affects the height but not
the slope of the gradient function.
In addition to N. E. Miller’s (1944) external versus
internal motivation account, there are two possible, not
mutually exclusive, accounts of gradient effects. Ac-
cording to the intensity account, some aspect of the
preevaluative representation of the relevant stimuli
shows a negative enhancement effect. There is at best
suggestive evidence from taste that intensity of bitter
(measured by physiological response) rises more
steeplythan for sweet (Pfaffman, 1960),and it seems to
usquitereasonablethatthiswouldalsoholdtrueforpain
orpleasure from the body surface.Because we can pre-
sumethatgettingcloserintimeor space toastimulusin-
creases the intensity of its representation, this intensity
negative bias effect could account for gradient effects.
However, one could also argue that, given the urgency
of negative inputs, the input might rise to a maximum
very rapidly, so that gradient effects might only occur
over a small range of time or space. There are sugges-
tions of a steeper function relating negative entities to
affectivenegativityinthecontagionliterature.Thephe-
nomenonof dose insensitivity, documentedprincipally
for negative contagion, indicates that very small doses
of contact with negative entities (e.g., germs, contact
with an undesirable other person) produce almost the
maximal effect (e.g., Rozin & Nemeroff, 1990; Rozin,
Markwith, & Nemeroff, 1992).
The negative gradient effect might also result from
an asymmetry later in the processing system. In the
face of equal recruitment of intensity with closeness, it
is possible that the function relating subjective inten-
sity to evaluation is what is more steep for negative
than for positive stimuli. We know of no direct evi-
dence on this relation. In either event, it is important to
realize, as Cacioppo and his colleagues have indicated
(Cacioppo & Bernston, 1994; Cacioppo, Gardner, &
Bernston, 1997, 1999), that although the slope is
steeper for negative events, at low levels, the absolute
304
ROZIN & ROYZMAN
value of positive affect is higher than is negative affect
(“positivity offset”).
Mood
Taylor (1991), on reviewing asymmetrical effects
of negative and positive events, concluded that expec-
tations of negative events are the strongest determi-
nants of mood. Insofar as negative events have been
equated with positive events, she suggested that nega-
tive effects still have a bigger effect on mood.
Memory
Forfineideasvanishfast,Whileallthegrossandfilthy
last. (W. I. Miller, 1997, p. 70 [Strephon and Chloe vv
233–234, Poetical Works, 525])
The evil that men do lives after them; the good is oft
interr’d with their bones. (Shakespeare, quoted in
Greenblatt, 1997, p. 1565)
These quotes notwithstanding, the existence of a
negativity bias (negative potency) in memory is con-
troversial. Baumeister et al. (in press), in their review
of the memory literature, concluded in favor of the ex-
istence of a negative bias in memory, at least for cer-
tain types of information. On the other hand, Taylor
(1991) and Matlin and Stang (1978) argued for greater
potency of positive memories. Our own consideration
of the literature inclines us to support a positivity bias
view in memory.
The widespread operation of negativity bias might
be expected to generalize to selective memory for neg-
ative experiences. This prediction might be strength-
ened by the greater attentional salience of negative
events (reviewed earlier) and abundant evidence for
deeper processing of negative events (reviewed in
Baumeister et al., in press). Although there are some
striking examples of predominance of negative bias in
recall, notably of early childhood memories (Blonskii,
1935/1994; Kreitler & Kreitler, 1968), in our view the
preponderance of evidence suggests a positivity bias in
this domain. The most thorough treatment is in a chap-
ter reviewing this very issue in a book devoted to dem-
onstrating a general positivity bias (Polyanna
Principle) by Matlin and Stang (1978). Well over 100
findings, from various paradigms, show a positivity
bias in a majority of cases. Matlin and Stang consid-
ered three accounts for this: advantages in short-term
memory, selective rehearsal, or compensatory pro-
cesses in long-term memory, and find evidence for
each. Most impressive is their multiple regression
analysis of 14 variables on degree of positivity bias. A
major effect is reported for delay of recall, such that
longer delays lead to more positive bias. Taylor
(1991), in a review, came to a similar conclusion as
Matlin and Stang and focused her account on compen-
satory responses that minimize negative memories,
which occur gradually over time (accounting for the
increasedpositivity bias with delay). Thus,on Taylor’s
reasonable view (consistent with Matlin and Stang),
the major reason for positivity bias is not that negative
events are inherently less memorable, but rather that
they are neutralized over time.
Results on both autobiographical memory and labo-
ratory studies suggest that over modest to long inter-
vals of recall, there is a positivity bias in memory. The
question remains as to whether this bias directly con-
tradicts the principle of negativity bias. There are two
processes that may be at work in these studies to en-
hance the appearance of a positivity bias. There is
abundant evidence that positive events are much more
frequent than negative events; hence, in an autobio-
graphical recall study, one would expect more positive
events. This is a substantial bias, but it has been elimi-
natedas a totalaccount by anumber of studiesin which
participant’s recall was checked against diary records
of positive and negative experiences that they were
asked to keep (Holmes, 1970; reviewed in Matlin &
Stang, 1978).
The second “bias” in recall is the very compensa-
tory processes that are aptly described by Taylor
(1991). Work is done to reduce the salience of negative
events in memory. We are inclined to believe that Tay-
lor identified the major reason for a positivity bias in
recall and that the mechanisms she invokes themselves
testify to the salience of negative events.
Contagion
Thedomainofcontagionofferswhat are perhaps the
moststrikinginstances of negativity bias. Itispartly for
this reason that the striking examples we introduced at
the beginning of this article are about contamination.
Thenegativity bias in contagion is evidenced in the ter-
minology alone: Negative contagion is represented in
the word “contamination” in English. There is no obvi-
ous opposite term for positive contamination (purified
has a much more general meaning and does not imply
minimal contact, as does contamination).
The law of contagion was put forward as one of the
laws of sympathetic magic by three anthropologists
around the turn of the century (Frazer, 1890/1922;
Mauss, 1902/1972; Tylor, 1871/1974). The basic idea
is that when entities contact, “essence” passes between
them and leaves a permanent trace (“once in contact,
always in contact”; reviewed in Rozin & Nemeroff,
1990;Nemeroff & Rozin, 2000). Minimal contact is all
305
NEGATIVITY BIAS
that is necessary to allow for the passage of essence.
The early anthropologists saw contagion as a belief of
“primitive” peoples, but it is now clear that it is univer-
sal (Rozin, Millman, & Nemeroff, 1986). Common ex-
amples for Americans include reluctance to consume
foods briefly contacted by worms or cockroaches or to
wear clothing that had previously been worn by a dis-
liked person.
Negativity bias in the potency domain is intu-
itively clear for Americans and has been demon-
strated in the laboratory (Rozin et al., 1986; Rozin,
Nemeroff, Wane, & Sherrod, 1989). In particular,
disliked people produce a larger contagion effect than
do liked people. In addition, as indicated in the intro-
duction, in the food domain, there is nothing nearly
as potent on the positive side as a cockroach or an
earthworm on the negative side.
Negativity bias in potency is also illustrated in the
Hindu caste system, where contact with lower castes
produces much more contagion than does contact with
higher castes. Among the Hua of New Guinea (Meigs,
1984), perhaps the most contagion sensitive of all cul-
tures studied, there is abundant positive contagion, but
negative contagion remains more powerful.
There is relatively little data on dominance effects
(greater effects of contaminants as opposed to purifi-
ers,over and above their rated potency). We have some
unpublished data (Rozin & Royzman, 2000) that indi-
cates that combinations of negative and positive conta-
gion (e.g., a sweater worn by a disliked and then a liked
person, or vice versa) is rated more negative than the
algebraic sum of the subjective evaluative ratings of
each sweater separately.
Itis notable that contagionisboth a domain ofmani-
festation of negativity bias and a theory for why it oc-
curs.Thatis,ingeneral,negativeeventsmayhavemore
penetrance or contagiousness than positive events.
Asymmetric weighting and racial purity. Cer-
tain practices of designating a racial underclass in rela-
tion to its ancestral roots proffer another compelling
(and politically consequential) example of negativity
bias at its bluntest. In the words of Neil Gotanda (cited
in Lopez, 1996), “The metaphor [for defining a racial
underclass] is one of purity and contamination: White
isunblemishedand pure, so one drop ofancestralBlack
blood renders one Black. Black is a contaminant that
overwhelms white ancestry” (p. 27). Of particular in-
terest, from this perspective, is the notorious “one drop
of blood” rule of racial categorization, having its for-
mal origins in provisions of the Code Noir (the “Negro
Code”)of 1685, designed,in part, tosafeguard the “pu-
rity” of the White race by eliminating “tainted” blood
(the rule enjoyed considerable vogue in certain parts of
postcolonial America, e.g., Alabama, Arkansas). The
rule is that “any known African ancestry renders one
Black” (Lopez, 1996, p. 27). As Lopez pointed out, in
accordance with this rule, “no ‘mixed-race’ applicant
was naturalized as White” (p. 27). There exists no his-
torical evidence for the positive equivalent of a
“one-drop”ordinance—that is, a statute whereby one’s
membership in a racially privileged class would be as-
sured by one’s being in possession of “one drop” of the
racially superior blood (a situation of some bearing
here is the determination, often for purposes of mar-
riage or succession, as to whether a person is of a
“royal” line; certainly, in this case, the royal blood can
represent less than half of the total “blood,” but we
know of nothing like a one-drop rule).
The Nuremburg laws, promulgated by the Nazis in
the1930s to define and persecuteJews, forced a defini-
tion for Jewishness. The one-drop rule here would
have been impossible to enforce; instead, the rule was
that one Jewish grandparent was sufficient for the des-
ignation as Jewish. It is interesting in this regard that in
the affirmative action debate in the United States,
where previously stigmatized and contaminating
groups, especially Blacks, are now given preference,
nothing like a one-drop rule has been instituted. It is
rather general resemblance and associations with
members of the previously stigmatized groups that
makes one eligible for privileged treatments. Thus,
when “Black was bad,” a one-drop rule justified inclu-
sion in the category, but a much more substantial link
is required for Black status now that, in some contexts,
“Black is good.”
Decision Making
Loss aversion, one of the most fundamental and
well-documentedbiases in information processing, is a
quintessential illustration of negativity bias in the form
of potency. The principle of loss aversion, based on the
prospect function, holds that losses are more negative
than corresponding gains are positive (Kahneman &
Tversky, 1979; Tversky & Kahneman, 1991). In its
boldest form, losing $10 is worse than winning $10 is
good. Although we are convinced of the general valid-
ity of loss aversion, and the prospect function that de-
scribes and predicts it, we confess that the
phenomenon is only realizable in some frameworks. In
particular, strict loss and gain of money does not reli-
ably demonstrate loss aversion (unpublished data by
the authors). Perhaps the most robust demonstration of
loss aversion is in the endowment effect (Kahneman et
al., 1990), in which the loss is framed as loss of a “pos-
session”and the gain as acquisition of the same posses-
sion. Under these circumstances, in a number of cases,
the loss of a possession, literally just given at random
to the participant, is valued at somewhere around twice
306
ROZIN & ROYZMAN
the monetary value as the gain of the same possession
by someone who does not currently possess it.
A general limitation of one class of loss aversion
studies is that they require an objective metric of value,
almost always money, against which to measure sub-
jective value. Money is a nicely scaled objective value,
but it may have some special properties that compro-
mise it as a metric. In particular, it may reduce loss
aversioneffects by virtue of its fungibility and promote
a “rational” mode of thinking (Rozin, Grant, Wein-
berg, & Parker, 1999).
