ArticlePDF AvailableLiterature Review

On Seeing Human: A Three-Factor Theory of Anthropomorphism

Authors:

Abstract

Anthropomorphism describes the tendency to imbue the real or imagined behavior of nonhuman agents with humanlike characteristics, motivations, intentions, or emotions. Although surprisingly common, anthropomorphism is not invariant. This article describes a theory to explain when people are likely to anthropomorphize and when they are not, focused on three psychological determinants--the accessibility and applicability of anthropocentric knowledge (elicited agent knowledge), the motivation to explain and understand the behavior of other agents (effectance motivation), and the desire for social contact and affiliation (sociality motivation). This theory predicts that people are more likely to anthropomorphize when anthropocentric knowledge is accessible and applicable, when motivated to be effective social agents, and when lacking a sense of social connection to other humans. These factors help to explain why anthropomorphism is so variable; organize diverse research; and offer testable predictions about dispositional, situational, developmental, and cultural influences on anthropomorphism. Discussion addresses extensions of this theory into the specific psychological processes underlying anthropomorphism, applications of this theory into robotics and human-computer interaction, and the insights offered by this theory into the inverse process of dehumanization.
On Seeing Human: A Three-Factor Theory of Anthropomorphism
Nicholas Epley, Adam Waytz, and John T. Cacioppo
University of Chicago
Anthropomorphism describes the tendency to imbue the real or imagined behavior of nonhuman agents
with humanlike characteristics, motivations, intentions, or emotions. Although surprisingly common,
anthropomorphism is not invariant. This article describes a theory to explain when people are likely to
anthropomorphize and when they are not, focused on three psychological determinants—the accessibility
and applicability of anthropocentric knowledge (elicited agent knowledge), the motivation to explain and
understand the behavior of other agents (effectance motivation), and the desire for social contact and
affiliation (sociality motivation). This theory predicts that people are more likely to anthropomorphize
when anthropocentric knowledge is accessible and applicable, when motivated to be effective social
agents, and when lacking a sense of social connection to other humans. These factors help to explain why
anthropomorphism is so variable; organize diverse research; and offer testable predictions about dispo-
sitional, situational, developmental, and cultural influences on anthropomorphism. Discussion addresses
extensions of this theory into the specific psychological processes underlying anthropomorphism,
applications of this theory into robotics and human–computer interaction, and the insights offered by this
theory into the inverse process of dehumanization.
Keywords: anthropomorphism, social cognition, animal cognition, agency, mind perception
Our planet is currently inhabited by approximately 1.75 million
known species with unique phylogenetic characteristics (UNEP
Biodiversity Assessment, as cited in Heywood, 1995), at least
10,000 distinct religions each describing its own set of supernat-
ural beings (D. Barrett, Kurian, & Johnson, 2001), and an expo-
nentially expanding number of technological devices designed to ease
the burdens of everyday life. This diversity of living, spiritual, and
technological agents is staggering, yet people’s representations of
these agents often take on a suspiciously familiar appearance. Ani-
mals are imbued with humanlike intentions, motivations, and goals.
Spiritual deities are embodied with fingers and facial hair, complete
with personality strengths and occasionally personality weaknesses.
And even the most technologically savvy have wondered, at least for
a moment, whether their computer is plotting against them.
Scholars from a wide array of disciplines have long noted that
people tend to see nonhuman agents as humanlike (Darwin, 1872/
2002; Feuerbach, 1873/2004; Freud, 1930/1989; Hume, 1757/
1956). Debates have ensued about whether such anthropomor-
phism represents accurate or fallacious thinking, whether
anthropomorphic descriptions have any place in scientific dis-
course, and whether anthropomorphism can account for phenom-
ena ranging from religious belief to effective marketing campaigns
(Aggarwal & McGill, in press). Empirical investigations have
demonstrated the ease with which people provide anthropomor-
phic descriptions of agents ranging from God (J. L. Barrett & Keil,
1996) to geometric shapes (Heider & Simmel, 1944) to moving
plants and computer-animated blobs (Morewedge, Preston, &
Wegner, 2007; see Guthrie, 1993, for additional examples). As we
note later, treating agents as human versus nonhuman has a pow-
erful impact on whether those agents are treated as moral agents
worthy of respect and concern or treated merely as objects, on how
people expect those agents to behave in the future, and on people’s
interpretations of these agents’ behavior in the present. Anthropo-
morphized agents can act as powerful agents of social connection
when human connection is lacking, and anthropomorphizing tech-
nological agents appears to aid in effectively learning how to use
those agents. But these existing observations and empirical re-
search do not provide a psychological account of anthropomor-
phism itself, nor do they explain or predict systematic variability in
the tendency to anthropomorphize nonhuman agents. This article
attempts to do so by synthesizing a widely dispersed literature that
predicts variability in anthropomorphism across dispositional, sit-
uational, cultural, and developmental sources of influence. This
account should provide testable predictions not only about anthro-
pomorphism—when people “see” nonhumans as human—but also
about the inverse process of dehumanization as well—when peo-
ple “see” human agents as nonhuman.
Anthropomorphism:
Historical Issues and Current Concerns
Imbuing the imagined or real behavior of nonhuman agents with
humanlike characteristics, motivations, intentions, and emotions is
Nicholas Epley, Graduate School of Business, University of Chicago;
Adam Waytz and John T. Cacioppo, Department of Psychology, University
of Chicago.
This research was supported by National Science Foundation Grant
SES0241544, National Institute of Aging Grant PO1 AG18911, and the
John Templeton Foundation. We thank Jasmine Kwong for editorial as-
sistance and Reid Hastie, Nicholas Haslam, Richard Thaler, Leaf Van
Boven, Daniel Wegner, and members of the Chicago Templeton Network
and University of Chicago Center for Decision Research for helpful com-
ments on this work.
Correspondence concerning this article should be addressed to Nicholas
Epley, University of Chicago, 5807 South Woodlawn Avenue, Chicago, IL
60637. E-mail: epley@chicagogsb.edu
Psychological Review Copyright 2007 by the American Psychological Association
2007, Vol. 114, No. 4, 864 886 0033-295X/07/$12.00 DOI: 10.1037/0033-295X.114.4.864
864
the essence of anthropomorphism. These nonhuman agents may
include anything that acts with apparent independence, including
nonhuman animals, natural forces, religious deities, and mechan-
ical or electronic devices. As the Oxford Dictionary (Soanes &
Stevenson, 2005) more simply puts it, anthropomorphism is the
“attribution of human characteristics or behavior to a god, animal,
or object” (p. 66).
Derived from the Greek words anthro pos (meaning “human”)
and morphe (meaning “shape” or “form”), anthropomorphism
involves more than simply attributing life to the nonliving (i.e.,
animism). Anthropomorphism involves going beyond behavioral
descriptions of imagined or observable actions (e.g., the dog is
affectionate) to represent an agent’s mental or physical character-
istics using humanlike descriptors (e.g., the dog loves me). At its
core, anthropomorphism entails attributing humanlike properties,
characteristics, or mental states to real or imagined nonhuman
agents and objects. The attributes involved in the perception of
mind—such as conscious experience, metacognition, and inten-
tions (Gray, Gray, & Wegner, 2007)—are therefore central to
anthropomorphism but are not exhaustive. Anthropomorphism
also entails attributing humanlike emotional states (Leyens et al.,
2003), behavioral characteristics, or humanlike forms to nonhu-
man agents. Xenophanes (6th century B.C., as cited in Lesher,
1992), for instance, was the first to use “anthropomorphism” when
describing the similarities between religious agents and their be-
lievers, noting that Greek gods were invariably fair skinned and
blue-eyed whereas African gods were invariably dark skinned and
dark-eyed (joking that cows would surely worship gods that were
strikingly cowlike). Anthropomorphism is therefore a process of
inference about unobservable characteristics of a nonhuman agent,
rather than descriptive reports of a nonhuman agent’s observable
or imagined behavior.
Very little of the existing research on anthropomorphism tries to
provide a psychological account of when and why people are
likely to anthropomorphize nonhuman agents, but rather it inves-
tigates the accuracy and functionality of these anthropomorphic
descriptions. Most existing research therefore addresses the actual
mental states and abilities of nonhuman agents by detailing the
extent to which these agents really are humanlike and the extent to
which anthropomorphism represents accurate insight versus fan-
ciful illusion (e.g., Cheney & Seyfarth, 1990; Hauser, 2000; Mor-
gan, 1894; Thorndike, as cited in Gates, 1949). Indeed, 104 of the
182 articles that emerge when “anthropomorphism” is searched in
the PsycINFO database (April 2007) involve discussions of, or
empirical work on, the capacities of nonhuman animals.
This existing research on anthropomorphism therefore tends to
overlook the very psychological phenomenon in its midst by
studying either the extent to which people anthropomorphize or the
accuracy of these beliefs rather than by providing a psychological
account of the beliefs themselves. Although interesting, the accu-
racy of anthropomorphic beliefs is simply orthogonal to under-
standing the psychological processes that give rise to these intui-
tive beliefs in the first place. Psychologically, anthropomorphism
has been generally considered an invariant and automatic psycho-
logical process that is simply a chronic feature of human judgment
(Guthrie, 1993; see also Mitchell, Thompson, & Miles, 1997). We
attempt to fill this gap by providing a psychological account of
anthropomorphism itself. The major benefit of this account is the
ability to predict variability in anthropomorphism, and therefore
variability in the consequences that follow from anthropomor-
phism as well. Some nonhuman agents are anthropomorphized
more than others. Children appear to anthropomorphize nonhuman
agents more than adults. Some people anthropomorphize nonhu-
man agents more than other people. Some situations seem to elicit
anthropomorphic beliefs more than others, and some cultures seem
especially fond of anthropomorphic descriptions compared with
others. A psychological account of anthropomorphism explains
why people are likely to anthropomorphize in the first place and in
so doing predicts when people are likely to anthropomorphize
nonhuman agents and when they are not.
SEEK: A Psychological Account of Anthropomorphism
Anthropomorphism represents a process of inductive inference
about nonhuman agents, and the basic cognitive operations that
perform such inferences should be no different for anthropomor-
phic inferences than for any other inductive inferences. These
basic cognitive operations include the acquisition of knowledge,
the activation or elicitation of stored knowledge, and the applica-
tion of activated knowledge to a given target (Higgins, 1996). This
last application process includes attempts to correct or adjust
highly accessible knowledge in order to integrate alternative
knowledge structures that are coactivated at the time of judgment.
Such correction processes are generally insufficient such that final
judgments are influenced in the direction of the most readily
accessible information. These process accounts are best known in
domains of dispositional inference, belief formation, and social
comparison (Gilbert, 1998) but have also been applied to judg-
mental anchoring (Epley, 2004), affective forecasting (Gilbert,
Gill, & Wilson, 2002), overconfidence (Griffin & Tversky, 1992),
self-assessment (Kruger, 1999), and a variety of egocentric biases
in social cognition (Epley, Keysar, Van Boven, & Gilovich, 2004;
Keysar & Barr, 2002).
As a basic process of induction, anthropomorphism works
through a similar process of starting with highly accessible knowl-
edge structures as an anchor or inductive base that may be subse-
quently corrected and applied to a nonhuman target. The extent to
which people anthropomorphize should therefore be determined
by the three major parts of the inductive process: (a) the likelihood
of activating, either chronically or situationally, knowledge about
humans when making inferences about nonhuman agents; (b) the
likelihood of correcting or adjusting anthropomorphic representa-
tions to accommodate nonanthropomorphic knowledge about non-
human agents; and (c) the likelihood of applying activated and
possibly corrected anthropomorphic representations to nonhuman
agents. Our account seeks to explain when and why knowledge
about the self or humans in general is likely to be highly accessible
and subsequently applied with little correction when reasoning
about nonhuman agents, and when that knowledge is likely to be
relatively inaccessible or unlikely to be applied. Our theoretical
account is therefore not a theory of the basic process that underlies
induction itself but is rather a theory of a particular kind of
inductive inference—the attribution of human characteristics or
traits to nonhuman agents (i.e., anthropomorphism)—and what we
believe are the three key psychological determinants that will
influence this particular kind of inference.
865
ANTHROPOMORPHISM
Key Psychological Determinants
Like any complicated phenomena, we propose that anthropo-
morphism is multiply determined, involving both cognitive and
motivational determinants. In anthropomorphism, knowledge
about humans in general, or self-knowledge more specifically,
functions as the known and often readily accessible base for
induction about the properties of unknown agents. A primary
determinant of anthropomorphism is therefore the elicitation of
agent knowledge itself. Knowledge about humans in general, or
the self in particular, is likely to serve as the basis for induction
primarily because such knowledge is acquired earlier and is more
richly detailed than knowledge about nonhuman agents. Such
acquired knowledge about human characteristics or qualities is
therefore likely to be more readily accessible at the time of
judgment. As knowledge about nonhuman agents is acquired,
however, knowledge about humans or the self should be less likely
to be used as a basis for induction simply because of the coacti-
vation (and perhaps eventual substitute activation) of alternate
knowledge structures at the time of judgment. Such activation of
alternate knowledge structures will then influence the application
of accessible knowledge to a given target either through a process
of correction to incorporate competing knowledge or through an
integration of accessible knowledge (Gilbert, 1991; Gilbert &
Malone, 1995; Trope & Gaunt, 2000). Understanding how knowl-
edge about human and nonhuman agents is acquired, activated (or
elicited), corrected, and applied to a target is therefore a central
feature of our theoretical account of anthropomorphism.
We propose that this cognitive mechanism of elicited agent
knowledge works in concert with two additional motivational
mechanisms— effectance and sociality. Effectance describes the
need to interact effectively with one’s environment (White, 1959).
Applied to anthropomorphism, effectance involves the motivation
to interact effectively with nonhuman agents (or perceived agents)
and operates in the service of enhancing one’s ability to explain
complex stimuli in the present and to predict the behavior of these
stimuli in the future. Attributing human characteristics and moti-
vations to nonhuman agents increases the ability to make sense of
an agent’s actions, reduces the uncertainty associated with an
agent, and increases confidence in predictions of this agent in the
future. At the very least, anthropomorphism provides a rich source
of testable hypotheses to guide a person’s behavior toward an
unknown agent or stimulus. The anxiety associated with uncer-
tainty and the importance of predicting an agent’s behavior should
therefore influence people’s tendency to anthropomorphize a non-
human agent. Sociality, in contrast, describes the need and desire
to establish social connections with other humans. Anthropomor-
phism enables satisfaction of this need by enabling a perceived
humanlike connection with nonhuman agents. In the absence of
social connection to other humans, in other words, people create
human agents out of nonhumans through anthropomorphism to
satisfy their motivation for social connection. This predicts that
anthropomorphism will increase when people feel a lack of social
connection to other humans and decrease when people feel a
strong sense of social connection.
These three determinants (Sociality, Effectance, and Elicited
agent Knowledge—SEEK), we propose, work in concert to in-
crease or decrease the extent to which a person anthropomorphizes
a nonhuman agent at any one time by altering the activation,
correction, or application of anthropomorphic knowledge to a
given target during this inductive reasoning process. The cognitive
factor of elicited agent knowledge specifies which cognitive vari-
ables, independent of motivational influences, are likely to render
knowledge about humans or the self highly accessible; when
correction of an anthropomorphic representation is likely and
when it is not; what cognitive factors are likely to increase the
application of anthropomorphic representations to nonhuman
agents; when accessible anthropomorphic representations are
likely to be corrected by competing representations and when they
are not; and when accessible anthropomorphic representations are
likely to be applied to a given target. Each part of this inductive
process, however, may be guided or modified by effectance or
sociality motivations. High levels of sociality motivation, we sug-
gest, can increase the accessibility of anthropomorphic or human-
like representations of nonhuman agents as a result of the attempt
to identify a potential source of social connection, diminish the
tendency to correct an initial anthropomorphic or humanlike rep-
resentation of a nonhuman agent, and thereby increase the ten-
dency to apply anthropomorphic representations to nonhuman
agents. High levels of effectance motivation should similarly exert
influence on the activation, correction, and application of anthro-
pomorphic representations. Each of these factors therefore makes
unique predictions about variables that will influence the tendency
to anthropomorphize nonhuman agents. In this article, we first
discuss predictions derived from each of these three factors inde-
pendently and later discuss predictions derived from the interrela-
tion between these factors.
Categories of Independent Variables
The major benefit of a psychological account of anthropomor-
phism is the ability to predict variability across the major catego-
ries of independent variables in everyday life, namely disposi-
tional, situational, developmental, and cultural. Our theory is
designed to make unique predictions about how specific indepen-
dent variables within each of these categories will influence an-
thropomorphism through each of the key psychological determi-
nants. Table 1 presents the specific independent variables within
each of the four major categories for each of the three factors that
we predict will influence anthropomorphism.
Dispositional variables, at least those described in this account,
are stable individual differences or personality traits that alter the
extent to which certain knowledge representations or motivational
states are chronically active. Situational variables represent tran-
sitory aspects of the environment that can alter the accessibility of
knowledge representations or can increase or decrease effectance
and sociality motivations. Finally, developmental and cultural
variables can alter anthropomorphism by influencing the content
of agent representations and the strength of effectance motivation
across developmental time and cultural place. We believe that
understanding these three psychological factors—sociality, effec-
tance, and elicited agent knowledge—and how they are influenced
by specific variables within each of these major categories of
influence— dispositional, situational, developmental, and
cultural—is critical for understanding anthropomorphism in daily
life.
866
EPLEY, WAYTZ, AND CACIOPPO
Varieties of Anthropomorphic Experience
Like any set of beliefs, from stereotypes to politics to simple
preferences, the strength of anthropomorphic beliefs can vary
along a continuum, from those held very strongly to those held
more weakly. Religious believers, for instance, frequently speak of
God’s will, intentions, or desires— beliefs about the mental states
of religious agents that are held by many with the deepest and most
sincere conviction. So too do people regularly speak of pets as
being thoughtful and considerate, and they would certainly re-
spond harshly if one were to suggest that they must be joking. But
weaker forms of anthropomorphism may emerge as well that
appear to be metaphorical ways of thinking rather than firmly held
beliefs that an agent has humanlike traits. People may try to
verbally cajole a stalled car into starting or explain that one’s
computer hates them. Even physicists may refer to particles as
charming or shy.
Our contention is certainly not that all instances of anthropo-
morphic language reflect the same degree or strength of explicit
belief about actual humanlike characteristics, just as researchers
studying stereotyping and prejudice don’t suggest that all of their
research participants hold their beliefs with the same degree of
explicit endorsement (Fazio & Olson, 2003), or that a person who
refuses to eat a piece of chocolate shaped like a cockroach some-
how confuses the chocolate for an actual cockroach (Rozin &
Nemeroff, 2002). Strong forms of anthropomorphism entail be-
having as if a nonhuman agent has humanlike traits or character-
istics along with explicit endorsement of those beliefs (such as
with religious agents), whereas weaker forms may only entail “as
if” metaphorical reasoning (such as with one’s malevolent com-
puter). We do contend, however, that our theoretical account
explains the occurrence of both strong and weak forms of anthro-
pomorphism equally well and that the antecedents to these weak
and strong forms of beliefs are the same and are captured well
within our theoretical account.
What is more, weak versions of anthropomorphism in which
inferences may appear to be simple metaphorical reasoning may
matter more than intuition would suggest. Metaphors that might
represent a very weak form of anthropomorphism can still have a
powerful impact on behavior, with people behaving toward agents
in ways that are consistent with these metaphors. As Lakoff and
Johnson (1980) noted, using metaphors such as “Inflation has
attacked the foundation of our economy” provides both a way of
thinking about an event and also “a way of acting towards it. The
INFLATION IS AN ADVERSARY metaphor therefore gives rise
to and personifies political and economic actions on the part of our
government; declaring war on inflation, setting targets, calling for
sacrifices, installing a new chain of command, etc.” (pp. 33–34).
Indeed, stock market analysts in one recent line of research quite
naturally invoked agentic and anthropomorphic descriptions of the
stock market (e.g., the NASDAQ flirted with the 2,000 mark), and
these descriptions influenced participants’ experimental invest-
ments in those markets in ways consistent with the metaphoric
implications (Morris, Sheldon, Ames, & Young, in press). The
difference between weak and strong versions of anthropomor-
phism, we suggest, is simply a matter of degree regarding the
strength and behavioral consequences of a belief, not a fundamen-
tal difference in kind.
Overview
Given the relatively small amount of research on anthropomor-
phic beliefs themselves, much of the supporting evidence that
follows comes not from research investigating how people think
about nonhuman agents but rather from research investigating how
people think about other humans. Anthropomorphism itself in-
volves a generalization from humans to nonhuman agents through
a process of induction, and the same mental processes involved in
thinking about other humans should also govern how people think
about nonhuman agents. Indeed, the same neural systems involved
in making judgments about other humans are also activated when
making anthropomorphic judgments about nonhuman agents (Cas-
telli, Happe´, Frith, & Frith, 2000; Iacoboni et al., 2004). And
people with deficits in the neural systems associated with difficul-
ties attributing mental states to human agents, such as autism,
show the same deficits when reasoning about nonhuman agents
(Heberlein & Adolphs, 2004). This evidence suggests that one of
the best ways to learn about when and why people are likely to
think of nonhuman agents as humanlike is to consider how people
think about other people.
The remainder of this article provides evidence for the three
psychological determinants of anthropomorphism, provides sup-
port for how operational variables from four major categories of
variance can impact each of these major mechanisms, and identi-
fies unique and novel predictions about how dispositional, situa-
tional, cultural, and developmental factors should influence an-
thropomorphism (see Table 1). We end this review by describing
interactions between the three psychological determinants and
describe several important extensions of this theory into human–
computer interaction and the inverse process of dehumanization.
Table 1
Key Psychological Determinants (Elicited Agent Knowledge, Effectance, and Sociality), Categories of Independent Variables, and
Predicted Sources of Influence From Specific Independent Variables on Anthropomorphism
Categories of
independent
variables
Key psychological determinants
Elicited agent knowledge Effectance motivation Sociality motivation
Dispositional Need for cognition Need for closure, desire for control Chronic loneliness
Situational Perceived similarity Anticipated interaction, apparent predictability Social disconnection
Developmental Acquisition of alternative theories Attaining competence Attachment
Cultural Experience, norms, and ideologies Uncertainty avoidance Individualism and collectivism
Note. The predicted sources of influence presented in this table, and in the article more generally, are meant to be illustrative rather than exhaustive.