The endowment effect allows one route around
money, because it involves loss or gain of the same
entity, and hence one can presume the objective value
is equal in either case. Framing is another route
around the money metric; the same objective transac-
tion can be framed as either a gain or a loss. Thus, it
is reported that people are more inclined to use credit
cards when the alternative is described as a discount
for cash (a gain) rather than a surcharge for use of the
credit card (a loss).
There is a second heuristic that can be seen to be
derivative from loss aversion. This is the principle
that there is risk aversion for gains and risk seeking
for losses. This bias can be derived from the idea that
a certain loss is particularly undesirable (loss aver-
sion) and hence encourages risk taking. Some of the
more robust findings in the study of heuristics and bi-
ases demonstrate the risk seeking for losses phenom-
enon. For example, people are much more likely to
take risks to avoid the certain loss of 400 out of 600
lives than they are to take risks in the face of saving
200 of 600 lives.
There is very little data on negativity dominance
in the loss aversion literature. This is undoubtedly be-
cause the prospect function does not predict
negativity dominance. That is, once one has subjec-
tively equated losses and gains, there is no reason to
predict that the combined outcome would be nega-
tive. Note that, although the bulk of data on loss aver-
sion comes from between-subject designs (which
avoid presentation of the “bald” facts to the partici-
pant), dominance studies require the direct compari-
son. In our experience, asking participants to judge
the net hedonic value of losing and gaining $100, or
losing and finding a possession, leads to surprise and,
for most participants, the obvious response of “zero.”
This can be avoided by using previously equated dif-
ferent entities, such as a food rated +4 and a musical
selection rated –4 on an hedonic scale. Another alter-
native is to simply combine negative and positive
events with previous subjective ratings and show that
the overall hedonic valuation is more negative than
the averaged evaluation of the components. We have
collected some data suggesting such a negative domi-
nance effect, using foods previously contacted in
preparation by, or sweaters worn by, both desirable
and undesirable people (Rozin & Royzman, 2000).
We also gathered informal data suggesting that peo-
pleregard pains asmore real than pleasures of equal in-
tensity. College students (n= 14) were asked to
imagine that they had before them a little red button
that, if pressed, would give 1 min of intense pain to one
individual and (Version A) 10 min of similarly intense
pleasure to another individual or (Version B) 1 min of
similarly intense pleasure to 10 individuals. All of the
participants declined “to purchase” pain with pleasure
under either scenario. When the students were asked to
put some number instead of 10 that would make them
change their mind, the lowest number was 800 for Ver-
sion A. Under Version B, all except one student said
that no number would suffice.
Development
There are no compelling predictions that the
negativity bias hypothesis makes about development,
other than the occurrence of the various manifesta-
tions of negativity bias features demonstrated in
adults. Very few of these have been probed or dem-
onstrated in children.
However, there is a feature of human development
that is supportive of the negative bias position. Bridges
(1932) observed infants in the 0- to 2-year-old range,
ratingbehaviorintermsofthepresenceof various emo-
tions.Theearliestexpressions,inthefirstmonthsoflife,
aregeneralexcitationanddistress,withnoclearpositive
expressions. Anger, disgust, and fear appear around 6
months of age. A positive expression (“delight”) was
apparentat3months,somewhatafterdistressappeared,
butnootherpositiveexpressionoractionsappearedun-
til 12 months, when elation and affection are reported.
Littleinformationisprovided aboutthecriteriaforthese
designations,sothisstudymustbetakenonlyassugges-
tive, but the point is surely worth exploring.
As mentioned earlier, Steiner (1979) reported that,
when adults judge emotions of videotaped infant faces,
they show better performance with responses to bitter
and general aversion than they do with positive faces
(induced by sugar in the mouth).
Impression Formation
It is in the domain of impressions of persons that
negative bias has its longest and fullest history in psy-
chology. The explicit references to the process of
negativity bias (Fiske, 1980; Kanouse & Hanson,
1972; Lewick et al., 1992; Peeters, 1971) have cen-
tered on the impressions phenomena, and negativity
bias theories in psychology also center in this area.
307
NEGATIVITY BIAS
There have been thorough reviews of negative bias in
impressions (Baumeister et al., in press; Fiske, 1980;
Skowronski & Carlston, 1989), so we will limit our
discussion here.
The core and paradigm setting study in this area is
Asch’s (1946) classic study. Asch employed lists of
terms describing people, using a between-subject de-
sign and typically varying only one feature of a de-
scription (presence or absence of a trait, or change in
ordering) between groups. Asch, himself, did not ex-
plore negativity bias, but he did promote the idea that
the total impression, and its valence, was not the alge-
braic sum of its components. This study, both concep-
tually and methodologically, gave birth to an
enormous literature on impressions of persons. With
respectto our theme, almost allof these studies involve
the combination of negative and positive traits and
hence fall under the rubric of negativity dominance.
A robust negative bias in impressions studies has
been consistently demonstrated, dating from Jordan
(1965) and Kanouse and Hanson (1972); the early
work is well reviewed by Fiske (1980). These and
other findings encouraged a variety of theoretical ac-
counts. Kanouse and Hanson introduced the lower fre-
quency of negative events and their consequent greater
extremity (from the positive baseline that results from
their lower frequency). Skowronski and Carlston
(1989) offered a comprehensive review of theories and
proposed a category diagnosticity theory. They em-
phasized the extent to which a characteristic is diag-
nostic of the category in question, which is to say, a
feature of most or all members of the category and as
few nonmembers as possible (a perfect discriminator,
like the placenta for mammals, having a diagnosticity
of 1.0). The claim is that highly diagnostic characteris-
tics will be more heavily weighted in impressions and
that, generally, extreme and negative behaviors are
more diagnostic. Although frequency and
diagnosticitymay often covary,they can be separated.
Frequency and diagnosticity theories both predict a
positive bias in cases where the positive trait is rarer.
Skowronski and Carlston (1992) demonstrated this ef-
fect, showing that dishonesty (more diagnostic than
honesty) dominates honesty, whereas high intelligence
(more diagnostic than stupidity or low intelligence)
dominates stupidity. A person who behaves highly in-
telligently on one occasion, and stupidly on three occa-
sions, is still seen as intelligent.
Fiske (1980) and Wojciszke, Brycz, and Boreknau
(1993) implicated extremity of a trait as another potent
influence and as a moderator of the negativity bias ef-
fect. Negativity bias effects appear much more clearly
when extreme traits, both positive and negative, are in-
cluded as instances. This result harkens back to the
steeper negative gradient phenomenon, referred to in
the taxonomy and discussed later.
More recently, Lupfer, Weeks, and Dupuis (2000)
noted that almost all of the evidence for negativity bias
in impressions comes from situations in which both
negative and positive traits or behaviors are attributed
to the target person. They demonstrated an absence of
negativity bias in comparisons of all positive attributes
versus all negative attributes impressions. This impor-
tant study has particular significance for the taxonomy
that we have presented, because it puts the negativity
bias in impressions phenomena squarely in the domain
ofnegativity dominance, ratherthan negative potency.
Empathy
Empathy is a nominally neutral term. According to
definition, it is a psychological process that “involves
sharing the perceived emotion of another—‘feeling
with’ another” (Eisenberg & Strayer, 1987, p. 5), “an
affective response that is more appropriate to the
other’s situation than to one’s own” (Hoffman, 1987,
p. 53), or “a cognitive awareness and understanding of
the emotions and feelings of another person” (Reber’s
Dictionary of Psychology, 1991, p. 238). However,
empathy is far more commonly used in the negative
sense, as an indication of one’s compassion, sympathy,
or pity. Ironically, the very same authors (e.g.,
Eisenberg, Hoffman) who define empathy with no re-
gard for valence give exclusively negative
instantiations of the phenomenon and experimental
models of empathy-elicitation and development (e.g.,
Batson & Coke, 1981; Hoffman, 1987) analyze it in
predominantly negative contexts.
Thus, there appears to be a glaring disparity be-
tween what Kenny (1963) would consider the “formal
object” (p. 189) of empathy (any emotional experience
of another sentient being) and its particular empirical
referent (another’s negative state, e.g., physical suffer-
ing, grief, fear, disappointment). Expressions of em-
pathic joy are somewhat unusual. Thus, if A tells me,
“I just got engaged,” it would be odd, if not downright
uncouth on my part, to reply, “You have my empathy”
precisely because of the implication that I view A’s en-
gagement as some kind of adversity. On the other
hand, on the purely linguistic level, we find that (a)
there is no term in English and at least a number of
other languages that specifically denotes “feeling well
(with or for) at the sight another’s happiness,” (b) there
are numerous terms that specifically denote a vicarious
affective response to another’s distress (the English
pity, sympathy, compassion; in Russian—sostrodanie
[cosuffering] and soboleznovanie [feeling another’s
pain]), and (c) the supposedly generic terms, like em-
pathy or the Russian sochustvie (feeling what another
feels), are, in practice, synonymous with the terms that
signify our affinity with another’s negative state (see
Royzman & Kumar, 2001).
308
ROZIN & ROYZMAN
The apparent preponderance of negative over
positive empathy is not restricted to the world of hu-
man adults. As Thompson (1987) commented in one
of the few passages in the psychological literature
that deal specifically with the positive–negative em-
pathy asymmetry,
There is reason to believe … that adults as well as young
children are more likely to respond empathetically to sa-
lient expressions of negative emotions in others. The hy-
pothesizedfunctionsofempathicarousalinhumanadap-
tation enlist empathy primarily in response to others’
distresscues.Similarly,fromtheethologicalperspective,
human infants are thought to be highly sensitive to social
cuesthat are relevant to protection from threat, andthis is
more likely to entail distress signals. (p. 139)
Rousseau (1762/1950) was so thoroughly imbued
with the practical aspect of this insight that he made
it one of the (less known) cornerstones of his peda-
gogical system:
First Maxim: It is not in the human heart to put our-
selves in the place of those who are happier than our-
selves,butonlyintheplaceofthosewhoaremosttobe
pitied. From this it follows that to incline a young man
to humanity, instead of making him admire the bril-
liant fate of others you must show him the sad sides of
things and make him fear them. (pp. 184–185)
To reiterate, although the “standard” definitions of
empathy stated at the beginning of this discussion en-
vision an empathic experience as something
unvalenced, in actuality, the empathic reaction ap-
pears to be principally to a narrowly defined set of
highly negative affects—for example, fear, grief, dis-
appointment, and so forth. It seems to be more diffi-
cult to be “contaminated” by, and react
empathetically to, another’s euphoria or pride. The
putative higher contagion of misery than joy seems to
be the reason why literary art is so heavily geared to-
ward the depiction of suffering (because aesthetic
contagion utilizes our more general empathic mecha-
nisms) and prosocial behavior is generally under-
stood as alleviating other people’s distress rather than
maximizing other people’s joy.
Another asymmetry in empathy has to do with the
targets of empathy. Instances of positive empathy are
disproportionately directed at individuals close to the
target person, whereas instances of negative empathy
extend broadly to people all over the world.
There is an adaptive account of the preponderance
of negative empathy. The experience of negative em-
pathy is likely to motivate specific helping behaviors
that, at least for related others, would be beneficial. On
the other hand, there is little in the way of response that
is warranted by the good fortune of others.
Moral Judgments
Whatever the theoretical account (and the fre-
quency or diagnostic account seems quite appropriate
here), the enormous negativity bias in judgments of
character is striking. The literature on morality in psy-
chology (and philosophy; see Kupperman, 1991) is
highly biased to analysis and experimentation with
scenarios involving a single moral judgment or of-
fense. These form, in many ways, the most tractable
basis for analysis and experimentation. However, this
emphasis has slighted the study of moral trajectories,
or character (Kupperman, 1991). It is in character, or
net worth, as it were, that the phenomena of negativity
dominance in morality play out, and it is thus in the
person perception literature, rather than the morality
literature, that we find relevant studies.
Rather extreme immoral acts have an almost in-
delible effect; for many people they are unforgivable.
This fact has been noted widely; we offer here two
literary sources:
So, oft it chances in particular men,
That for some vicious mole of nature in them,
As, in their birth—wherein they are not guilty,
Since nature cannot choose his origin—
By the o’ergrowth of some complexion,
Oft breaking down the pales and forts of reason,
Or by some habit that too much o’er-leavens
The form of plausive manners, that these men,
Carrying, I say, the stamp of one defect,
Being nature’s livery, or fortune’s star,—
Their virtues else—be they as pure as grace,
As infinite as man may undergo—
Shall in the general censure take corruption
From that particular fault: the dram of eale
Doth all the noble substance of a doubt
To his own scandal.
(Shakespeare, quoted in Greenblatt, 1997, Hamlet, Act
I, Scene 4, p. 1682)
Consideralsothenotionof “the tragic flaw,” typically a
single failing in an otherwise admirable character, that
brings about the ruin of the character in classic Greek
drama and later drama as well:
Oh! I can feel it now: nought can soothe
us midst our worldly cares, but the conscience!
But if, through chance, it’s scarred, by but a single
stain, … a single stain … then woe and misery!
As from a deadly sore, the soul then burns, the heart is
drenched in venom, and reprove, as if some pealing
hammer, fills the ears.
One’s sick all over, and the head is whirling,
and bloody lads appear before the eyes … And one
would flee, save one can find no shelter … agony!
He is wretched most whose conscience is unclean.
(Pushkin, 1825/1978, Boris Godunov, p. 219)
309
NEGATIVITY BIAS
Consider also the slip of the tongue, or single ques-
tionable action (re Gary Hart’s admission of one indis-
cretion) that ruins a promising career. With reference
to status in street gangs, Anderson (1984) noted that
“In street culture, especially among young people, re-
spect is viewed as almost an external entity that is
hard-won but easily lost, and so must constantly be
guarded” (p. 82). The fragility of moral purity in the
Hindu system is yet another example—hard to attain
and maintain, easy to lose by a single polluting act.
It seems to us that frequency and diagnostic ac-
counts, although clearly relevant and with some ex-
planatory power, are not able, on their own, to account
for the negativity bias in moral judgment. After all, he-
roic acts are surely as infrequent and diagnostic as im-
moral acts, yet murder is rarely balanced, in the
judgment of people, by a single heroic act. In a recent
survey with a few hundred introductory psychology
undergraduates, we asked how many lives a person
would have to save, each on individual occasions and
each at risk to his or her own life, to be forgiven for the
murder of one person. The median was 25 (Kurzer,
Rozin, & Royzman, 2000).
When we shift the focus from moral incidents to
moral trajectories—that is, to character—we almost
invariably place ourselves in the framework of
negativity dominance because most lives are a mixture
of morally admirable and questionable acts. It is possi-
ble, referring back to Lupfer et al. (2000), that there
would not be a negative bias in the comparison of mor-
ally exemplary and totally immoral characters.
Negativity Dominance and
Cognitive Distortions
One of the fundamental tenets of the cognitive
model of depression (A. T. Beck, 1976) is that indi-
viduals suffering from this disorder exhibit a system-
atic negative bias in the processing of complex
valenced information (i.e., the information that has
both positive and negative characteristics). This ten-
dency leads them to pay inordinate attention to the
negative aspects of a situation, an object, or a person,
while assiduously ignoring or discounting the ob-
ject’s positive aspects or attributes. This pathological
tendency (which expresses itself in a variety of pro-
cessing errors or “cognitive distortions,” including
the disqualifying the positive distortion, the magnifi-
cation or minimization distortion, the mental filter
distortion, and the tunnel vision distortion; J. S. Beck,
1995) bears a striking resemblance to what we dis-
cussed earlier as the synchronic subtype of negativity
dominance, the principle of evaluation positing that
“the holistic perception/appraisal of integrated nega-
tive and positive events (or objects, individuals,
hedonic episodes, personality traits, etc.) is more
negative than the algebraic sum of the subjective va-
lences of those individual entities.” We speculate that
these cognitive distortions represent an abnormally
amplified, persistent, self-regarding version of the
penchant for negativity dominance that is part of a
normal evaluative response. If this conjecture is cor-
rect, one may predict that subclinical individual dif-
ferences in negativity bias will be an index of a
person’s potential for depressive ideation.
Negative Differentiation
There is substantial evidence suggesting that re-
sponses to negative events are more differentiated and
complex. The organism probably has more appraisal to
do on negative events, because the response options
are more varied (fight, flight, slow withdrawal, or
freezing), as opposed to the straightforward approach
response to positive events. The evidence that is rele-
vant (some of it reviewed in Peeters & Czapinski,
1990, and Taylor, 1991), falls into three categories,
which we will consider in turn.
Negativeeventselicitmoreattributionalactivity.
Negative events elicit more causal attribution than
positive events (Bohner, Bless, Schwarz, & Strack,
1988) and are perceived as more complex (docu-
mented and reviewed in Peeters & Czapinksi, 1990).
Across cultures, people seem to seek more explana-
tions for negative than positive events (A. Fiske, per-
sonal communication).
Roese and Olson (1997), following on norm the-
ory (Kahneman & Miller, 1986), demonstrated that
counterfactual thinking, a form of concern about and
replaying a past outcome, is more likely in the con-
text of negative events and negative emotions. People
seem to be particularly motivated to undo past un-
pleasantness. Kahneman and Miller (1986) also noted
that the “affective response to an event is enhanced if
its causes are abnormal” (p. 145). Because negative
events are rarer, they may fall under this account, as
being more abnormal. Baumeister et al. (in press)
considered this feature of negativity bias in much
more detail.
Language: Greater cognitive elaboration of
negative events. The vocabulary used to describe
the qualities of physical pain is far richer and more var-
ied than that employed to depict the qualia associated
with physical pleasure, suggesting that our cognition is
more complex, elaborated, and fine-tuned when it co-
310
ROZIN & ROYZMAN
mes to the occurrences of the former. This is our list of
pain and pleasure descriptors:
Pain (31 words): deep, intense, drilling, boring,
dull, sharp, aching, burning, cutting, pinching,
piercing, tearing, twitching, shooting, raking,
gnawing, itching, stabbing, nipping, sticking,
thrusting, hard, throbbing, (dragging), penetrat-
ing, lingering, fitful, radiating, (breathtaking),
bitter, pricking, (vicious), (monotonous), bit-
ing, incising.
Pleasure (14 words; partly based on a review of
erotic literature): intense, thrilling, sharp, deli-
cious, exquisite, deep, fluttering, lingering, radi-
ating,sumptuous, breathtaking, electrifying,del-
icate, sweet.
The substantially greater (350 msec) reaction time
in matching a negative adjective to the designation
negative, as opposed to a positive adjective to the des-
ignation positive, is most easily interpreted as more
complex or deeper processing of the negative, al-
though there are other accounts, such as simply revers-
ing a natural tendency (Osgood & Hoosain, 1983).
Emotions: Greater number of categories and
response options on the negative side. “All happy
families resemble one another; every unhappy family
is miserable in its own way” (Tolstoy, 1875/1998, p.
1). The elicitors of negative emotions are conceptu-
ally more varied, and there are more response out-
comes that are appropriate for negative elicitors. (On
the contrary, it is probably true that there is more be-
tween-individual variation in what elicits positive af-
fect, as opposed to negative affect; the elicitors of the
latter, although highly varied, tend to be more similar
across people. The idiosyncratic nature of most plea-
sures is precisely what makes their induction so diffi-
cult in a laboratory and makes public policy directed
at improving pleasures somewhat problematic.) The
multiplicity of responses implies a more complex
processing system, at least at the final stages before
output. Because emotions are often interpreted as ac-
tion tendencies, the wider range of potential re-
sponses to negative events links directly to the obser-
vation of a larger number of negative than positive
emotions. None of this escaped the founder of experi-
mental psychology, Wilhelm Wundt (1896), who
wrote the following:
Obviouslylanguagehasproducedamuchgreatervari-
ety of names for unpleasant emotions than for pleasur-
able. In fact, observation renders it probable that
unpleasurable emotions exhibit a greater variety of
typical forms of occurrence, and that their different
forms are really more numerous. (p. 180)
Similarly, Titchener (1908) noted in his introductory
text that “ … language has but few words to express
pleasurable emotions” because of more varied “bodily
manifestations of unpleasant emotions” (p. 243).
Carlson(1966)reviewedemotiontermsin172intro-
ductory texts, published over the period from 1877 to
1960. He recorded 20 pleasant emotion terms, and 30
unpleasant emotion terms, and reported that 74.8% of
the pages in the emotion chapters referred to negative
emotions. Furthermore, when students were asked to
name emotion terms, 35.2% were positive and 61%
were negative. Averill (1980) as well reported fewer
positive emotions from nonverbal indicators and re-
ported that 62% of a large number of adjective emotion
terms were rated as negative. He also reported a weak
tendencyforpositiveemotionalwordstobeusedmore.
There is a larger number of “basic” negative emo-
tions as opposed to “basic” positive emotions, a point
noted by Izard (1971). We present here a list of the ma-
jor taxonomies of emotion (Table 1), all of Western or-
igin except that from the ancient Hindu texts, the
Natyashastra (Hejmadi, 1999). Unlike all other taxon-
omies,the Hindu list has equal numbers of positive and
negative emotions. This difference is surely worth ex-
ploring and may have something to do with the fact
that this is the only taxonomy that comes from a “col-
lectivist” society.
The most straightforward account of the thrust of
the results described is that the greater set of response
options in the negative domain promotes a more differ-
entiated set of emotional expressions because their
contact should signal appropriate action to
conspecifics. On the other hand, Averill (1980) sug-
gestedan interesting additional influence. Emotion im-
plies a sense of diminished responsibility (at least in
English), and this frame discourages classification of
positive acts as produced by emotion. People are rarely
described as overcome by charity.
Language: A negativity bias in lexicalization of
negative events. There are a number of com-
mon-usageone-word terms with negativeconnotations
whose positive opposites, although conceptually con-
ceivable, simply do not have any simple, one-word lin-
guistic representation. Consider, for example, “risk” as
it is commonly used in law or medicine (in the sense of
a future outcome that is both negative and uncertain,
i.e., “bad chance”). Opportunity apparently does not
qualify as an opposite. One would be hard-pressed to
say (at least in English), “he has a 50% opportunity of
walkingagain.”Apparently there is no singleunambig-
uous word for “a chance of some propitious occur-
311
NEGATIVITY BIAS
312
Table 1.Emotion Taxonomies
Source Date Anger Fear Sad Disgust Contempt Guilt Shame Interest Curiosity Surprise Happy Joy Love Other Positive
Natya-sastraaA.D. 200 XXXX X X 4
b
Darwin 1872 XXXXX XX X X X 1
c
W. James 1890 X X X1
d
Woodworth 1938 X X X X X X X
Tomkins 1963 XXXXX XX X X
Izard 1971 XXXXX XXX X X
Ekman & Friesen 1975 XXXX X X X
Plutchik 1980 XXXX X X X 1
e
Shaver et al. 1987 X X X XX
aHejmadi, 2000. bFour additional positive emotions: awe, heroism, peace, humor and amusement. cOne additional positive emotion: pride. dOne additional positive emotion: mirth. eOne additional positive
emotion: acceptance.
rence” either in English or in almost all of the 17 lan-
guages we surveyed, all of which have a synonym for
risk (Rozin, Berman, & Royzman, 2001). The same
“asymmetry” holds for such commonly used terms as
“accident,” “catastrophe,” and “murderer.” “Savior,”
the best opposite for murderer, has strong religious
connotations.Inaddition, the word is normallyused for
a person who saves many lives, whereas murderer is
reasonably and usually applied to the loss of one life
(Rozin et al., 2001).