867
ANTHROPOMORPHISM
Cognitive Determinants of Anthropomorphism
Elicited Agent Knowledge
Inductive inference operates much like an intuitive scientist,
taking data from behavioral observations, mental simulations, or
verbal reports and interpreting them within a framework of acces-
sible hypotheses or theories (e.g., Barsalou, 1983; Holland, Ho-
lyoak, Nisbett, & Thagard, 1986; Murphy & Medin, 1985; Nisbett
& Ross, 1980). Inferences about other agents are therefore not only
a product of the other agent’s actual or imagined behavior but also
a product of knowledge representations accessible to the perceiver
at the time of judgment and subsequently applied to a given target
(Higgins, 1996). As such, anthropomorphism will be predicted in
large part by cognitive factors that determine the likelihood of
activating, either chronically or situationally, knowledge about
humans when making inferences about nonhuman agents; the
likelihood of correcting this anthropomorphic knowledge; and the
likelihood of applying knowledge about humans to nonhuman
agents.
There are three basic reasons to believe that knowledge about
humans, or the self in general, is likely to serve as the base for
inductive reasoning when considering nonhuman agents. First,
simple physical constraints mean that humans have direct and
immediate access to the phenomenological experience of being a
human but do not have such immediate access to the phenomeno-
logical experience of any nonhuman agent. A person cannot truly
know what it is like to be a bat (Nagel, 1974), a sloth (Gould,
1998), or any other nonhuman agent for that matter simply because
humans’ sensory experiences are in here, not in there. This invari-
ant feature of sensory apparati means that knowledge about human
experience will be directly experienced and thus be acquired more
easily, more completely, and more quickly than any knowledge
(however indirect) about what it is like to be a nonhuman agent. Of
course, this physical constraint applies to knowing what it is like
to be another person as well, and research has repeatedly demon-
strated that people make inferences about others’ mental states by
relying inordinately on their own mental states as a starting point
for induction (e.g., Epley, Keysar, et al., 2004; Keysar & Barr,
2002; Nickerson, 1999). Using one’s own mental states and char-
acteristics as a guide when reasoning about other humans is
egocentrism. Using one’s own mental states and characteristics as
a guide when reasoning about nonhuman agents is anthropomor-
phism.
Second, watching another agent’s action appears to activate a
phenomenological experience directly consistent with the agent’s
action, providing a default that is likely to guide subsequent
reasoning about that agent. Neuroscientific evidence has identified
a system of mirror neurons in the prefrontal cortex whose sole job
is to mimic the same neural regions that would be active if the
perceiving agent had performed these actions him or herself (e.g.,
Buccino et al., 2001; for a review, see Rizzolatti & Craighero,
2004). This mirror neuron system is the foundation for people’s
ability to empathize— or simultaneously experience—another per-
son’s emotional state (Carr, Iacoboni, Dubeau, Mazziotta, &
Lenzi, 2003). This rapid and automatic mirror system is therefore
likely to provide a rapid phenomenological experience that would
need to be undone or overcome by subsequent reasoning.
Finally, newborn infants are notorious in their need for intensive
care giving. Except in extremely rare cases, this care is provided
by other humans. The social life of infants in nearly every human
culture on the planet, no matter how primitive, is therefore dom-
inated by exposure to and contact with other humans. This expo-
sure and contact is exactly the kind of experience necessary to
create detailed, interconnected, and rich representations of human
characteristics and traits and to create relatively vague, discon-
nected, and sparse representations of nonhuman agents’ character-
istics and traits. As children age and are exposed to a wider array
of nonhuman agents, richer representations of these agents are
more likely to develop, and anthropomorphism toward such agents
should diminish as a result of the coactivation of alternate nonan-
thropomorphic representations.
Beyond these basic theoretical reasons, a wide variety of em-
pirical findings also suggest that knowledge about humans in
general, or self-knowledge in particular, is likely to serve as a
readily accessible base for induction when reasoning about non-
human agents. The dominance of self-knowledge or egocentric
experience when reasoning about other agents is evident very early
in human development. Children younger than 4 years of age, for
instance, do not distinguish between what they know and what
others know (Perner, 1991; Wimmer & Perner, 1983), fail to
provide enough information to identify referents in ambiguous
communication (Deutsch & Pechmann, 1982; Sonnenschein &
Whitehurst, 1984), and fail to distinguish between the way an
object appears to them and its objective reality (Flavell, 1986). As
Piaget (1929) characterized, at “the starting point in the life of
thought, we find a protoplasmic consciousness unable to make any
distinction between the self and things” (p. 235).
A key step in cognitive development involves learning to dis-
tinguish the self from other humans, and presumably from nonhu-
man agents as well (e.g., Amsterdam, 1972; Flavell, 1973; Gopnik
& Meltzoff, 1994; Piaget, 1929, 1932, 1962, 1969; Piaget &
Inhelder, 1948). Understanding others as distinct entities requires
first developing a more sophisticated self-concept, and then using
this knowledge to simulate another agent’ experiences and there-
fore infer another person’s mental states (e.g., Heal, 1986; Melt-
zoff & Brooks, 2001; Piaget, 1929). The development of this
capacity further contributes to the emergence of distinct cognitive
representations of oneself, others, and the superordinate category
of human beings more generally. In addition to the self-concept,
categorical knowledge about humans in general also provides an
elaborate and accessible cognitive representation that may be eas-
ily applied when making inferences about nonhuman agents
(Carey, 1985; Inagaki & Hatano, 1987).
Although most full-grown adults develop the ability to recog-
nize that others have unique mental states of their own, they do not
appear to outgrow their childish ways altogether. A person’s own
knowledge and phenomenological experience are so automatically
accessible and richly organized that they continue to serve as an
automatic base for induction that needs to be overcome and cor-
rected when reasoning about others, rather than being a childhood
tendency that is outgrown. Insufficient correction of this egocen-
tric base appears to explain, at least in part, the robustness of
egocentric biases even among full-grown adults (Epley Keysar, et
al., 2004; Epley, Morewedge, & Keysar, 2004; Keysar & Barr,
2002).
Reducing either the motivation or cognitive capacity to engage
in this effortful correction process increases egocentric biases in
judgment, whereas increasing the motivation or capacity to engage
868
EPLEY, WAYTZ, AND CACIOPPO
in effortful correction reduces egocentric biases (Epley, Keysar, et
al., 2004). This suggests that people reason about the mental states
of others through a process of egocentric simulation and correct
that simulation to incorporate abstract knowledge about the mental
states of others only when they have the capacity, motivation, and
requisite representations about others to do so (see also Meltzoff &
Brooks, 2001). Such a correction or adjustment process is likely to
operate on a test– operate–test– exit fashion (G. A. Miller, Ga-
lanter, & Pribram, 1960), in which one corrects or adjusts a default
judgment some amount and then tests whether the adjusted value
if plausible. If plausible, adjustment terminates. If not plausible,
another adjustment away from the default is made, its plausibility
assessed, and so on until a satisfactory value is reached (Epley &
Gilovich, 2006).
When reasoning about the mental states of nonhuman agents,
this simulation process is likely to take on a similar form, with
people using themselves as a guide and only correcting that ego-
centric simulation when they have the motivation, capacity, and
requisite nonegocentric information to do so. Our model therefore
predicts that variability in the acquisition, activation, adjustment,
and subsequent application of these egocentric or homocentric
defaults in induction should therefore influence the tendency to
anthropomorphize nonhuman agents. In the next sections we de-
rive predictions of such variability from dispositional, situational,
developmental, and cultural influences.
Dispositional Influences: Need for Cognition
The ready accessibility of self-knowledge and one’s own phe-
nomenology makes an anthropomorphic inference a likely intui-
tive anchor or starting point when reasoning about nonhuman
agents, and correction of this anchor is possible to the extent that
people are motivated and able to do so. This implies that chronic
differences in the extent to which people are motivated to expend
such attentional resources should influence the extent to which
people anthropomorphize nonhuman agents. Such stable and
chronic differences are indexed as a person’s need for cognition.
Those who are high in need for cognition tend to enjoy engaging
in effortful thinking and are more likely to overcome readily
accessible defaults in judgment more than those low in need for
cognition (Cacioppo, Petty, Feinstein, & Jarvis, 1996; D’Agostino
& Fincher-Kiefer, 1992; Epley & Gilovich, 2005).
Engaging in more effortful processing when considering other
agents should lead to less reliance on readily accessible egocentric
or anthropomorphic information and instead lead to an increased
activation of alternate representations that may be more applicable
to a given stimulus through a process of correction or integration.
Indeed, those high in need for cognition tend to respond equally
quickly when answering questions about themselves versus others,
whereas those low in need for cognition are markedly faster to
answer questions about themselves than to answer questions about
others (Mueller, Haupt, & Grove, 1988). These findings suggest
that those high in need for cognition should show weaker evidence
of anthropomorphism, assuming the presence of alternate nonan-
thropomorphic representations exists and can be accessed.
Situational Influences: Perceived Similarity
People tend to rely on egocentric knowledge more heavily when
reasoning about other humans when the target appears similar to
the self. When the target appears dissimilar, people instead rely on
alternate forms of information to make inferences about their
mental states, such as stereotypes (Ames, 2004). This is consistent
with our SEEK model of anthropomorphism, in which the per-
ceived similarity of targets to humans should likewise influence
the extent to which people anthropomorphize nonhuman agents.
Any apparent similarity of the stimulus with one’s concept of the
self or of human implies that unknown properties of the stimulus
should, in Rips’s (1975) terms, “mirror the distribution” (p. 665) of
other properties known to be possessed by humans. Readily ob-
servable humanlike features should therefore influence the acces-
sibility of egocentric or anthropomorphic knowledge structures,
thereby increasing the likelihood that such knowledge is applied to
a nonhuman target of judgment (Mussweiler, 2003).
At least two dimensions of similarity seem particularly impor-
tant for anthropomorphism to occur—similarity in motion and
similarity in morphology. Children as young as 12 months appear
to understand that autonomous motion is a defining feature of
living agents (Poulin-Dubois, Lepage, & Ferland, 1996), and chil-
dren until at least age 5 use motion to determine whether a
stimulus is alive (Richards & Siegler, 1986). Among adults, non-
human agents such as toy robots and amorphous blobs are attrib-
uted mental states when they move at speeds that approximate
normal human motion but appear mindless when they move either
faster or slower than normal human motion (Morewedge et al.,
2007). Plants, for instance, appear to move intentionally toward the
sun only when time-lapsed photography gives their movement a
humanlike speed (Morewedge et al., 2007). And hummingbirds
suddenly appear more deliberate and thoughtful when their natural
quickness (and apparent mindless behavior) is slowed to a human-
like speed. Motion that appears biologically driven (Dittrich &
Lea, 1994) or socially interactive (Bassili, 1976; Heider & Simmel,
1944; S. C. Johnson, Slaughter, & Carey, 1998; Michotte, 1946/
1963) even influences anthropomorphic descriptions of such sim-
ple stimuli as two-dimensional shapes and letters.
Morphological similarity is simply the extent to which a non-
human agent’s observable features look humanlike. The more
similar in appearance, the more people are likely to use themselves
as a source of induction and anthropomorphize these nonhuman
agents. Children as young as 9 months, for instance, behave as if
they attribute intentions to an action when it is performed by a
humanlike hand, but not when performed by a wooden rod (Wood-
ward, 1999). Robots and other mechanical devices are also anthro-
pomorphized more readily when given humanlike faces and bodies
(Burgoon et al., 2000; DiSalvo, Gemperle, Forlizzi, & Kiesler,
2002). Marketers design anthropomorphic products and characters
in advertisements to elicit desired emotions from consumers, to
convey specific social relations, and to increase sales (Welsh,
2006). And nonhuman agents in literature and in everyday dis-
course are attributed mental states consistent with their morpho-
logical similarity to humans (Guthrie, 1993). Judgments of simi-
larity therefore influence anthropomorphism by activating human-
or self-relevant representations or by activating specific stereo-
types associated with humans. The importance of morphological
similarity for attributing mental states to other agents appears early
in human life and does not appear to diminish with age (S. C.