Language: A positive bias that supports the
negative bias frequency or information account.
Within the impressions literature, there is a docu-
mented positivity bias; this appears for the ability (as
opposed to the moral ) domain. This makes sense in
terms of the frequency or diagnostic approach to posi-
tive and negative biases; immoral behaviors and out-
standing ability-based accomplishments are both rare
events. We now point to a second domain where
positivity bias is evident, a finding also susceptible to
the frequency or diagnostic account. The positive bias
in language has been noted on a number of occasions,
particularly by Osgood and his collaborators (Boucher
& Osgood, 1969; Osgood, 1979; Osgood, May, &
Miron,1975)andinthebookdevotedtopositivebiasor
the “Polyanna Principle” by Matlin and Stang (1978).
In the domain of language, with evidence primarily
but not entirely for English, a positive bias appears in
two ways: (a) positive words (usually adjectives, with
a clear evaluative sign) occur much more frequently
than negative words (reviewed in Matlin & Stang,
1978), and (b) positive adjectives “dominate” their
negative opposite in a number of ways to be described
(reviewed in Matlin & Stang, 1978). We argue that at
least some of this effect derives from the same causes
that have been invoked to explain negativity bias.
The evaluative dimension of words and concepts is
one of their paramount features, as suggested by the
ancientChinese opposition ofyin and yang.This oppo-
sition emerges regularly, as the first factor extracted in
analysis of the attributes of words, confirming
Wundt’s (1896) initial suggestion of evaluation, po-
tency, and activity as the three basic dimensions
(Osgood, 1979). In a study across over 20 languages,
there were about twice as many evaluative qualifiers
commonly used as potency qualifiers (e.g., big) and
more potency than activity (Osgood et al., 1975).
Across languages, good is the most common qualifier,
and the evaluative factor accounts for most of the vari-
ance in word attributes and meanings.
The higher positive frequency effect. Boucher
and Osgood (1969), examining 13 languages, stated and
confirmed what they call the “Polyanna hypothesis”:
“Thereisauniversalhumantendencytouseevaluatively
positive words more frequently than evaluatively nega-
tive words in communication.” Osgood et al. (1975) re-
ported the frequency and breadth (contexts) of use of set
of adjectives across over 20 diverse languages, using a
composite score that increases with both frequency and
breadth.“Good” scores higher than “bad” in all cultures,
and “good” scores highest of all 60 common adjectives
studied, in 8 of 22 languages (Osgood et al., 1975).
Matlin and Stang (1978) reviewed these and many other
studies, including many of their own, and made a very
strong case for more frequent occurrence of positive
words, principally but not entirely in English. However,
theyconcludedthatintermsofnumberofdifferentnega-
tive and positive words, findings are mixed and there is
no clear conclusion.
Positivity dominance. The various word
positivity dominance effects described here are demon-
strated for all or almost all of 17 languages in our recent
study(Rozinetal., 2001). However, each was described,
sometimes for English, sometimes for many languages,
in at least one prior study (and many of the points we
raise were documented by Matlin & Stang, 1978):
1. The unmarked positive. Greenberg (1966)
showed, from many languages, that positive adjectives
are usually unmarked (e.g., happy, pleasant, intelli-
gent)and that negative adjectives can usually be gener-
ated from them by negating the positive attribute (un-
happy, unpleasant, unintelligent). Furthermore, when
negativeadjectivesdoexist in their own right (e.g.,sad,
aversive, stupid), they are often not marked to generate
a positive term (e.g., unsad, unaversive, unstupid).
Matlin and Stang (1978) reviewed many other exam-
ples of this asymmetry. We confirmed the asymmetry
with interviews of native speakers of 17 different lan-
guages (Rozin et al., 2001). This can be accounted for
onthe general principle of evolution of an efficient lan-
guage: More common words would be expected to be
shorter and more basic.
2. The positive term typically defines the dimen-
sion. Results from many languages suggest that typi-
cally, the positive term in a positive negative pair is
the term that is used to define the dimension defined
by the pair (Greenberg, 1966; Matlin & Stang, 1978;
Rozin et al., 2001). Thus, in English, the happy–sad
dimension is called happiness, and the strong–weak
dimension is called strength. The positive term tends
to be used to describe both the positive extreme and
the whole dimension.
3. The marked positive term is more extreme than
the marked negative term. Perhaps as a consequence
of the fact that the positive term usually defines the
313
NEGATIVITY BIAS
dimension, the marked (negated) positive term usu-
ally represents the negative end of the dimension,
whereas the marked (negated) negative term repre-
sents the neutral point. Thus, unhappy means sad,
whereas unsad means neutral. This asymmetry ap-
peared in most of the 17 languages we sampled
(Rozin et al., 2001).
4. Positive precedence in frozen opposite compari-
sons. Freezes are fixed traditional word orders, such as
“cat and mouse” or “now and then.” There are no syn-
tactic or obvious semantic rules that constrain the or-
dering, but there is a strong sense that a particular order
is correct (Cooper & Ross, 1975). Cooper and Ross
heldthat, in such ordersof opposite valencedterms, the
positive term usually comes first, as in more or less,
happy or sad,orwin or lose. The authors proposed that
positive words are easier to process, and that there is
some greater efficiency in leading with them. But they
also reported that the order is often reversed in Hindi.
Our study of 17 languages indicated that in almost all
cases for word pairs and for languages, the positive
term typically leads (Rozin et al., 2001).
Theory
We identify three types of theories of negativity
bias: adaptive or /evolutionary accounts, focusing on
the adaptive value of the principles; developmental
theories, focusing on the ontogeny of the principles;
and mechanistic theories, which try to account for the
instantiation and manifestations of the principle, on-
line. The presence of negativity bias, in various forms,
inboth animals andhumans, encourages thepromotion
of at least some accounts that do not require the media-
tion of language or human culture.
The Adaptive Value of Negativity Bias
There are four theoretical accounts of the adaptive
value of negativity bias. These adaptive accounts are
mutually reinforcing and are all likely to be operative,
and derive from us and from a number of prior authors
(Cacioppo et al., 1994, 1997; Lewick et al., 1992; Tay-
lor, 1991):
1. Negative potency. In the extreme, negative
events are more threatening than are positive events
beneficial.Theclear example here is death,afinal, irre-
versibleevent.Avoiding risks of death must beamatter
of the highest priority in the evolutionary scheme; the
peak of vigilance and investment would well be ori-
ented to escape death. It is true that reproduction is the
final measure of evolutionary success, but there are
usually multiple opportunities to reproduce, and death
terminates these options. Of course, there are cases
where death (as sacrifice to protect more reproduc-
tively capable relatives) contributes to inclusive fit-
ness, especially when the organism in question is no
longercapable of reproduction.Insofar as deathand re-
productive potential shape our decision processes and,
in particular, negativity bias, these considerations sug-
gest that the extent of our loss aversion may be pro-
grammed to vary with age. In particular, because
women’sreproductive capacity declinesmore with age
than does men’s reproductive capacity, gender differ-
encesinlossaversionmightappear at the time of meno-
pause, and there may be general age effects, as well. Of
course, such effects might be moderated, especially in
humans, by the many ways in which older adults can
improvethefitnessoftheirchildrenandgrandchildren.
2. Greater negative informational complexity. In
general, positive entities are things to be approached
and engaged. The link between appraisal and action is
typically straightforward. Negative events, on the
other hand, require a more sophisticated appraisal, be-
cause the options for action are more varied. These in-
clude approach (as in certain forms of threat, as
mediated by the emotion of anger), freezing (some-
times appropriate in fear situations), withdrawal
(sometimes appropriate when resources at hand are not
adequate to deal with a situation), and fleeing (e.g., in
the face of certain fear related threats). It is for this rea-
son (mentioned earlier) that, at least in Western emo-
tion taxonomies, there are more negative emotions
(interpreted here as action tendencies; see Frijda,
1986) than positive emotions.
3. Negative events often develop more rapidly and
require a rapid response. The model, of course, is pred-
ator threat. This is not a time for trial and error.
4. Negative events are more contagious and hence
have more negative potential. The basic model here is
the germ, for which there is not an obvious positive par-
allel.Minimalexposuretogermscanleadtomaximalef-
fects,because germs, unlike mostother negative or posi-
tive entities, are self replicating. We suggest that this
germ feature may be the origin of negative contagion ef-
fects, which has, by a process of preadaptation, spread
through other domains of life (such as morality) just as a
germwould(Rozin,Haidt,McCauley,&Imada,1997).
Developmental Theories
Given the adaptive advantage of particular vigi-
lance with respect to negative events, it is quite reason-
able to suppose that the negative bias is a built-in
predisposition. Its presence in animals lends support to
this idea. Furthermore, the opportunities for gradual
learning to avoid death-threatening events may be
minimal. Menzies and Clarke (1995) argued for innate
314
ROZIN & ROYZMAN
accounts of phobias and phobic predispositions, on the
grounds of the terminal consequences of many nega-
tive events (like falling off a cliff). Innate fear to enti-
ties such as enlarging looming objects suggests that
there is at least some inborn equipment designed to fa-
cilitate exit from, or termination of, threatening situa-
tions. There is little data available, but we expect that
organisms are genetically predisposed toward greater
weighting and attention to negative events. On the
other hand, the organism has ample opportunity to
learn that most events are positive, and that rarer, nega-
tive events, are more threatening and offer more adap-
tive options for response.
Neural Aspects of Negativity Bias
There is now abundant evidence, from studies of
behavior in animals (e.g., Berridge & Grill, 1984)
and humans (e.g., Cacioppo et al., 1994, 1997, 1999;
Diener & Emmons, 1985; Watson, Clark, &
Tellegen, 1988), that the negative and positive affect
systems may operate quite independently at some im-
portant level prior to the generation of responses. As
Cacioppo et al. noted, the constraints of action, often
manifested as approach and withdrawal, may force a
negative and positive summation at the response
level, but there is good evidence that the negative and
positive are separately represented at an earlier stage,
which they call the stage of evaluative categorization.
This evidence includes work on the nervous system.
Ito, Larsen, Smith, and Cacioppo (1998) demon-
strated larger event-related potentials for negative
events in neutral contexts than for equally rare,
equally valenced positive events in the same neutral
contexts. Also, there seems to be an evaluative divi-
sion of labor in the cortices, with the right side of the
brain associated with negative affect (Davidson,
1991). This separation would presumably facilitate
operation of a system that enhanced negative affect,
relative to positive affect. As Cacioppo et al. pointed
out, separate loci for the accumulation of positive or
negative effects allow for separate currency functions
(incremental rise in affect with input) for positive and
negative systems. This arrangement also allows for
the demonstrated somewhat independent variation of
negative and positive affect, which can both be high
or both be low at any given moment, although they
are often reciprocally related.
Mechanistic Theories of
Negativity Bias
The first two theories we present arise principally
from the impressions literature (see discussions by
Fiske, 1980, and Skowronski & Carlston, 1989) and
were briefly discussed in the aforementioned treatment
of impressions.
Frequency and diagnosticity. Organisms are
built to attend to change and to attend to more informa-
tive events. The demonstrated much lower frequency
of negative than positive events makes the negative
events more informative. Hence, this general informa-
tionalbias would work in theservice of negativity bias,
as has been noted by a number of investigators (Fiske,
1980; Kanouse & Hanson, 1972; Lewick et al., 1992;
Peeters, 1971; Peeters & Czapinski, 1990; Skowronski
&Carlston,1989). This position gains supportfrom the
factthat a positivity bias is observed in instances where
positive events (such as manifestations of very high
ability) are rarer (Skowronski & Carlston, 1992).
Range Frequency
As with the informative event formulation, another
general “adaptation” principle leads to a negative bias
in low frequency negative domains. Based on adapta-
tion level theory (Helson, 1964), and its expansion into
range-frequency theory by Parducci (1995), it follows
from the higher frequency of positive events that the
evaluative neutral point moves in the positive direc-
tion. A consequence of that, assuming that the initial
magnitude of positive and negative events is about
equal, is that negative events become more potent be-
cause they are now further from the acquired neutral
(formerly positive) baseline than are positive events.
Of course, even if negative events are originally more
potent, this process would act to further enhance rela-
tive negative potency. Along with the informational
value view just discussed, this position would predict a
positive bias in domains where there is a higher fre-
quency of negative events, such as performance on dif-
ficult tasks or gambling.
Contagion
We have commented on the high negative conta-
gious potential of negative events, perhaps originating
from reactions to germs. This fact about the world
could be mirrored in the way the organism and brain
process negative events. Negative events may inher-
ently (or conceivably by acquisition) be more conta-
gious, generalize more to neighboring domains, and be
more resistant to elimination. There is clear evidence
from the contagion literature (Nemeroff & Rozin,
2000; Rozin & Nemeroff, 1990; Rozin et al., 1989) in-
dicating that negative entities transfer properties by
315
NEGATIVITY BIAS
contactmuch more thando positive entities.Contagion
is particularly appropriate as an account for situations
in which both positive and negative factors are pres-
ent—that is, for cases of negativity dominance. We
note that both the most robust and most common in-
stances of negativity bias, including in the impressions
literature (e.g., Lupfer et al., 2000), involve such nega-
tive–positive combinations.
Purity and Perfection
Perfection and purity instantiate actually sought
goals of humans in many domains. Although we typi-
cally settle for something that is less than perfect, we
often judge things in terms of their distance from per-
fection and sometimes dichotomize into the tiny per-
fect, and enormous imperfect, categories. This is
clearlyexpressed in various religious traditionsthat fo-
cus on purity. The human penchant for certainty may
be another manifestation of this principle; people will
pay much more to eliminate a risk than to reduce the
risk, when the actual reduction in risk is the same in
both cases (as in reducing a risk from 1% to .5% vs.
.5% to 0%; Baron, Gowda, & Kunreuther, 1993). It is
also common knowledge that among most collectors,
“perfect” (unused, unsullied in any way) items demand
by far the highest prices.
Tolstoy (1875/1998) began Anna Karenina with the
statement, “All happy families resemble one another;
every unhappy family is miserable in its own way” (p.
1). One meaning of this trenchant quote is that it is very
hard to achieve perfection, and there are many ways to
fail to achieve it. Blemishes can spoil kosher meat, or
beautiful faces, or gems. Another statement of the
same idea is in the proverb, “a chain is as strong as its
weakest link.” If perfection is a sought-after goal, then
we are very vulnerable to loss of this desirable state in
myriads of different ways. The perfect standard primes
negativity bias.
Theory and the Negativity Bias
Taxonomy
The taxonomy of negativity biases we present pro-
vides a structure for the generation and evaluation of
theories. Particular theories may be more appropriate
for particular types of negativity bias. Thus, contagion
is particularly appropriate for negativity dominance
and negative potency and has little to say about differ-
entiation. On the other hand, the asymmetry of death is
most related to potency accounts, and the multiple op-
tions for response to negative events applies uniquely
to negative differentiation. The frequency and
diagnosticity accounts may relate to both potency and
dominance, as perhaps do the impressions phenomena
that gave birth to them.
Negative potency, which has been implicitly the
predominant theory prior to this article, may actually
have two instantiations, as suggested by the work of
Cacioppo et al. (1994, 1997, 1999). The enhancement
of negativity may occur at both the level of integration
of negative and positive representations, in preparation
for response, and at some prior level. Furthermore, the
definition of potency may have two instantiations, as
suggested by some of the empirical research, espe-
cially on impressions. In the narrow form, as described
in loss aversion, potency means that objectively equal
oppositely valenced stimuli are subjectively unequal,
with the negative stronger. However, it also appears
that even when there is subjective equality, negative
stimuli still seem more potent. The question is whether
this residual negative potency is only manifested in
combination and hence falls under negativity domi-
nance, or might operate in isolation.
Conclusion
We do not offer a general, simple theory of
negativity bias. We believe that more than one phe-
nomenon is involved and doubt that there is one theory
toaccount for all(in accord with Taylor, 1991). Rather,
atthe adaptive, developmental, and mechanistic levels,
we presume there are multiple contributions to the
bias; negativity bias is both multidetermined and
overdetermined.However, it is far from universal. Fur-
thermore, as Taylor noted, this very bias may be the
cause of a range of adaptations to modulate the re-
sponse to strong negative effects after they occur. This
couldbe one accountof the apparent lack of greater po-
tency for negative memories.
We believe that organisms have evolved to deal
with both the most frequent and the most important
events in their lives; when frequency and importance
are negatively correlated, as we believe is generally the
casefor negative and positive events,there are likely to
be complexities in systems that deal with appraisal or
response to negative and positive events. This is ex-
actly what we see. The diagnosticity account, in its
fully elaborated form (e.g., Skowronski & Carlston,
1989), is adequate to explain some of the occurrences
of positivity and negativity biases. The fact that it pre-
dicts the inversion from the moral to the ability domain
testifies to the potency of frequency and diagnosticity.
On the other hand, in our view, this type of account is
not helpful in certain domains, such as those that in-
volve negative differentiation or those that invoke con-
tagion. We are inclined to believe that it will be
necessary to invoke both purity as an important human
idealand greater contagiousnessof negative entities, to
316
ROZIN & ROYZMAN
account for some of the instances of negativity bias.
We conclude that what is required is an analysis in the
framework established by Peeters and his colleagues
(Peeters, 1971, 1989; Peeters & Czapinksi, 1990) and
more recently by Cacioppo and his colleagues
(Cacioppo et al., 1994, 1997, 1999), an interplay of
positive and negative biases.
One of the major tasks of future research in this area
will be to designate those domains in which negativity
bias and positivity bias are manifested. We want to
leave the reader with the sense that there really is a
negativity bias, a meaningful and adaptive one, in
much of human and animal cognition and behavior.
There is still some taxonomic work to do, and a great
deal of analysis. In our hope of making a better world,
it should be important to understand and perhaps inter-
vene in some manifestations of negativity bias.
References
Anderson, E. (1984). The code of the streets. Atlantic Monthly,5,
81–94.
Asch, S. (1946). Forming impressions of personality. Journal of Ab-
normal and Social Psychology,41, 258–290.
Averill, J. R. (1980). On the paucity of positive emotions. In K. R.
Blankstein, P. Pliner, & J. Polivy (Eds.), Advances in the study
of communication and affect, Vol. 6 (pp. 7–45). New York: Ple-
num.
Baeyens, F., Crombez, G., Van den Bergh, O., & Eelen, P. (1988).
Onceincontactalways in contact: Evaluativeconditioningisre-
sistant to extinction. Advances in Behavior Research and Ther-
apy,10, 179–199.
Baeyens, F., Eelen, P., Van den Bergh, O., & Crombez, G. (1990).
Flavor–flavor and color–flavor conditioning in humans.
Learning and Motivation, 21, 434–455.
Baron, J., Gowda, R., & Kunreuther, H. (1993). Attitudes toward
managinghazardouswaste:Whatshould be cleaned upandwho
should pay for it? Risk Analysis,13, 183–192.
Batson,D.,&Coke,J.S.(1981).Empathy:A source of altruistic mo-
tivation for helping. In J. P. Rushton & R. M. Sorrentino (Eds.),
Altruism and helping behavior: Social, personality, and devel-
opmental perspectives (pp. 167–211). Hillsdale, NJ: Lawrence
Erlbaum Associates, Inc.
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (in
press). Bad is stronger. Review of General Psychology.
Beck,A.T.(1976).Cognitivetherapyandemotionaldisorders.New
York: International Universities Press.
Beck,J.S.(1995).Cognitive therapy: Basicsandbeyond.NewYork:
Guilford.
Bernstein, I. L. (1978). Learned taste aversions in children receiving
chemotherapy. Science,200, 1302–1303.
Berridge, K. C., & Grill, H. J. (1984). Isohedonic tastes support a
two-dimensional hypothesis of palatability. Appetite,5,
221–231.
Blonskii,P.P.(1994).A critique oftheFreudiantheoryof child sexu-
ality. In V. M. Leibin (Ed.), Zigmund Freud: Psychoanalysis
and the Russian thought (pp. 362–372). Moscow: Republic.
Bloom,H.S.,&Price,H.D.(1975).Voterresponseto short-run eco-
nomic conditions: The asymmetric effect of prosperity and re-
cession. The American Political Science Review,69,
1240–1254.
Bohner, G., Bless, H., Schwarz, N., & Strack, F. (1988). What trig-
gers causal attributions? The impact of valence and subjective
probability. European Journal of Social Psychology,18,
335–345.
Boucher,J.,&Osgood,C.E.(1969).The polyanna hypothesis. Jour-
nal of Verbal Learning and Verbal Behavior,8, 1–8.
Brickman, P., Coates, D., & Janoff-Bulman, R. (1978). Lottery win-
nersandaccidentvictims:Ishappinessrelative?Journalof Per-
sonality and Social Psychology,36, 917–927.
Bridges, K. M. B. (1932). Emotional development in early infancy.
Child Development,3, 324–341.
Brown, J. S. (1948). Gradients of approach and avoidance responses
andtheirrelationtolevel of motivation.JournalofComparative
and Physiological Psychology, 41, 450–465.
Cacioppo,J.T.,&Bernston,G. G. (1994). Relationship betweenatti-
tudes and evaluative space: A critical review, with emphasis on
the separability of positive and negative substrates. Psychologi-
cal Bulletin,115, 401–423.
Cacioppo, J. T., Gardner, W. L., & Bernston, G. G. (1997). Beyond
bipolar conceptualizations and measures: The case of attitudes
and evaluative space. Personality and Social Psychology Re-
view,1, 3–25.
Cacioppo,J.T.,Gardner,W.L., & Berntson, G. G.(1999).Theaffect
system has parallel and integrative processing components:
Form follows function. Journal of Personality and Social Psy-
chology,76, 839–855.
Carlson, E. R. (1966). The affective tone of psychology. Journal of
General Psychology,75, 65–78.
Cometto-Muniz,J.E.,&Cain, W. S. (1992).Sensoryirritation:Rela-
tion to indoor air pollution. In W. G. Tucker, B. P. Leaderer, L.
Molhave, & W. S. Cain (Eds.), Sources of indoor air contami-
nants: Characterizing emissions and health impacts (pp.