Johnson, 2003).
869
ANTHROPOMORPHISM
Developmental Influences: Acquisition of Alternate
Theories
Assessing the similarity of a nonhuman agent to humans or to
the self requires that assessors have a sense of humans or self to
begin with and do not already have a fully elaborated representa-
tion of the nonhuman agent to use in place of induction. Such
representations are at least partly acquired through a process of
direct or indirect learning and are therefore subject to a variety of
developmental influences. As already mentioned, most children
appear to reason egocentrically before developing the tendency to
reason from another person’s perspective, and some have further
argued that children also develop a conceptual understanding of
the self before an understanding of others (e.g., Chandler, 1977; P.
Harris, 1992; Piaget, 1929, 1932, 1962, 1969; Piaget & Inhelder,
1948; Shantz, 1975; Tomasello, 1999). Most important for the
present theoretical account is simply that the self clearly emerges
in most children as an elaborate and early guide for inductive
reasoning about other agents.
Not all children, however, develop these representations, and
perturbations in the development of knowledge about the self and
others provide insights into the process of anthropomorphism
itself. Autistic individuals who have difficulty acquiring the un-
derstanding that other humans have mental states at all tend to
describe commonly anthropomorphized objects in purely mechan-
ical terms (Castelli, Frith, Happe´, & Frith, 2002). Acquiring the
representation of others and the self as mental agents—a theory of
mind—appears critical for anthropomorphism to occur. What is
more, individuals with autism also do not appear to use their own
beliefs egocentrically as a guide to others’ beliefs (e.g., Baron-
Cohen, Leslie, & Frith, 1985; Baron-Cohen, Tager-Flusberg, &
Cohen, 1993), a finding that may be due to difficulties representing
the self and problems introspecting about one’s own mental states
(Carruthers, 1996; Pennington & Ozonoff, 1996). The lack of
sophisticated theories about the self, about others, and about
human-typical mental capacities demonstrated in people with au-
tism seem to at least partly explain their lack of anthropomorphism
in descriptions of animated stimuli. Research on this population
suggests that a person cannot project properties typical of humans
onto nonhuman agents without first developing representations of
oneself and of others.
Not only does development bring an understanding of the self
and other humans, but it also brings with it an understanding of
nonhuman agents. Because social experience in early human life is
primarily with other human agents (including the self), understand-
ing of other agents in nonanthropomorphic terms should develop
later in life. Indeed, young children show a particular penchant for
attributing life and mental states to nearly all stimuli in the envi-
ronment, from angry clouds to a happy sun to a stuffed animal that
is treated as a living companion (Piaget, 1929). Such extreme
anthropomorphism is not shown by most adults primarily because
they acquire alternate representations of these agents either
through direct or indirect experience and are therefore more likely
to use these representations when reasoning about nonhuman
agents. Anthropomorphism should therefore vary over the course
of development as a function of the representation of the self, other
humans, and nonhuman agents, being most anthropomorphic early
in life as representations of the self and others are learned and then
diminishing as alternate representations of nonhuman agents are
acquired.
Notice, however, that the development of alternate representa-
tions does not necessarily mean that adults will outgrow their
childish ways altogether. For instance, adults tend to show con-
siderably weaker egocentric biases in judgment than do children,
but this difference appears to arise because adults are more likely
to correct an automatic egocentric interpretation of the world than
are children rather than because adults are less likely to start with
an egocentric interpretation to begin with (Epley, Morewedge, &
Keysar, 2004). The reduction in anthropomorphism that comes
with age may similarly reflect the deliberate correction of an
automatic anthropomorphic interpretation, rather than a change in
the general extent to which adults are likely to anthropomorphize
nonhuman agents per se. Consistent with this possibility, adults’
explicitly reported beliefs about God tend to differ quite markedly
from their intuitively recalled memories of God’s actions. In
Protestant religions, for instance, God is explicitly described as
disembodied, unbound by space and time, and omniscient, all
clearly nonanthropomorphic traits. People’s explicit descriptions
of God tend to match these beliefs, but their more informal
descriptions of God’s actual behavior show strong evidence of
anthropomorphism, with God moving from one place to another,
answering prayers in a serial fashion, or being unaware of certain
events in the world (J. L. Barrett & Keil, 1996). Religious believ-
ers’ intuitive and automatic inferences about God’s nature appear
to be considerably more anthropomorphic than their deliberate and
explicit reports. People may acquire nonanthropomorphic repre-
sentations and knowledge about nonhuman agents ranging from
dogs to gods to electronic gadgets, but this does not guarantee that
people will use them.
Cultural Influences: Experience, Norms, and Ideologies
Culture influences anthropomorphism, we predict, in much the
same way as development by influencing representations of the
self, of other humans, and of nonhuman agents. Culture therefore
alters both the acquisition and accessibility of egocentric or ho-
mocentric representations during the process of inductive infer-
ence. Such cultural influence appears to emerge fairly quickly, as
children as young as 4 years from Christian backgrounds differ-
entiate between the capabilities of God, humans, and dogs
(J. L.Barrett, Newman, & Richert, 2003). Similar results emerge
among Mayan children 4 –7 years old for culturally prescribed
beliefs about God versus humans (Knight, Sousa, Barrett, & Atran,
2004).
Culture may exert its influence by providing distinct norms and
ideologies about how people relate to others and the natural world,
or by influencing the general level of experience with particular
nonhuman agents and the acquisition of nonanthropomorphic rep-
resentations. Nonindustrialized cultures tend to rely more closely
on the natural world for their daily needs, and members of those
cultures would be more likely than members of more modern or
urbanized cultures to interact with and use nonhuman animals.
Members of nonindustrialized cultures are therefore more likely to
acquire independent representations of these nonhuman animals
and be less likely, according to our theory, to anthropomorphize
these agents than would members of modern cultures. Members of
modern cultures, in contrast, are more likely to interact with
870
EPLEY, WAYTZ, AND CACIOPPO
sophisticated mechanical devices (cars, computers, etc.) than are
members of nonindustrialized cultures and would therefore be
more likely to acquire nonanthropomorphic representations of
their inner workings and less likely to anthropomorphize these
mechanical agents. The fewer distinct cognitive representations
one has developed for nonhuman agents, the more likely repre-
sentations of the self or humans in general are to become activated
when encountering any nonhuman agent and the greater the like-
lihood that human attributes are projected toward the stimulus as a
result.
Indirect support for this hypothesis comes from separate lines of
research that have previously appeared inconsistent with each
other. One line demonstrates anthropomorphism when reasoning
about the biological functioning of nonhuman animals among
children from industrialized cultures (namely, the United States;
Carey, 1985). A second line demonstrates that children from less
developed and rural populations show relatively little anthropo-
morphism when reasoning about local nonhuman animals (Atran
et al., 2001; Ross, Medin, Coley, & Atran, 2003). Whereas indi-
viduals from industrialized cultures reasoned about nonhuman
animals on the basis of their knowledge of humans, individuals
from less industrialized cultures reasoned about nonhuman ani-
mals using their more advanced knowledge of the animal world.
These results are consistent with the accessibility of agent repre-
sentations influencing anthropomorphism (Medin & Atran, 2004)
and demonstrate how culture can influence anthropomorphism.
Although a cognitive representation of humans might be equally
accessible for individuals from both urban industrialized and rural
underdeveloped populations, the extent to which this concept is
deemed applicable for reasoning about a nonhuman agent may
differ. Differences in perceived applicability of this concept may
then account for cross-cultural differences in nearly identical pro-
cesses of inductive reasoning about a nonhuman animal. The
presence of alternative knowledge structures for the perceiver
appears to explain at least some of the cross-cultural differences
that emerge in reasoning about nonhuman animals on the basis of
knowledge about humans. When alternate knowledge structures
are accessible, individuals devote attention to judging their appli-
cability to a particular stimulus (Higgins, 1989). For children from
rural cultures, knowledge of multiple biological kinds (additional
to humans) may be more readily accessible and deemed applicable
for reasoning about nonhuman animals. For urban industrialized
children, knowledge of biological kinds is more limited, and so
these children instead rely on the readily accessible concept of
human agency to reason about nonhuman animals.
Motivational Determinants of Anthropomorphism
We propose that anthropomorphism is, at its core, a process of
induction that utilizes existing knowledge representations to guide
inferences about the properties, characteristics, and mental states
of nonhuman agents. We suggest that this inductive process of
anthropomorphism can also be substantially influenced by two
major motivational factors. The first is effectance—the motivation
to interact effectively in one’s environment. Effectance motivation
entails understanding, predicting, and reducing uncertainty about
one’s environment and the agents that inhabit it. The second is
sociality—the motivation for social contact, social connection, and
social approval from other agents (human or otherwise). We dis-
cuss each of these motivational determinants in turn, along with
the predictions made by each regarding dispositional, situational,
developmental, and cultural sources of variability for anthropo-
morphism.
The critical distinction between motivational and cognitive de-
terminants in our theory is that motivational determinants operate
like typical drive states, increasing in strength as a function of
deprivation and decreasing when the drive is satisfied, whereas
cognitive determinants operate off typical principles of cognitive
activation, increasing in strength at the point of activation and then
decreasing over time. Like hunger or thirst, depriving people of a
sense of efficacy or social connection should activate a goal to
repair or decrease this deprived state. This goal should then remain
active, or even increase in strength over time, until the goal is
satisfied (Bargh, Gollwitzer, Lee-Chai, Barndollar, & Tro¨tschel,
2001; Heckhausen, 1991; Wicklund & Gollwitzer, 1982). Motiva-
tional states can influence basic cognitive operations such as
attention and cognitive accessibility and can therefore indepen-
dently influence the higher order cognitive processes involved in
social cognition (Maner et al., 2005). Anthropomorphism, we
suggest, serves as one way of satisfying a goal to feel efficacious
in one’s environment, or to increase a sense of social connection,
and should therefore increase as a function of these two motiva-
tional states.
Effectance Motivation as a Determinant of
Anthropomorphism
Developmental psychologists (e.g., Gopnik, Meltzoff, & Kuhl,
2001) have made it patently clear that William James (1890/1950)
underestimated the faculties of human infants when he suggested
that their first sensory experiences were a “blooming, buzzing
confusion” (p. 442). But what James’s sentiment did capture is the
overwhelming complexity and uncertainty that exists in the child’s
environment and the inherent difficulty in making sense of that
complexity from scratch. From natural phenomena such as the
weather and tides, to biological phenomena such as death and
reproduction, to social phenomena such as the behavior of human
and nonhuman agents, observations of the external world are
fraught with uncertainty and ambiguity that must be explained in
order to operate effectively in one’s environment by children and
adults alike. Following the Hurricane Katrina disaster of 2005, for
instance, New Orleans Mayor Ray Nagin explained the city’s
misfortune by arguing, “Surely God is mad at America. Surely
he’s not approving of us being in Iraq under false pretense. But
surely he’s upset at Black America, too. We’re not taking care of
ourselves” (Martel, 2006, p. A04). Although Nagin later retracted
these statements under political pressure, the intuitive attempt to
make sense of this natural disaster highlights the possible appeal of
anthropomorphism to reduce uncertainty and increase comprehen-
sion of events in one’s environment.
Anthropomorphism provides an intuitive and readily accessible
method for reducing uncertainty in contexts in which alternative
nonanthropomorphic models of agency do not exist (such as those
provided by science or culture). Charles Darwin, for instance,
advocated anthropomorphism as a necessary tool for understand-
ing nonhuman agents. So too did Donald Hebb (1946), a seminal
contributor to cognitive psychology and neuroscience, who argued
871
ANTHROPOMORPHISM
for anthropomorphism as a procedural tool in his laboratory stud-
ies of chimpanzees as a practical aid to understanding:
A thoroughgoing attempt to avoid anthropomorphic description in the
study of temperament was made over a two-year period at the Yerkes
laboratories. All that resulted was an almost endless series of specific
acts in which no order or meaning could be found. On the other hand,
by the use of frankly anthropomorphic concepts of emotion and
attitude one could quickly and easily describe the peculiarities of
individual animals . . . . Whatever the anthropomorphic terminology
may seem to imply about conscious states in chimpanzee, it provides
an intelligible and practical guide to behavior. (p. 88)
Notice that Hebb’s experience suggests that anthropomorphism
can aid understanding regardless of its accuracy by serving a more
utilitarian function. This utilitarian function is perhaps best de-
scribed by Dennett (1987) as the “intentional stance,” whereby
clearly unintentional agents (such as thermostats) are attributed
humanlike intentions simply to increase the ease with which peo-
ple can reason about those agents or communicate about them with
others, and thereby interact with them more effectively. Indeed,
Kiesler and Goetz (2002) have argued that an anthropomorphic
model for robots may help advance knowledge in the human–
computer interaction domain just as anthropomorphic theories
have arguably enhanced psychological studies of nonhuman ani-
mals.
These arguments suggest that anthropomorphism may be uti-
lized to increase the predictability and comprehension of what
would otherwise be an uncertain world, much in the way that
egocentric knowledge about one’s own preferences can serve as a
useful guide to another person’s preference in the absence of any
additional information (Dawes & Mulford, 1996). Anthropomor-
phism should therefore be influenced by the human motivation to
resolve uncertainty, seek meaning, and feel efficacious. This gen-
eral motivation to “interact effectively with [one’s] environment”
(White, 1959, p. 297) deemed effectance motivation by White,
suggests that humans are driven to master their environments (e.g.,
Harter, 1978) by increasing the environment’s predictability (e.g.,
Averill, 1973; Lazarus, 1966) and apparent controllability (e.g.,
Burger & Cooper, 1979; Rothbaum, Weisz, & Snyder, 1982;
Rotter, 1966). Knowledge about the self in particular, or humans in
general, can serve as a rich representation for gaining a sense of
predictability and controllability when reasoning about nonhuman
agents. Indeed, induction itself has been conceptualized as satis-
fying one’s effectance motivation by including “all inferential
processes that expand knowledge in the face of uncertainty” (Hol-
land et al., 1986, p. 1). Anthropomorphism, as a specialized
process of induction, should therefore be influenced by one’s
effectance motivation. When effectance motivation is high, anthro-
pomorphism should increase. When effectance motivation is low,
anthropomorphism should decrease.
These predictions are in line with at least some classic philo-
sophical and historical accounts of anthropomorphism in religious
agents and supernatural explanations for everyday phenomena
(e.g., Feuerbach, 1873/2004; Freud, 1927/1961; Hume, 1757/
1956; Marx, 1844/1959). The tendency to see humanlike figures
among the constellations, humanlike religious agents guiding
weather patterns, or humanlike ghosts and spirits causing madness
and psychological dysfunction has long been explained as a logical
attempt by humans to understand and predict the complicated
world around them using existing knowledge about human agents
(see Preus, 1987). “The whole frame of nature bespeaks an intel-
ligent author,” Hume (1757/1956, p. 21) observed, an inference
many have suggested arises intuitively from the sheer complexity
and diversity of life on the planet. It is almost certainly no accident
that the intelligent author in many religious systems resembles an
extremely intelligent human.
The predicted role of effectance motivation in the context of
nonhuman agents, however, goes beyond these intuitive observa-
tions and suggests that anthropomorphism should be heightened by
two major factors. First, anthropomorphism should be heightened
by uncertainty regarding the behavior of a real or presumed non-
human agent, usually activated by observing the agent’s behavior
or by observing behavior attributed to a nonhuman agent (such as
with natural phenomena attributed to religious agents). This un-
certainty may arise because the agent is novel and unknown,
because the agent appears unpredictable, because the agent vio-
lates one’s expectations, or because the causal mechanisms under-
lying an observed behavior are unknown or unobservable. Second,
by the incentives associated with accurately understanding or
predicting the behavior of a nonhuman agent. When incentives for
understanding and predictability are high, anthropomorphism
should increase. When incentives are low, anthropomorphism
should decrease. Agents perceived as threatening or able to influ-
ence one’s welfare, for instance, may be anthropomorphized more
readily than nonthreatening or powerless agents. Agents one is
likely to interact with in the future are likely to be anthropomor-
phized more readily than agents unlikely to be seen again. These
two main factors make specific predictions regarding dispositional,
situational, development, and cultural influences on the role of
effectance motivation in anthropomorphism.
Dispositional influences: Need for closure and desire for con-
trol. Effectance motivation entails the desire to reduce uncer-
tainty and ambiguity, at least in part with the goal of attaining a
sense of predictability and control in one’s environment. Stable
individual differences have been identified and validated for both
of these features of effectance motivation, and both should influ-
ence the tendency to anthropomorphize nonhuman agents.
The need for closure (e.g., Dittes, 1961; Kruglanski & Mayse-
less, 1988; Kruglanski & Webster, 1996) indexes stable individual
differences in the extent to which a person desires “an answer on
a given topic . . . compared to confusion and ambiguity” (Kruglan-
ski, 1990, p. 337; Kruglanski & Webster, 1991). Those high in
need for closure tend to demonstrate primacy effects in forming
impressions of stimuli, relying on early information in making
judgments of other people instead of relying on more recent
information (e.g., Kruglanski & Freund, 1983; Kruglanski & Web-
ster, 1991) and tend to seize on immediately accessible informa-
tion when making judgments about others in an attempt to reach a
quick resolution (Kruglanski & Webster, 1996). This tendency
toward primacy in impression formation suggests that need for
closure indexes a drive toward attaining a sense of comprehension
and understanding within a situation. Coupled with the tendency to
seize upon information immediately accessible in judgment, this
motivation for comprehension and understanding suggests that
those high in need for closure should be more likely to activate
anthropomorphic representations and less likely to subsequently
correct those representations when making an inference about a
872
EPLEY, WAYTZ, AND CACIOPPO
nonhuman agent. Individuals high in need for closure may there-
fore be particularly inclined to anthropomorphism.
Similarly, individuals differ consistently in their desire for con-
trol—the “extent to which people generally are motivated to see
themselves in control of the events in their lives” (Burger, 1992, p.
6)—and this variable should motivate anthropomorphism as a
means both to organize the present and to establish predictability
in future interactions with a nonhuman agent. Those with a strong
desire for control exhibit more vigorous attributional activities to
explain others’ behavior, usually focusing on typically anthropo-
morphic concepts such as intentions and desires (Burger & He-
mans, 1988; Liu & Steele, 1986; Pittman & Pittman, 1980). What
is more, this attributional activity appears to increase the feeling of
efficacy in social interaction. Desire for control should therefore
facilitate the activation and application of anthropomorphic repre-
sentations in the service of attaining both an explanation for an
agent’s behavior in the present and an increased sense of predict-
ability for an agent’s behavior in the future.
Situational influences: Anticipated future interaction and ap-
parent predictability. The motivation to interact effectively with
agents in one’s environment should be activated primarily for
agents one actually expects to encounter in the environment. The
likelihood of future interaction should therefore serve as a situa-
tional influence on anthropomorphism because it alters the incen-
tives associated with understanding an agent in the present and
predicting its behavior in the future and increases the likelihood of
activating existing anthropomorphic representations when reason-
ing about nonhuman agents. Agents one is expected to interact
with in the future should be anthropomorphized more extensively
than agents one never expects to see again.
When reasoning about other humans, anticipated future interac-
tion increases the amount of information people seek about others
(Berger & Douglas, 1981; Berscheid, Graziano, Monson, &
Dermer, 1976; Kellerman & Reynolds, 1990), increases the like-
lihood of making dispositional inferences about others (Feldman &
Ruble, 1981; D. T. Miller, Norman, & Wright, 1978), increases the
egocentric projection of one’s beliefs and attitudes onto others (N.
Miller & Marks, 1982), and generally increases people’s interests
in knowing others’ thoughts and motivations (Douglas, 1990). All
of these results are consistent with future interaction increasing the
tendency to seek additional understanding about an individual and
a tendency to activate and apply accessible (egocentric) knowledge
structures to enable that understanding. When reasoning about
nonhuman agents, this additional information is likely to focus on
the agents’ intentions, goals, emotional states, and underlying
personality characteristics to gain a sense of predictability and
understanding (Gilbert, 1998; Heider & Simmel, 1944), thereby
increasing the extent to which people anthropomorphize these
nonhuman agents.
In addition to altering the incentives associated with effective
interaction, situational variables that introduce uncertainty regard-
ing a nonhuman agent should also stimulate effectance motivation
and increase anthropomorphism. There are a variety of situational
factors that may increase uncertainty, but perhaps the most com-
mon is when agents violate one’s expectations. Such expectancy
violations require one to rethink preexisting beliefs about an
agent’s behavior or mental states and are likely to stimulate an-
thropomorphic thoughts about an agent’s underlying intentions,
motivations, or goals—the very features commonly associated
with anthropomorphism. When reasoning about other humans,
expectancy violations can activate counterfactual thinking (Sanna
& Turley, 1996) and tend to increase the amount of information
people seek regarding a target person (Russell, Buckworth, Case,
& Upshaw, 1977).
Although expectancy violations do not necessitate an increase in
anthropomorphic thinking if people are anthropomorphizing a
nonhuman agent already, expectancy violations are likely to in-
crease the attention paid to a nonhuman agent’s goals, intentions,
or underlying motivations and therefore produce a general increase
in anthropomorphism. One’s dog may not be anthropomorphized
until it violates a routine command and begins behaving less like
an agent under one’s own control and more like an agent guided by
its own goals or intentions. Or one’s computer may seem like a
complicated but mindless device when it enables word processing
exactly as it is programmed to do but may quickly acquire malev-
olent intentions and motives the moment it crashes or freezes
unexpectedly. Indeed, the more participants in one experiment
reported their computers to malfunction, the more willing they
were to attribute mental states, beliefs, and desires to these com-
puters (Morewedge, 2007). In another experiment, participants
were more likely to attribute intentionality to the movement of
marbles when they were unable to control them (with a magnet)
than when they were able to control them (J. L. Barrett & Johnson,
2003). Humanlike mental states are unnecessary to explain an
agent that behaves perfectly predictably or is controllable but may
be activated and applied to make sense of an agent once it starts
behaving unpredictably.
More direct support for this prediction comes from two recent
experiments investigating anthropomorphism directly (Epley,
Waytz, Akalis, & Cacioppo, in press). In one, 132 participants
were asked to watch a 30-s video of and evaluate the behavior of
two different dogs, one of which had been rated as significantly
more unpredictable and uncontrollable than the other by an inde-
pendent sample of 54 participants. Participants watched this video
twice, rating one dog after watching the video the first time and the
other dog after watching it the second time. Participants then
evaluated both dogs on four items related to anthropomorphism:
the extent to which each dog is “aware of its emotions,” has “a
conscious will,” and has “a personality” and an item asking them
to rate the dog on its similarity to other life forms with one side of
the scale representing “bacteria” and the other representing “hu-
mans.” Participants also completed a scale to measure their “desire
for control” (Burger & Cooper, 1979). These items were highly
intercorrelated and therefore standardized and averaged together to
create a single composite measure. As expected, participants rated
the unpredictable dog as significantly more humanlike on this
composite measure than the predictable dog, and those high in
desire for control rated the dogs as more humanlike than those low
in desire for control. These two main effects were qualified by a
marginally significant interaction ( p .052), demonstrating that
those high in desire for control were especially likely to anthro-
pomorphize the unpredictable dog.