137–151). New York Academy of Sciences.
Cooper, W. E., & Ross, J. R. (1975). World order. Unpublished
manuscript.
Czapinski, J. (1985). Negativity bias in psychology: An analysis of
Polish publications. Polish Psychological Bulletin, 16, 27–44.
Davidson, R. J. (1991). Cerebral asymmetry and affective disorders:
A developmental perspective. In D. Cicchetti & S. L. Toth
(Eds.), Internalizing and externalizing expressions of dysfunc-
tion: Rochester symposium on developmental psychopathology
(Vol. 2, pp. 123–154). Hillsdale, NJ: Lawrence Erlbaum Asso-
ciates, Inc.
Diener, E., & Emmons, R. A. (1985). The independence of positive
and negative affect. Journal of Personality and Social Psychol-
ogy,47, 1105–1117.
Eisenberg,N.,&Strayer,J.(1987).Criticalissuesinthestudy of em-
pathy. In N. Eisenberg & J. Strayer (Eds.), Empathy and its de-
velopment(pp.3–13).New York: CambridgeUniversityPress.
Ekman,P.,&Friesen, W. V. (1975).Unmaskingtheface. Englewood
Cliffs, NJ: Prentice Hall.
Fiske, S. T. (1980). Attention and weight in person perception: The
impact of negative and extreme behavior. Journal of Personal-
ity & Social Psychology,38, 889–906.
Frazer, J. G. (1922). The golden bough: A study in magic and reli-
gion. New York: Macmillan. (Original work published 1890)
Frederick, S., & Loewenstein, G. (1999). Hedonic adaptation. In D.
Kahneman, E. Diener, & N. Schwarz (Eds.), Well being. The
foundations of hedonic psychology (pp. 302–329). New York:
Russell Sage Foundation.
Frijda,N.H.(1986).Theemotions.Cambridge,England:Cambridge
University Press.
Galef, B. G., Jr. (1989). Enduring social enhancement of rats’ prefer-
ences for the palatable and the piquant. Appetite,13, 81–92.
Garb, J., & Stunkard, A. J. (1974). Taste aversions in man. American
Journal of Psychiatry,131, 1204–1207.
317
NEGATIVITY BIAS
Garcia, J., Hankins, W. G., & Rusiniak, K. W. (1974). Behavioral
regulation of the milieu interne in man and rat. Science,185,
824–831.
Greenberg, J. H. (1966). Language universals. In T. A. Sebeok (Ed.),
Current trends in linguistics; III. Theoretical foundations (pp.
61–112). The Hague, Netherlands: Mouton.
Greenblatt, S. (Ed). (1997). Norton Shakespeare. New York: W. W.
Norton.
Grill, H. J., & Norgren, R. (1978). The taste reactivity test: I. Oro-fa-
cialresponsestogustatorystimuli in neurologically normalrats.
Brain Research,143, 263–279.
Hansen,C.H.,& Hansen, R.D.(1988).Findingthe face inthecrowd:
An anger superiority effect. Journal of Personality & Social
Psychology,54, 917–924.
Hejmadi, A. (1999). Classical Hindu Indian emotion theory: Impli-
cations for interdisciplinary and cross-cultural research.
Manuscript submitted for publication.
Helson, H. (1964). Adaptation-level theory: An experimental and
systematicapproachtobehavior. New York:HarperandRow.
Hoffman, M. L. (1987). The contribution of empathy to justice and
moral judgment. In N. Eisenberg & J. Strayer (Eds.), Empathy
and its development (pp. 47–80). New York: Cambridge Uni-
versity Press.
Holmes, D. S. (1970). Differential changes in affective intensity and
the forgetting of unpleasant personal experiences. Journal of
Personality and Social Psychology,15, 234–239.
Ito,T.A.,Larsen,J. T., Smith, N. K.,&Cacioppo,J.T. (1998). Nega-
tive information weighs more heavily on the brain: The
negativitybiasinevaluativecategorizations. Journal of Person-
ality & Social Psychology,75, 887–900.
Izard, C. E. (1971). The face of emotion. New York: Appleton-Cen-
tury-Crofts.
James, W. (1890). The principles of psychology. New York: Holt.
Jordan,N.(1965).Theasymmetry of likinganddisliking.Aphenom-
enon meriting further reflection and research. Public Opinion
Quarterly, 29, 315–322.
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental
testsoftheendowmenteffectandthe Coase theorem. Journal of
Political Economy,98, 1325–1348.
Kahneman,D.,&Miller,D.(1986).Normtheory.PsychologicalRe-
view,93, 136–153.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis
of decisions under risk. Econometrica,47, 263–291.
Kanouse,D.E.,&Hanson,L. (1972). Negativity inevaluations.InE.
E. Jones, D. E. Kanouse, S. Valins, H. H. Kelley, R. E. Nisbett,
& B. Weiner (Eds.), Attribution: Perceiving the causes of be-
havior (pp. 47–62). Morristown, NJ: General Learning Press.
Kenny, A. (1963). Action, emotion, and will. London: Routledge &
Kegan Paul.
Kreitler, H., & Kreitler, S. (1968). Unhappy memories of the “happy
past”: Studies in cognitive dissonance. British Journal of Psy-
chology,59, 157–166.
Kupperman, J. J. (1991). Character. New York: Oxford University
Press.
Kurzer, N. C., Rozin, P., & Royzman, E. (2000). [Individual differ-
ences in negativity bias]. Unpublished raw data.
Levey, A. B., & Martin, I. (1975). Classical conditioning of human
“evaluative” responses. Behaviour Research and Therapy, 13,
221–226.
Lewick, M., Czapinski, J., & Peeters, G. (1992). Positive–negative
asymmetryor“Whentheheartneedsa reason.” European Jour-
nal of Social Psychology,22, 425–434.
Logue, A. W., Ophir, I., & Strauss, K. E. (1981). The acquisition of
taste aversions in humans. Behavior Research and Therapy,19,
319–333.
Lopez, I. F. H. (1996). White by law: The legal construction of race.
New York: New York University Press.
Lupfer,M.B.,Weeks,M.,&Dupuis,S. (2000). How pervasive is the
negativity bias in judgments based on character appraisal? Per-
sonality and Social Psychology Bulletin,26, 1353–1366.
Martin,I.,&Levey,A. B. (1978). Evaluativeconditioning.Advances
in Behavior Research and Therapy, 1, 57–102.
Matlin, M. W., & Stang, D. J. (1978). The polyanna principle. Selec-
tivity in language, memory, and thought. Cambridge, MA:
Schenkman.
Mauss,M.(1972).Ageneraltheoryofmagic(R. Brain, Trans.). New
York: Norton. (Original work published 1902)
McCauley, C. R. (1998). When screen violence is not attractive. In J.
Goldstein(Ed.),Whywewatch.Theattractionsofviolententer-
tainment (pp. 144–162). New York: Oxford University Press.
Meigs, A. S. (1984). Food, sex, and pollution: A New Guinea reli-
gion. New Brunswick, NJ: Rutgers University Press.
Menzies, R. G., & Clarke, J. C. (1995). The etiology of phobias: A
nonassociativeaccount.ClinicalPsychologyReview,15, 23–48.
Miller, N. E. (1944). Experimental studies of conflict. In J. McV.
Hunt (Ed.). Personality and the behavior disorders (pp.
431–465). New York: Ronald.
Miller, W. I. (1997). The anatomy of disgust. Cambridge, MA: Har-
vard University Press.
Nemeroff, C., & Rozin, P. (2000). The makings of the magical mind.
In K. S. Rosengren, C. N. Johnson, & P. L. Harris (Eds.), Imag-
ining the impossible: Magical, scientific, and religious thinking
inchildren(pp.1–34). New York:CambridgeUniversityPress.
Oesterreich, T. K. (1974). Possession: Demonical and other.
Secaucus, NJ: The Citadel Press.
Osgood,C.E. (1979).Fromyangand yintoandorbutincross cultural
perspective. International Journal of Psychology, 14, 1–35.
Osgood, C. E., & Hoosain, R. (1983). Polyanna II: Two types of
negativity. Journal of Psychology,113, 151–160.
Osgood, C. E., May, W. H., & Miron, M. S. (1975). Cross-cultural
universals of affective meaning. Urbana: University of Illinois
Press.
Parducci,A.(1995).Happiness,pleasure, and judgment. Thecontex-
tual theory and its applications. Mahwah, NJ: Lawrence
Erlbaum Associates, Inc.
Peeters, G. (1971). The positive–negative asymmetry: On cognitive
consistency and positivity bias. European Journal of Social
Psychology,1, 455–474.
Peeters, G. (1989). Evaluative inference in social cognition: The role
ofdirectversusindirectevaluationand positive–negative asym-
metry. European Journal of Social Psychology,21, 131–146.
Peeters, G., & Czapinski, J. (1990). Positive–negative asymmetry in
evaluations: The distinction between affective and informa-
tional effects. In W. Stroebe & M. Hewstone (Eds.), European
review of social psychology (Vol. 1, pp. 33–60). New York:
Wiley.
Pelchat, M. L., & Rozin, P. (1982). The special role of nausea in the
acquisition of food dislikes by humans. Appetite,3, 341–351.
Pfaffman, C. (1960). The pleasures of sensation. Psychological Re-
view,67, 253–268.
Plutchik, R. (1980). Emotion: A psychoevolutionary synthesis. New
York: Harper & Row.
Pratto, L., & John, O. P. (1991). Automatic vigilance: The atten-
tion-grabbing power of negative social information. Journal of
Personality and Social Psychology,81, 380–391.
Pushkin, A. S. (1978). Boris Godunov. In M. M. Alekseev (Ed.), Se-
lected works in two volumes: Volume 2 (p. 184). Moscow: Cre-
ative Fiction. (Original work published 1825)
Rachman, S., & Hodgson, R. J. (1968). Experimentally-induced
“sexual fetishism”: Replication and development. Psychologi-
cal Record,18, 25–27.
318
ROZIN & ROYZMAN
Reber, A. S. (1991). The Penguin dictionary of psychology. New
York: Penguin.
Roese, N. J., & Olson, J. M. (1997). Counterfactual thinking: The in-
tersection of affect and function. Advances in Experimental So-
cial Psychology,29, 1–59.
Rokeach, M., & Rothman, G. (1965). The principle of belief congru-
ence and the congruity principle as models of cognitive interac-
tion. Psychological Review, 72, 128–142.
Rousseau, J. J. (1950). Emile. London: J. M. Dent & Sons LTD.
(Original work published 1762)
Royzman,E.B.(2000).Are experiments possible? Thelimitationsof
aposterioricontrolinexperimental behavior analysis. Theory&
Psychology, 10, 171–196.
Royzman, E. B., & Kumar, R. (2001). On the relative preponderance
of empathic sorrow and its relation to commonsense morality.
New Ideas in Psychology, 19, 131–144.
Rozin, P. (1982). Human food selection: The interaction of biology,
culture and individual experience. In L. M. Barker (Ed.), The
psychobiology of human food selection (pp. 225–254). Bridge-
port, CT: AVI.
Rozin, P. (1990). Getting to like the burn of chili pepper: Biological,
psychological and cultural perspectives. In B. G. Green, J. R.
Mason,&M.R.Kare (Eds.), Chemicalsenses,Volume2:Irrita-
tion (pp. 231–269). New York: Marcel Dekker.
Rozin, P. (1999). Preadaptation and the puzzles and properties of
pleasure. In D. Kahneman, E. Diener, & N. Schwarz (Eds.),
Well-being. The foundations of hedonic psychology (pp.