A second experiment extended these findings by measuring the
anthropomorphism of electronic gadgets (Waytz, Cacioppo, &
Epley, 2007), by manipulating predictability independent of the
particular stimulus, and by asking participants to rate the gadgets
on both anthropomorphic qualities and nonanthropomorphic qual-
ities to demonstrate that effectance motivation influences anthro-
873
ANTHROPOMORPHISM
pomorphism specifically rather than any rating of a nonhuman
agent. In this experiment, 32 participants evaluated 30 different
consumer devices on five items measuring anthropomorphism
(“To what extent does [this device] have intentions, experience
emotion, have free will, have consciousness, and have a mind of its
own?”) and three behavioral descriptors that do not reflect anthro-
pomorphism “To what extent is [this device] attractive, efficient,
and strong?”). These devices (such as Clocky, a fuzzy alarm clock
on small wheels) were described either as being relatively predict-
able and controllable (e.g., “You can program Clocky so that when
you press snooze it runs away from you or you can program it so
that when you press snooze, it will jump on top of you.”) or as
being relatively unpredictable and uncontrollable (e.g., “When you
press snooze, Clocky either runs away from you, or it jumps on top
of you . . . [and is] unpredictable in this way.”). Participants read
the predictable description for half of the devices and the unpre-
dictable description for the other half (counterbalanced across
participants).
As expected, participants rated the devices significantly higher
on the anthropomorphism measures when they were described as
unpredictable than when they were described as predictable. This
pattern did not emerge on any of the single nonanthropomorphic
ratings or on the composite average of all three ratings. This
difference produced the predicted significant interaction, demon-
strating that predictability only influenced ratings on the anthro-
pomorphic items in the expected direction. Both predictability and
control are related to effectance motivation, and agents that pos-
sess neither of these appear to induce anthropomorphism.
The one limitation to this general prediction comes from more
extreme cases in which an agent is so unpredictable that it appears
completely random and therefore mindless. However, we suspect
that such cases of extreme unpredictability are rather rare given
people’s well-documented penchant for providing nonrandom and
systematic accounts of truly random behavior, such as random
shooting in a basketball game being ascribed to a systematic “hot
hand” (Gilovich, Vallone, & Tversky, 1985) or the random roll of
a dice to any number of superstitious mechanisms (e.g., Heltzer &
Vyse, 1994; Vyse, 1997; Zusne & Jones, 1982).
Developmental influences: Attaining competence. If effec-
tance motivation entails the desire to interact effectively with one’s
environment by increasing comprehension in the present and a
sense of predictability and control for the future, then the strength
of this motivation should be influenced both by the extent to which
one experiences a sense of comprehension of the environment and
the incentives associated with accurate predictions and by effective
control. Both of these are likely to vary systematically over one’s
lifespan.
Children may not be in the midst of a “blooming, buzzing,
confusion,” but few will debate that infants have more to learn
about their external environment than adults. Children appear more
likely to anthropomorphize than adults (e.g., Bering & Bjorklund,
2004), and we suggest that children’s penchant for anthropomor-
phism is at least partly a function of their heightened effectance
motivation. As White (1959) suggested, young children are often
“occupied with the agreeable task of developing an effective
familiarity with his environment . . . build[ing] up an increased
competence in dealing with the environment” (p. 321) by con-
stantly exploring and manipulating their surroundings (Harlow,
1953; Piaget, 1952; Woodworth, 1958). While attempting to gain
this familiarity and competence, children attribute intentions and
causal agency widely to the simplest and most abstract of nonhu-
man stimuli (Dasser, Ulbaek, & Premack, 1989; Gergely, Na´-
dasdy, Csibra, & ´ro´, 1995; S. C. Johnson, Booth, & O’Hearn,
2001; Leslie, 1982; Leslie & Keeble, 1987; Scholl & Tremoulet,
2000; Shimizu & Johnson, 2004). Because ascribing these mental
characteristics aids children’s attempts to make sense of a wide
variety of stimuli, those who have not yet attained a full sense of
competence with their environment should be particularly likely to
activate anthropomorphic representations and less likely to subse-
quently correct or adjust those representations. Anthropomorphism
should therefore be especially helpful in these early stages of life
as a means of reducing uncertainty.
The opposite pattern of effectance motivation may emerge,
however, for the incentives associated with predicting and control-
ling ones’ environment. As children become more independent and
in control of their own and others’ actions, the need to maintain a
sense of predictability and control may increase with age. This
increase would suggest that effectance motivation would be more
readily activated among adults than children and that anthropo-
morphism of nonhuman agents would therefore increase as well.
This speculation suggests that anthropomorphism may occur in
daily life at reasonably high levels for children and adults, albeit
for different reasons. Children may be motivated to attain com-
prehension and understanding more often than adults, and adults
may be more motivated to attain a sense of predictability and
control overall than children. We offer this predicted interaction
between uncertainty and control as a tentative prediction that is
clearly in need of empirical investigation.
Cultural influences: Uncertainty avoidance. As mentioned
earlier, culture influences anthropomorphism through the same
mechanisms as do developmental influences, but the specific in-
dependent variables that manipulate effectance motivation at the
cultural level differ from those at the individual level. At the
cultural level, uncertainty avoidance (akin to individual need for
closure) was one of the five unique value dimensions to emerge
from factor analyses of over 116,000 questionnaire responses from
66 nationalities in 50 different countries (Hofstede, 1980, 2001).
Uncertainty avoidance represents “the extent to which the mem-
bers of a culture feel threatened by uncertain or unknown situa-
tions” (Hofstede, 2001, p. 161) and is tightly linked to effectance
motivation. This cross-cultural variable should therefore influence
anthropomorphism in the same manner that need for closure in-
fluences individual-level responses. Namely, individuals from cul-
tures that score high on this dimension should be more prone to
activate and apply anthropomorphic representations as a means to
establish comprehensibility and predictability than those from cul-
tures that score low. People in cultures high in uncertainty avoid-
ance “look for structure . . . which makes events clearly interpret-
able and predictable” (Hofstede, 2001, p. 148), and
anthropomorphism is likely to be one readily available method of
attaining this structure and reducing the anxiety from uncertainty.
Findings at least consistent with this hypothesis come from
differences in the religious traditions of cultures high versus low in
uncertainty avoidance. Cultures high in uncertainty avoidance, for
instance, tend to believe more strongly in the theological concept
of immanence—the extent to which a god is “incorporated, is
immanent, in persons, organizations, or various objects in the
natural world” (Swanson, 1967, p. 1)—that reflects a more anthro-
874
EPLEY, WAYTZ, AND CACIOPPO
pomorphic and embodied view of religious agents. Immanence
implies the possibility of a personal and intimate relationship with
religious agents for those who devote themselves to their faith and
implies a more tangible image of religious agents. This doctrine of
immanence is explicitly endorsed by Catholicism but not by Prot-
estantism, and all 20 of the Christian countries above the median
in uncertainty avoidance in Hofstede’s (2001) analysis were pre-
dominantly Catholic. In contrast, all but 3 of the low uncertainty
avoidant Christian countries in Hofstede’s analysis were predom-
inantly Protestant or mixed. Although it is nearly impossible to
find a religion that doesn’t endorse some anthropomorphic notion
of religious agents (Guthrie, 1993), these anthropomorphic beliefs
exist on a continuum, and cultural values that endorse certainty and
an avoidance of ambiguity should motivate depictions of a more
humanlike religious agent. The same should be true of more
secular nonhuman agents as well.
Additional suggestive evidence also comes from apparent cul-
tural differences in anthropomorphism between Japanese and
American scientists (Asquith, 1996), countries ranked, respec-
tively, in the top 10 and bottom 10 of Hofstede’s (2001) uncer-
tainty avoidance index. American scientists, for instance, have
criticized Japanese researchers for their highly anthropomorphic
descriptions of primates (Asquith, 1986; de Waal, 2003). Although
these differences may arise for a variety of reasons, it is at least
possible that cultural differences in uncertainty avoidance between
these two groups of sciences were playing an important role, and
the Japanese effort to anthropomorphize was a deliberate effort to
increase predictability and understanding of nonhuman primates as
both Darwin and Hebb suggested was practically useful to do.
Sociality Motivation as a Determinant of
Anthropomorphism
Aristotle noted some time ago that “man is a social animal,” and
those attempting to understand the human condition in the inter-
vening centuries have not simply echoed this basic sentiment but
have enhanced it (e.g., Bowlby, 1969, 1973; Freud, 1930/1989;
Maslow, 1954, 1968). Establishing and maintaining a sense of
social connection with others demonstrates all the hallmarks of a
basic human need akin to hunger or thirst—it appears to be
universal across individuals and cultures, is evident at birth, cre-
ates strong negative affect and adverse health outcomes when
thwarted, and elicits a search for social connection when one is
deprived or excluded from it (for a review, see Baumeister &
Leary, 1995). For nearly all of human evolutionary history, being
shunned or ostracized by other humans was tantamount to a death
sentence for one’s genetic inheritance, either because an individual
was unable to survive or was unable to reproduce. Possessing a
keen sensitivity to one’s social relations, and a strong motivation
to establish and maintain social connections, therefore confers
obvious survival and reproductive benefits (Axelrod & Hamilton,
1981; Buss, 1991; Cacioppo, Hawkley, et al., 2006; Harlow, 1958;
Kling, Lancaster, & Benitone, 1970).
This social monitoring system appears to work through some of
the very same neural mechanisms as physical pain, meaning that
depriving a person of social connection hurts in much the same
way as physical pain (MacDonald & Leary, 2005). What is more,
social pain also leads people to actively search for effective ways
to alleviate this pain, in much the same way as people seek to
alleviate physical pain. Those experiencing social pain from a lack
of social connection, compared with those not experiencing social
pain, spontaneously imagine important social relationships
(Twenge, Catanese, & Baumeister, 2003), focus their attention on
social symbols and reminders of loved ones such as photographs
and mementos (Gardner, Pickett, & Knowles, 2005), attend to
others’ social behavior more carefully (Pickett, Gardner, &
Knowles, 2004), and seek out new contacts with other people
(Maner, DeWall, Baumeister, & Schaller, 2007).
Importantly for anthropomorphism, this need for social connec-
tion appears to be satisfied by connections with two of the most
commonly anthropomorphized nonhuman agents, namely pets and
religious agents. Elderly pet owners, for instance, appear to be
buffered from the negative impact of stressful life events and visit
their doctor less often compared with elderly people without pets
(Siegel, 1990), and those diagnosed with AIDS are less likely to
become depressed if they own a pet than if they do not (Siegel,
Angulo, Detels, Wesch, & Mullen, 1999). In fact, the presence of
one’s pet in stressful evaluative circumstances can have a more
positive influence on reducing anxiety and psychophysical re-
sponses to stress than the presence of one’s spouse (Allen, Blas-
covich, & Mendes, 2002). So too do people form strong attach-
ments to religious agents, which take on the same attachment
characteristics as their relationships with other people (Birgegard
& Granqvist, 2004). Those who suffer a profound loss of social
connection, namely through the death of a spouse, tend to increase
their connection to religious agents, such as God (Brown, Nesse,
House, & Utz, 2004), and those who already feel a strong connec-
tion to a religious agent appear to be buffered from common
symptoms of social pain, including grief (McIntosh, Silver, &
Wortman, 1993; Walsh, King, Jones, Tookman, & Blizard, 2002),
depression (Azhar & Varma, 1995), and more general dissatisfac-
tion with life (McFadden, 1995).
This need to establish and maintain a sense of social connection
with others, and the apparent ease with which nonhuman agents
can satisfy this need, forms the basis of the second motivational
determinant of anthropomorphism. In particular, we suggest that
sociality motivation increases the tendency to anthropomorphize
nonhuman agents in two distinct ways. First, sociality motivation
increases the baseline accessibility of social cues including hu-
manlike traits and characteristics (Gardner, Pickett, Jefferis, &
Knowles, 2005), thereby increasing the tendency to perceive hu-
manlike characteristics and traits even in nonhuman agents. Con-
siderable evidence suggests that people selectively attend to
human-typical features from birth, showing a preference for face-
like stimuli compared with nonface stimuli (Mondloch et al.,
1999), and 4-month-olds prefer looking at motion displays that
resemble human walking compared with displays simulating ran-
dom motion (Bertenthal, 1993). Studies with older participants
also demonstrate a biased allocation of processing resources to-
ward social stimuli (Farah, Wilson, Drain, Tanaka, 1998; Ito &
Cacioppo, 2000). Perhaps more important, depriving people of a
sense of social connection increases attention to social cues in the
environment, presumably as a result of an increased interest in
forming new relationships (Pickett et al., 2004).
Second, sociality motivation increases the tendency to anthro-
pomorphize nonhuman agents by increasing the tendency to ac-
tively search for sources of social connection in one’s environ-
ment. A person feeling lonely, isolated, or lacking social
875
ANTHROPOMORPHISM
connection may attempt to recover from this social pain by an-
thropomorphizing nonhuman agents— essentially creating social
connection by making it up in nonhuman agents. This clever
method of turning ambiguous nonhuman agents into avenues for
social connection may help to alleviate social pain and a sense of
disconnection when one is most in need. Notice, however, that this
particular mechanism is unlikely to influence anthropomorphism
in general but rather increases anthropomorphic descriptions that
facilitate social connection more specifically. One may be more
likely to perceive a pet as thoughtful and considerate when one is
feeling alone and isolated but may be unlikely (or even less likely)
to also perceive the pet as vindictive and deceitful. Variation in
one’s sense of social connection stemming from dispositional,
situational, developmental, or cultural sources should therefore
influence anthropomorphism by altering the accessibility of
human-typical traits, by reducing the tendency to correct an intu-
itive anthropomorphic inference, or by increasing the tendency to
apply anthropomorphic representations to nonhuman agents.
Dispositional influences: Chronic loneliness. Although most
people are well acquainted with feelings of loneliness and social
isolation, some are more chronically acquainted than others
(Boomsma, Willemsen, Dolan, Hawkley, & Cacioppo, 2005; Ca-
cioppo, Hughes, Waite, Hawkley, & Thisted, 2006). All else equal,
those who are chronically lonely should be more likely to anthro-
pomorphize nonhuman agents than those who are more chronically
connected. For chronically lonely individuals who are routinely
seeking agents of social connection, anthropomorphic representa-
tions will be more readily accessible and thus more likely to be
activated and applied to nonhumans. This should be especially true
for novel nonhuman agents for whom the influence of chronic
loneliness should be the clearest. Although theories about anthro-
pomorphism and loneliness have existed since the time of Freud
(1927/1961), very little research has tested this dispositional pre-
diction directly.
In one exception (Epley, Waytz, et al., in press), 164 university
undergraduates were asked to rank order the traits that best de-
scribed their pet (or a friend’s pet, if they didn’t have one of their
own) from a list that included supportive anthropomorphic traits
(thoughtful, considerate, sympathetic), nonsupportive anthropo-
morphic traits (devious, embarrassable, jealous, creative), and be-
havioral traits (aggressive, agile, active, energetic, fearful, lethar-
gic, muscular). There was a small but statistically reliable (r
.18, p .02) negative relationship between one’s loneliness as
measured by the Revised UCLA Loneliness Scale and the average
rank of the supportive anthropomorphic traits. The more lonely
participants reported being, the higher they ranked the supportive
anthropomorphic traits as accurate descriptors of either their own
or a friend’s pet. There was no significant correlation (and slightly
in the opposite direction), however, between chronic loneliness
and rankings of either the nonsupportive anthropomorphic traits
(r .10, p .17) or the behavioral traits (r .07, p .35), and
none of these individual items approached significance. These data
are at the very least consistent with our theoretical predictions
about the relationship between sociality motivation and anthropo-
morphism.
Situational factors: Social disconnection. Hungry people seek
food. Thirsty people seek water. And socially disconnected people
seek social connection. As just discussed, the strength of these
drive states vary consistently from one person to another, but they
also vary from one moment to another. People feel hungry and
thirsty over time as they go without food and water. And people
become socially disconnected over time as they go without mean-
ingful social contact. Isolation, exclusion, or a feeling of discon-
nection from others is the hallmark of social pain, and people
experiencing this social pain take active steps toward alleviation
by seeking out meaningful social connections with others (Maner
et al., 2007). One method for regaining social connection, we have
suggested, is to anthropomorphize nonhuman agents, essentially
creating humans out of nonhumans. People may be especially
likely, then, to anthropomorphize nonhuman agents when they are
feeling socially disconnected.
Most readers will resonate with this prediction from daily life, as
stereotypes about the relationships between loneliness and anthro-
pomorphism are widespread. From the elderly person who treats
his or her cat as a bit too much like a spouse to cinematic
depictions such as Cast Away in which the shipwrecked protago-
nist (Tom Hanks) anthropomorphizes a volleyball (named Wilson)
after being marooned on an island, those who are lacking human
connection appear to seek it out in nonhuman connections. Louka-
tos (1976) even suggested that the personified names given to
coastal rocks and capes by Greek sailors were attempts to create
social contact in a context of extreme isolation. It may be no
coincidence that those who are deeply religious, such as monks
and priests, often go into isolation from other humans in order to
more effectively find God.
Consistent with these observations, experimental evidence sug-
gests that momentary feelings of social rejection or isolation can
increase the tendency to anthropomorphize one’s pet in a manner
that may effectively alleviate social disconnection and also in-
crease the extent which people report believing in anthropomor-
phized supernatural agents (Epley, Akalis, Waytz, & Cacioppo, in
press). In one experiment, participants were induced to feel lonely,
afraid, or neither by watching video clips intended to induce these
emotional experiences. Participants in the lonely condition were
asked to identify with the protagonist in Cast Away just at the
moment when he discovers his true isolation and loneliness. Par-
ticipants in the fear condition watched a fear-inducing clip from
the Silence of the Lambs. Participants in the control condition
watched a clip that was neither socially isolating nor fear inducing
from Major League. Participants then picked the three traits that
best described their own pet (or a well-known pet) out of a list that
included supportive anthropomorphic traits (thoughtful, consider-
ate, sympathetic), nonsupportive anthropomorphic traits (devious,
embarrassable, jealous, creative), and behavioral traits (aggressive,
agile, active, energetic, fearful, lethargic, muscular). Participants
also indicted the extent to which they believed in a set of com-
monly anthropomorphized religious agents (God, angels, ghosts,
and the devil).
Those induced to feel lonely, compared with those in both the
fear and control conditions, were more likely to select supportive
anthropomorphic traits to describe their pet, with no differences
emerging on either the nonsupportive anthropomorphic or behav-
ioral traits. This suggests that anthropomorphism of these pets,
when possible, may be oriented toward creating more socially
supportive agents that can especially alleviate the pain of social
disconnection. Participants induced to feel lonely also reported a
stronger belief in the entire set of religious agents than either those
induced to feel fear or those in the control condition. It is inter-
876
EPLEY, WAYTZ, AND CACIOPPO
esting to note that belief in these individual supernatural agents
was highly correlated, so people did not distinguish between
supportive supernatural agents (God, angels, perhaps ghosts) and
nonsupportive agents (the devil). These findings suggest that situ-
ationally activated isolation or loneliness can increase anthropo-
morphism in a manner that may alleviate social disconnection and
that these results are not simply attempts to repair a negative mood
because participants in the fear condition in this experiment looked
no different than participants in the control condition on these
measures.
These findings are consistent with naturally occurring incidents
that induce social disconnection, such as the loss of a spouse or
loved one. Such events are particularly likely to increase the
strength of one’s religious beliefs (e.g., Glick, Weiss, & Parks,
1974; McIntosh et al., 1993; Spilka, Hood, & Gorsuch, 1985;
Wuthnow, Christiano, & Kuzlowski, 1980), and those who are
widowed appear to use their religious beliefs as a method to cope
with their loss (Michael, Crowther, Schmid, & Allen, 2003).
Consistent with Freud’s (1927/1961, 1930/1989) suggestion that
religion can act as a social buffer against loss, increasing the
strength of one’s religious beliefs appears to be a useful tactic for
coping with loneliness (Rokach & Brock, 1998), divorce (Cain,
1988), or merely being single (Granqvist & Hagekull, 2000).
Coping appears to be especially facilitated by construing God in
particularly anthropomorphic terms as a partner or collaborator
rather than as a more personally distant social agent (Pargament,
1997; Pargament et al., 1988; Pargament, Smith, Koenig, & Perez,
1998). In a less extreme example, a Newsweek poll of religious
beliefs in America found that 40% of survey respondents indicated
that they felt closer to God when praying alone and only 2%
indicated feeling closer to God when praying with others (reported
in Adler, 2005). Religious agents—like pets or other anthropomor-
phized agents—clearly serve as effective sources of social con-
nection when human connection is lacking. Lacking social support
from other humans, our theory predicts that people may try to
create a sense of connection by anthropomorphizing nonhuman
agents.
Developmental influences: Attachment. The extent to which a
person is likely to feel socially isolated versus socially connected
depends on the quality of their social relationships. Systematic
differences in the quality of social relationships produced by
developmental influences are reflected in a person’s attachment
style (Ainsworth, Blehar, Waters, & Wall, 1978; Bowlby, 1969,
1973; Hazan & Shaver, 1987). Attachment styles are likely to be
related to anthropomorphism by altering the extent to which a
person is likely to be actively seeking information about social
relationships. In particular, those with insecure–anxious attach-
ment styles with close others are likely to be vigilantly monitoring
and appraising the environment for cues that suggest a supportive
and interested relationship partner, whereas those with secure
attachment styles are likely to be less vigilant (Fraley, Niedenthal,
Marks, Brumbaugh, & Vicary 2006; Mikulincer & Shaver, 2003).
Much like those who feel chronically or momentarily lonely, those
with insecure–anxious attachment styles may compensate by seek-
ing more secure or stable relationships from nonhuman agents and
may do so by anthropomorphizing nonhuman animals or religious
agents more intensely to create appropriate agents of social con-
nection. The active search for social connection is likely to render
anthropomorphic representations that would alleviate social dis-
connection highly accessible, thereby increasing the likelihood that
they are applied to a given target.
Although research has not investigated this hypothesis directly,
existing research suggests that those with insecure attachment
styles seek compensatory relationships from parasocial or nonhu-
man agents (especially religious agents). Those with insecure–
anxious attachment styles, for instance, are more likely than those
with secure attachment styles to form perceived social bonds with
television characters (Cole & Leets, 1999) and are more likely to
report an increase in religious belief over a given time period than
those with secure attachment styles (Kirkpatrick & Shaver, 1990).