109–133). New York: Russell Sage Foundation.
Rozin, P., Berman, L., & Royzman, E. (2001). Positivity and
negativity bias in language: Evidence from 17 languages.
Manuscript submitted for publication.
Rozin,P.,Grant,H., Weinberg, S.,&Parker,S.(1999). “Headversus
Heart”:Effectofmonetaryframeson expression of sympathetic
magical concerns. Manuscript submitted for publication.
Rozin, P., Gruss, L., & Berk, G. (1979). The reversal of innate aver-
sions: Attempts to induce a preference for chili peppers in rats.
Journal of Comparative and Physiological Psychology,93,
1001–1014.
Rozin, P., Haidt, J., McCauley, C. R., & Imada, S. (1997). The cul-
tural evolution of disgust. In H. M. Macbeth (Ed.), Food prefer-
ences and taste: Continuity and change (pp. 65–82). Oxford,
England: Berghahn.
Rozin, P., & Kalat, J. W. (1971). Specific hungers and poison avoid-
ance as adaptive specializations of learning. Psychological Re-
view,78, 459–486.
Rozin, P., & Kennel, K. (1983). Acquired preferences for piquant
foods by chimpanzees. Appetite,4, 69–77.
Rozin,P.,Kurzer,N.,&Royzman,E. (2000). [Unpublished observa-
tions].
Rozin, P., Markwith, M., & Nemeroff, C. (1992). Magical contagion
beliefsandfearofAIDS.JournalofAppliedSocial Psychology,
22, 1081–1092.
Rozin,P.,Millman,L.,&Nemeroff,C. (1986). Operation of the laws
of sympathetic magic in disgust and other domains. Journal of
Personality and Social Psychology,50, 703–712.
Rozin, P., & Nemeroff, C. J. (1990). The laws of sympathetic magic:
A psychological analysis of similarity and contagion. In J.
Stigler,G.Herdt,&R.A.Shweder (Eds.), Cultural psychology:
Essays on comparative human development (pp. 205–232).
Cambridge, England: Cambridge University Press.
Rozin, P., Nemeroff, C., Wane, M., & Sherrod, A. (1989). Operation
of the sympathetic magical law of contagion in interpersonal at-
titudes among Americans. Bulletin of the Psychonomic Society,
27, 367–370.
Rozin, P., & Royzman, E. (2000). [Negativity dominance]. Unpub-
lished raw data.
Rozin,P.,&Schiller,D.(1980).Thenatureandacquisitionofapref-
erence for chili pepper by humans. Motivation and Emotion,4,
77–101.
Scheier,M.F.,&Carver, C. S. (1985).Optimism,coping,andhealth:
Assessment and implications of generalized outcome expectan-
cies. Health Psychology,4, 219–247.
Schopenhauer, A. (1995). The world as will and representation (Vol.
II) (E. F. S. Payne, Trans.). New York: Dover. (Original work
published 1844)
Schulz,R.,Bookwala,J., Knapp, J.E.,Scheier,M.,& Williamson, G.
M. (1996). Pessimism, age, and cancer mortality. Psychology
and Aging,11, 304–309.
Scitovsky, T. (1974). The joyless economy. New York: Oxford Uni-
versity Press.
Sclafani,A.,&Nissenbaum,J.W.(1988).Robustconditionedflavor
preferences produced by intragastric starch infusions in rats.
American Journal of Physiology, 255, R672–R675.
Seligman, M. E. P. (1970). On the generality of the laws of learning.
Psychological Review,77, 406–418.
Seligman, M. E. P. (1971). Phobias and preparedness. Behavior
Therapy,2, 307–320.
Shakespeare, W. (1906). Hamlet. In W. Shakespeare (Ed.), Trag-
edies(pp.482–574).London:J. M. Dent&SonsLTD.(Original
work published 1602)
Shaver,P.,Schwartz,J., Kirson, D.,&O’Connor,C. (1987). Emotion
knowledge: Further exploration of a prototype approach. Jour-
nal of Personality and Social Psychology,52, 1061–1086.
Skowronski,J.J.,&Carlston,D.E.(1989).Negativityandextremity
biases in impression formation: A review of explanations. Psy-
chological Bulletin,105, 131–142.
Skowronski, J. J., & Carlston, D. E. (1992). Caught in the act: When
impressions based on highly diagnostic behaviours are resistant
to contradiction. European Journal of Social Psychology,22,
435–452.
Solomon, R. L., & Wynne, L. C. (1954). Traumatic avoidance learn-
ing: The principles of anxiety conservation and partial irrevers-
ibility. Psychological Review,61, 353–385.
Steiner, J. E. (1979). Human facial expressions in response to taste
andsmellstimulation.InH.W.Reese&L. P. Lipsitt (Eds.), Ad-
vances in child development and behavior (Vol. 13, pp.
257–295). New York: Academic.
Stevenson, H. N. C. (1954). Status evaluation in the Hindu caste sys-
tem. Journal of the Royal Anthropological Institute of Great
Britain and Ireland,84, 45–65.
Taylor, S. E. (1991). Asymmetrical effects of positive and negative
events: The mobilization-minimization hypothesis. Psycholog-
ical Bulletin,110, 67–85.
Thompson, R. A. (1987). Empathy and emotional understanding:
Theearlydevelopmentofempathy.In N. Eisenberg &J.Strayer
(Eds.),Empathyanditsdevelopment(pp.119–146).NewYork:
Cambridge University Press.
Titchener, E. B. (1908). An outline of psychology. New York:
Macmillan.
Tolstoy, L. N. (1998). Anna Karenina (L. Maude & A. Maude,
Trans.). Oxford, England: Oxford University Press. (Original
work published 1875)
Tomkins, S. S. (1963). Affect, imagery, consciousness: Vol. 2. The
negative affects. New York: Springer.
Troland, L. T. (1928). The fundamentals of human motivation. New
York: Van Nostrand.
Tversky, A., & Kahneman, D. (1991). Loss aversion in riskless
choice:Areference-dependentmodel.The Quarterly Journal of
Economics, 106, 1039–1061.
Tylor, E. B. (1974). Primitive culture: Researches into the develop-
ment of mythology, philosophy, religion, art and custom. New
York: Gordon Press. (Original work published 1871)
319
NEGATIVITY BIAS
Warren, R. P., & Pfaffman, C. (1959). Early experience and taste
aversion. Journal of Comparative and Physiological Psychol-
ogy,52, 263–266.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and
validation of brief measures of positive and negative affect:
The PANAS scales. Journal of Personality and Social Psy-
chology,54, 1063–1070.
Wojciszke, B., Brycz, H., & Boreknau, P. (1993). Effects of informa-
tion content and evaluative extremity on positivity and negativity
biases. Journal of Personality and Social Psychology,64,
327–335.
Woodworth, R. S. (1938). Experimental psychology. New York:
Holt.
Wrzesniewski, A., Rozin, P., & Bennett, G. (in press). Working, playing,
andeating:Makingthemostof most moments. In C.L.M.Keyes&J.
Haidt (Eds.), Flourishing: The positive person and the good life.
Washington, DC: American Psychological Association.
Wundt,W.(1897).Outlinesof psychology (C. H.Judd,Trans.).Leip-
zig, Germany: Wilhelm Engelmann.
Zahorik,D.(1979).Learnedchangesinpreferencesforchemical
stimuli:asymmetricaleffectsofpositiveandnegativeconse-
quences,andspeciesdifferencesinlearning. In J. Solmsand
R.L.Hall (Eds.), Preferencebehaviourand chemoreception
(pp. 233–243). London: Information Retrieval.
Zellner,D.A.,Rozin,P.,Aron, M., & Kulish,C.(1983).Conditioned
enhancement of human’s liking for flavors by pairing with
sweetness. Learning and Motivation,14, 338–350.
320
ROZIN & ROYZMAN
... Додатно, промене у њуз фидовима (news feeds), или како се другачије колоквијално каже "на зидовима" (on the wall), на "лајни" или "кућним страницама" (home page), којима је корисник изложен на друштвеним медијима, показале су да изазивају промене осећања онлајн корисника (Ferrara & Yang, 2015). Људи генерално реагују интензивније на негативна осећања, што проучава теорија негативног мишљења (theory of negativity bias) (Baumeister et al., 2001;Derks et al., 2008;Rozin & Royzman, 2001). У експерименту је доказано да негативна осећања имају већи утицај када се истовремено интензивирају и позитивна и негативна осећања код неке особе (Liebrecht et al., 2019). ...
... Additionally, alterations in the news feeds that users are exposed to on social media have shown a changing impact on the emotions of online users (Ferrara & Yang, 2015). People tend to react more intensively to negative emotions, which is explored by the theory of negativity bias (Baumeister et al., 2001;Derks et al., 2008;Rozin & Royzman, 2001). In an experiment by Liebrecht et al. (2019), it has been proved that negative emotions have a higher effect when both positive and negative emotions are intensified. ...
Article
Full-text available
Both tech companies and AI algorithms exercise immense power in today's globally interconnected world, which is based on big data and digital footprints of online users. This paper analyses the transfer of power from societies to tech companies and algorithms with the aim of examining whether recommender algorithms can be considered a public good. Deployed methods include content analysis and literature reviews. The study has found that control exercised over public opinion, decisions and moods of online users is unprecedented to such a high degree in human history. The above-mentioned control is based on the impact of both tech companies and algorithms. The limitation of this research is the lack of quantitative analysis. Future research should concentrate on defining recommender algorithms as a public good and analyzing how different media content, including virtual reality, affects citizens' psychology.
... Thereby, from a sponsor perspective, it is particularly important to consider the potential downside of customer engagement outcomes because consumers generally process negative information about a brand more intensely than positive information and are more likely to store it in their long-term memories (Yu et al., 2018). This phenomenon is referred to as "negativity bias," which states that consumers generally evaluate negative events as more prominent and powerful than positive events (Rozin and Royzman, 2001). Furthermore, latest research has shown that consumers' sponsorship evaluations depend on the authenticity of a sponsor-sponsee relationship (Shoffner and Koo, 2020). ...
... This finding supports Crompton's (2015) assertion that there are certain risks associated to sport sponsorship that can lead to negative publicity damaging the sponsors' reputation. Considering the potential of negative effects is especially important because of the so-called negativity bias (Rozin and Royzman, 2001). This study's results on the higher probabilities of negative CEB following sponsors' activations compared to positive CEB in all scenarios confirm the assumption that negative events dominate positive events (Yu et al., 2018). ...
Article
Purpose In marketing research, it is widely acknowledged that customer engagement leads to higher reputation, lower costs and increased revenues for firms. However, there are still open questions on how sport sponsorship can drive customer engagement. It is hypothesized that sponsors' activations correlate with customer engagement toward the sponsor. Specifically, the roles of sponsorship authenticity and attitudes toward the sponsor have received little attention in this context. Accordingly, this study aims to test the effects of sponsors' activations on customer engagement disposition (CED) and customer engagement behavior (CEB) by considering the roles of sponsorship authenticity and attitudes toward the sponsor. Design/methodology/approach An online experiment with a factorial between-subjects design with 529 total participants was conducted. Data were analyzed through analysis of covariance (ANCOVA) and binary regression analysis. Findings Sponsors' activations can lead to positive or negative CEB, depending on how sport consumers evaluate the activation. Sponsorship authenticity reduces or enhances CEB following a sponsor's activation. Moreover, consumers' prior attitudes toward the sponsor influence the relationship between sponsors' activations and CED. The findings further showed that CED leads to CEB. Originality/value This research contributes to the sport sponsorship literature by empirically proving that sponsors' activations increase customer engagement toward the sponsors. Moreover, this is the first study testing consequences of sponsors' activations in relation to sponsorship authenticity and consumers' attitudes. Furthermore, the authors enrich the customer engagement literature by discussing the sponsors' activations as a marketing strategy to increase customer engagement and consequently firms' performance.
... Although we hypothesized a similar bidirectional relationship for positive affect words, there are a few reasons this monodirectional relationship may emerge. People tend to give greater attention to negative than positive information (Baumeister et al., 2001;Rozin & Royzman, 2001), and recent studies have shown stronger influence of negative than positive language on social media posts (Schöne et al., 2021). This may lead children (especially anxious children, Abend et al., 2018;Fu & Pérez-Edgar, 2019;Puliafico & Kendall, 2006) to not be as sensitive to taking on positive affective language expressed by their mothers as they do for negative language. ...
Article
Humans influence each other’s emotions. The spread of emotion is well documented across behavioral, psychophysiological, and neuroscientific levels of analysis, but might this influence also be evident in language (e.g., are people more likely to use emotion words after hearing someone else use them)? The current study tests whether mothers and children influence each other’s use of affective language. From 2018-2020, children aged 6-12 who met diagnostic criteria for anxiety disorders and their mothers (N=93 dyads) completed a challenging puzzle task while being video recorded. Analyses of transcriptions revealed that mothers and children indeed influenced each other’s language. Bidirectional influence was observed for use of negative affect words: Mothers were more likely to use negative affect words if their child had just used negative affect words (over and above mothers' own language on their previous turn), and children were similarly influenced by mother affect word use. A similar bidirectional relation emerged for linguistic distance, a measure related to effective emotion regulation and mental health. However, the significance of the child-to-mother direction of influence for these two variables varied depending on correction threshold and should thus be verified in future research. Nonetheless, these findings extend understanding of emotional influence by showing turn-by-turn relations between use of affective language.
... Positive emotions signal safety and the fulfillment of primary needs, while negative emotions have an alarm function (Tugade, 2010). Negative emotions are therefore more informative than positive emotions because the price of failing to notice a major threat easily outweighs the costs of a missed opportunity (Öhman & Mineka, 2001;Rozin & Royzman, 2001). In line with this, the dynamics of negative emotions have also been found to play a larger role in explaining psychological well-being in adults (Houben et al., 2015). ...
Article
Full-text available
Research on emotion dynamics as indices of emotion functioning has become muddled by conceptual confusion, methodological heterogeneity, and seemingly conflicting results. One way to address this chaos is the study of profiles of emotion dynamics across 12 emotions and how they differ between 246 adolescents. The interpretation of these dynamic profiles was guided by auxiliary variables including age, personality, depressive symptoms, and social experiences. Method: During 6 days, 246 adolescents (Mage=14.20, 65% female) rated 9 times daily the intensity of 12 emotions (cheerful, happy, energetic, joyful, content, relaxed, anxious, worried, irritable, insecure, down, and guilty), and their social experiences with family, friends, and classmates. Additional baseline measures included neuroticism, extraversion (JEPQR-S), and depressive symptoms (CES-D). A three-mode principal component analysis (3MPCA Tucker3-based) model was estimated on the person-specific dynamic parameters of emotional intensity (mean), variability (standard deviation), instability (mean squared successive difference), and inertia (autocorrelation). Results: The 3MPCA identified three emotion-mode components (positive affect, negative affect, and irritability), three dynamic-mode components (emotional intensity, lability, and inertia). Five individual-mode components captured interactions between these modes, of which positive affect explained most variation in the data. These emotion dynamic profiles correlated differently with social experiences. Additional 3MPCA model structures based on imputed data (correcting missing autocorrelations) and affect scale composites (low and high arousal positive and negative affect) showed strong resemblance. Conclusion: The identified emotion dynamic profiles capture meaningful interpersonal differences in adolescents’ emotional experiences and change. Future work should focus on irritability and positive affect as these were uniquely informative in adolescents’ emotional experiences.
... During observations, participants may have been more likely to express frustrations rather than positive experiencesthe "negativity bias" psychological phenomonon. 33 Clinicians at HospitalQ were simultaneously adjusting to a new hospital in addition to digitization. Additionally, observations were predominantly conducted during mornings (approximately 7 a.m. to 2 p.m.), potentially missing critical clinical behaviors related to care digitization outside of those hours. ...
Article
Full-text available
Background Understanding electronic medical record (EMR) implementation in digital hospitals has focused on retrospective “work as imagined” experiences of multidisciplinary clinicians, rather than “work as done” behaviors. Our research question was “what is the behavior of multidisciplinary clinicians during the transition to a new digital hospital?” Objectives The aim of the study is to: (1) Observe clinical behavior of multidisciplinary clinicians in a new digital hospital using ethnography. (2) Develop a thematic framework of clinical behavior in a new digital hospital. Methods The setting was the go-live of a greenfield 182-bed digital specialist public hospital in Queensland, Australia. Participants were multidisciplinary clinicians (allied health, nursing, medical, and pharmacy). Clinical ethnographic observations were conducted between March and April 2021 (approximately 1 month post-EMR implementation). Observers shadowed clinicians in real-time performing a diverse range of routine clinical activities and recorded any clinical behavior related to interaction with the digital hospital. Data were analyzed in two phases: (1) content analysis using machine learning (Leximancer v4.5); (2) researcher-led interpretation of the text analytics to generate contextual meaning and finalize themes. Results A total of 55 multidisciplinary clinicians (41.8% allied health, 23.6% nursing, 20% medical, 14.6% pharmacy) were observed across 58 hours and 99 individual patient encounters. Five themes were derived: (1) Workflows for clinical documentation; (2) Navigating a digital hospital; (3) Digital efficiencies; (4) Digital challenges; (5) Patient experience. There was no observed harm attributable to the digital transition. Clinicians primarily used blended digital and paper workflows to achieve clinical goals. The EMR was generally used seamlessly. New digital workflows affected clinical productivity and caused frustration. Digitization enabled multitasking, clinical opportunism, and benefits to patient safety; however, clinicians were hesitant to trust digital information. Conclusion This study improves our real-time understanding of the digital disruption of health care and can guide clinicians, managers, and health services toward digital transformation strategies based upon “work as done.”
... By extension, because negative and positive life events influence negative and positive affective states respectively (Zautra & Reich, 1983), the undesirable nature of inconsistent leadership should be related to negative, but not positive, affect. Second, negative events and interpersonal interactions are generally more influential than their positive counterparts in affecting well-being related outcomes (Baumeister et al., 2001;Rook, 2001;Rozin & Royzman, 2001). ...
Article
When public health emergencies occur, a large amount of low-credibility information is widely disseminated by social bots, and public sentiment is easily manipulated by social bots, which may pose a potential threat to the public opinion ecology of social media. Therefore, exploring how social bots affect the mechanism of information diffusion in social networks is a key strategy for network governance. This study combines machine learning methods and causal regression methods to explore how social bots influence information diffusion in social networks with theoretical support. Specifically, combining stakeholder perspective and emotional contagion theory, we proposed several questions and hypotheses to investigate the influence of social bots. Then, the study obtained 144,314 pieces of public opinion data related to COVID-19 in J city from March 1, 2022, to April 18, 2022, on Weibo, and selected 185,782 pieces of data related to the outbreak of COVID-19 in X city from December 9, 2021, to January 10, 2022, as supplement and verification. A comparative analysis of different data sets revealed the following findings. Firstly, through the STM topic model, it is found that some topics posted by social bots are significantly different from those posted by humans, and social bots play an important role in certain topics. Secondly, based on regression analysis, the study found that social bots tend to transmit information with negative sentiments more than positive sentiments. Thirdly, the study verifies the specific distribution of social bots in sentimental transmission through network analysis and finds that social bots are weaker than human users in the ability to spread negative sentiments. Finally, the Granger causality test is used to confirm that the sentiments of humans and bots can predict each other in time series. The results provide practical suggestions for emergency management under sudden public opinion and provide a useful reference for the identification and analysis of social bots, which is conducive to the maintenance of network security and the stability of social order.
Article
Full-text available
Since the advent of the Internet has changed how risk information develops and disseminates, citizens’ risk perception alters correspondingly. Although extant studies have explored the impacts of Internet use on citizens, only a few have focused on citizens’ perception of social safety exclusively. This study examined the effect of Internet use and Internet use frequency on citizens’ perception of social safety with 2017 China Social Survey data. It found that Internet use and Internet use frequency have a significant negative impact on citizens’ social safety perception. Compared to non-users, the probability of perceiving society as very safe decreases significantly by 2.3% for Internet users. Subsequently, this study avoided the endogeneity issues by employing the Bioprobit and CMP methods, and the robustness check adopted propensity score matching methods. In general, it supported results in benchmark regression. The heterogeneity analysis indicated that Internet use had a higher negative impact on citizens in the western region, rural household registration, and middle-aged groups. It suggests that the establishment of a comprehensive mechanism to regulate online information involving governments, Internet industrial organizations, and Internet users may improve governance effectiveness. In addition, education targeting Internet literacy is required to enable a more rational citizen participation on the Internet.
Chapter
How can we get the most out of our close relationships? Research in the area of personal relationships continues to grow, but most prior work has emphasized how to overcome negative aspects. This volume demonstrates that a good relationship is more than simply the absence of a bad relationship, and that establishing and maintaining optimal relationships entails enacting a set of processes that are distinct from merely avoiding negative or harmful behaviors. Drawing on recent relationship science to explore issues such as intimacy, attachment, passion, sacrifice, and compassionate goals, the essays in this volume emphasize the positive features that allow relationships to flourish. In doing so, they integrate several theoretical perspectives, concepts, and mechanisms that produce optimal relationships. The volume also includes a section on intensive and abbreviated interventions that have been empirically validated to be effective in promoting the positive features of close relationships.
Article
Indicators are important sources of information about problems across many policy areas. However, despite a growing number of indicators across most policy areas, such as health care, business promotion, or environmental protection, we still know little about if, how, and when such indicators affect the policy agenda. This article develops a theoretical answer to these questions and examines the implications using a new large-n dataset with 220,000 parliamentary questions asked by government and opposition MPs in Australia, Belgium, Denmark, Germany, France, Italy, and Spain. The data contain information on political attention to 17 problems, such as unemployment, C02 emission, and crime from 1960 to 2015. Across this wealth of data, the article demonstrates that politicians respond to the severity and development of problem indicators over time and in comparison to other countries. Results also show that politicians respond much more when problem indicators develop negatively than when they develop positively.
Chapter
This book raises the idea of a distinct discipline of cultural psychology, the study of the ways that psyche and culture, subject and object, and person and world make up each other. Cultural Psychology is a collection of essays from leading scholars in anthropology, psychology, and linguistics who examine these relationships with special reference to core areas of human development: cognition, learning, self, personality dynamics, and gender. The chapters critically examine such questions as: Is there an intrinsic psychic unity to humankind? Can cultural traditions transform the human psyche, resulting less in psychic unity than in ethnic divergences in mind, self, and emotion? Are psychological processes local or specific to the sociocultural environments in which they are embedded? The volume is an outgrowth of the internationally known Chicago Symposia on Culture and Human Development. It will appeal to an interdisciplinary audience of anthropologists, psychologists, linguists, historians, philosophers and hermeneutists interested in the prospects for a distinct discipline of cultural psychology.