Those with insecure–avoidant attachment styles—those who have
insecure attachments and actively avoid social contact— generally
report stronger religious beliefs, report a stronger relationship with
God, and are more likely to report sudden religious conversions
later in life (Kirkpatrick & Shaver, 1990). Those with insecure
attachment styles appear to actively seek social surrogates to form
the basis of more secure social relationships, and anthropomor-
phizing non-human agents is one potentially successful strategy
for obtaining such relationships.
Cultural influences: Individualism and collectivism. Culture
is the conduit through which people experience social connection
and relational connection, both at the local and at the national
level. Cultural differences in relational affiliations are most evident
in the distinction between individualist and collectivist value sys-
tems (e.g., Triandis, 1995). Although sociality appears to be a
relatively universal motivation (Baumeister & Leary, 1995), the
value placed on social relations and affiliation appears to be
greater in collectivist cultures (Markus & Kitayama, 1991; Sh-
weder & Bourne, 1984). The quality of one’s relations with others
is a central feature of the self in collectivist cultures, whereas
autonomy, privacy, and individual initiative are regarded highly in
addition to social needs within individualist cultures. Loneliness
may therefore be a less chronic condition among people living in
collectivist cultures, but the reaction to momentary isolation or
social exclusion may be more severe in these cultures. The impact
of culture on sociality motivation and anthropomorphism may be
somewhat nuanced.
The focus on social relations among collectivist cultures means
that the general level of felt social connection and support should
be higher among collectivist cultures than among individualistic
cultures and therefore should see more anthropomorphism result-
ing from general sociality motivation among individualistic cul-
tures than among collectivistic cultures. However, the greater
focus on social relations among collectivistic cultures means that
members of collectivistic cultures may respond more intensely to
momentary social isolation than members from individualistic
cultures (D. P. Johnson & Mullins, 1987) and therefore be more
likely to anthropomorphize in response to temporary feelings of
social isolation or loneliness. These hypotheses would be in line
with our SEEK theory of anthropomorphism, although experimen-
tal research investigating anthropomorphism across cultures has
yet to be conducted systematically.
Reiterations and Future Directions
Anthropomorphism in the perception of nonhuman agents is
commonly observed but is poorly understood. Existing scholarly
treatments have noted the extent to which people anthropomor-
877
ANTHROPOMORPHISM
phize nonhuman agents or have investigated the accuracy of an-
thropomorphic beliefs by studying the actual mental states of at
least some nonhuman agents. But none of this existing research
provides a psychological account of anthropomorphism itself to
explain when people are likely to see humanlike characteristics or
traits in nonhuman agents and when they are not. This article offers
such an account.
Drawing from research examining beliefs about nonhuman an-
imals, technological agents, religious deities, and other humans,
we have suggested that anthropomorphism is largely determined
by the operation of three major factors—the accessibility and
applicability of egocentric or homocentric knowledge (elicited
agent knowledge), the motivation to be effective social agents
(effectance motivation), and the motivation for social connection
(sociality motivation). These three factors in conjunction help to
explain why anthropomorphism is both a common and a wide-
spread phenomenon and, more important, predicts variability
across four major categories of independent variables (disposi-
tional, situational, developmental, and cultural).
Because experimental research on anthropomorphism is in its
infancy, much of the evidence amassed to support our overall
theory is extrapolated from research investigating inductive infer-
ences about other human agents. Research on anthropomorphism,
per se, is relatively scattered throughout the scientific literature and
nowhere near as developed as related research programs in social
cognition regarding other humans. This means that certain ele-
ments of our theoretical account are better supported by existing
scientific research than others, and many of the predictions we
have offered are as yet untested.
There are also at least three important issues within our theo-
retical account that we believe warrant direct research attention:
(a) the specificity of the elicited agent knowledge that guides
anthropomorphism, (b) empirical evidence of the inductive process
guiding anthropomorphism itself, and (c) greater understanding of
the interrelations between the three key psychological determi-
nants. We discuss each briefly in turn.
The Specificity of Elicited Agent Knowledge
The core cognitive component of our theoretical account is the
agent knowledge elicited at the time of judgment, and we have
suggested throughout this article that both categorical knowledge
about humans in general and egocentric knowledge about the self
in particular may serve as an inductive base for anthropomor-
phism. This naturally raises questions about which base is used
more routinely, and there is at present no empirical evidence that
would provide a clear answer. We suggest, however, three likely
determinants of whether general or specific knowledge is accessed.
First is the level at which a person categorizes another agent.
Broad categorizations of an agent made, for instance, at the level
of a species are likely to make broad categorical knowledge about
humans in general more accessible, whereas low level categoriza-
tions made, for instance, at the individual level of one’s unique pet
are likely to activate more specific egocentric knowledge about the
self. Determinants of the level at which an agent is categorized or
construed should therefore determine the level of agent knowledge
activated. Second, stimuli that are similar to other humans but not
to oneself should increase the likelihood of homocentric rather
than egocentric attributions. Finally, members of individualistic
cultures tend to access egocentric information and use their own
egocentric perspective more readily than members from collectiv-
istic cultures (Cohen & Gunz, 2002; Wu & Keysar, 2007). Mem-
bers of individualistic cultures should therefore be more likely to
use egocentric representations, whereas members of collectivistic
cultures should be more likely to use homocentric representations.
Differentiating the activation and use of these two forms of knowl-
edge should enable more specific predictions about the exact
nature of anthropomorphic judgments.
Underlying Process Evidence
We have suggested that anthropomorphism uses the same basic
mechanisms of inductive inference that guide many attributional
phenomena in social cognition. In particular, we have suggested
that anthropomorphic knowledge is often rapidly activated for a
variety or reasons, and final judgments are biased in the direction
of this information compared with information that is not as
readily accessible because subsequent correction processes are
generally insufficient. Under such conditions, our account—like
those from other domains—predicts that those without sufficient
time or cognitive resources to engage in effortful correction will
show stronger evidence of anthropomorphism than those given
more time or ability to engage in more effortful thought. Such
evidence would clearly demonstrate that people are reasoning
about nonhuman agents from an inductive base of homocentric or
egocentric knowledge and would provide compelling evidence that
the process underlying the attribution of humanlike dispositions
and traits to nonhuman agents (anthropomorphism) is indeed
highly similar to the process underlying the attribution of human
dispositions and traits to other humans (social cognition).
Interrelation Between Key Psychological Determinants
We have thus far discussed each of the psychological determi-
nants of anthropomorphism largely in isolation but have suggested
that they work together and function synergistically in daily life.
The inductive process of anthropomorphism works through the
agent knowledge that is activated, applied, or adjusted when rea-
soning about nonhuman agents, but each part of this process can be
guided or modified by motivations to understand and effectively
interact with another agent or to perceive it as a source of social
connection (Higgins, 1996). In this sense, acquiring and develop-
ing anthropocentric knowledge that can be activated and applied
may be seen as a necessary condition for anthropomorphism to
occur.
This relationship between the cognitive and motivational factors
has two important implications for anthropomorphism. First, ef-
fectance and sociality motivation increase the likelihood of acti-
vating and directly applying anthropocentric knowledge to a non-
human target but may operate differently if nonanthropomorphic
knowledge representations are readily accessible. Effectance mo-
tivation is likely to increase the use of these alternate structures to
attain a sense of understanding, thereby reducing the likelihood of
anthropomorphism. For example, the Yukatek Mayan children
described earlier (Medin & Atran, 2004) showed no preference for
“human” as a base for induction toward other animals because they
simply possessed better conceptual alternatives for reasoning
about nonhuman animals. The urban industrialized children who
878
EPLEY, WAYTZ, AND CACIOPPO
more readily exhibit anthropocentrism (Carey, 1985), on the other
hand, relied on “human” as a guide to other animals because the
concept human best served their comprehension needs. Sociality
motivation, however, increases the general tendency to seek, no-
tice, and attend to social behaviors in other agents, and sociality
motivation is therefore likely to increase the tendency to anthro-
pomorphize even if nonanthropocentric knowledge structures are
readily accessible. A skilled computer technician is unlikely to
anthropomorphize the well-known box of circuits compared with a
novice computer user who lacks such knowledge, but the techni-
cian may nonetheless treat the computer as a pal if human social
connection is severely lacking. Effectance should therefore in-
crease anthropomorphism when nonanthropocentric knowledge is
unavailable, but sociality motivation should increase anthropomor-
phism even in contexts when nonanthropocentric knowledge is
readily accessible.
Second, these cognitive and motivational processes are likely to
follow distinctly different time courses of activation and applica-
tion and therefore have interesting implications for the expression
of anthropomorphism. In particular, knowledge accessibility typ-
ically decays quickly over time (except in instance of chronic
accessibility) and can be corrected or overcome by competing
knowledge structures. As mentioned earlier, this suggests that
anthropomorphic inferences may serve as an automatic base for
induction that may be subsequently overcome by more deliberate
and conscious reasoning. Motivational processes, in contrast, typ-
ically operate like drive states in the service of either general or
specific goals and follow the opposite temporal pattern by increas-
ing in strength over time until the activated goal is satisfied, at
which point the motivation diminishes. Once activated, effectance
and sociality motivation should therefore increase the tendency to
anthropomorphize nonhuman agents until uncertainty or social
deprivation is overcome. Factors that impact the accessibility of
egocentric or homocentric knowledge should therefore influence
anthropomorphism relatively quickly and be relatively localized in
time and across stimuli, whereas factors that impact the motiva-
tional determinants of effectance and sociality may have longer
lasting and wider ranging effects until these motivational goals are
fully achieved (Bargh et al., 2001). Although we have specified
conditions that account for variability in anthropomorphism and
described some of the underlying psychological processes, further
elucidation of the specific psychological processes that guide
anthropomorphism through these cognitive and motivational fac-
tors is an especially promising area for future research.
Applications: Robotics and Human–Computer Interaction
Beyond testing unique predictions about anthropomorphism and
the psychological processes that guide it, we believe this theory
also helps to inform applied fields involving human interaction
with technological agents. From computers to interactive naviga-
tion systems to home alarms, life in the modern developed world
involves frequent interaction with an ever-expanding array of
lifelike technological and virtual agents (more frequent interaction
for many, in fact, than with biological agents; Nowak, 2001). Our
theory provides insights about how to conceptually design such
agents to both facilitate understanding and learning and increase
liking through the creation of a virtual social connection.
The inner workings of most modern technological agents are
every bit as obtuse as the mental states of biological agents, but the
incentives for understanding and effectively interacting with such
agents are very high. This effectance motivation coupled with a
general lack of understanding means that the tendency to anthro-
pomorphize the workings of many nonhuman agents may be
especially high.
Anthropomorphism can enable a sense of efficacy with these
agents, a sense that actually increases one’s apparent competence
interacting with these agents. Indeed, artificial intelligence design
appears to benefit from attempts to elicit anthropomorphism from
users. People rate robots who appear to exhibit playful behavior as
being more extroverted and outgoing than robots that appeared
more serious, and people were more likely to cooperate and work
with the playful robots than with the serious robots (Kiesler &
Goetz, 2002). The nature of anthropomorphic traits projected onto
these robots significantly influenced subsequent interactions. More
closely related to effectance motivation, anthropomorphic virtual
assistants within software programs (such as in Microsoft Word)
improve users ability to learn the software by increasing their
ability to cope with information overload (Moreale & Watt, 2004).
Facilitating anthropomorphism may therefore serve as an effective
method for improving the usefulness of certain technological
agents.
Facilitating anthropomorphism may also increase the usefulness
of technological agents by creating social bonds that increase a
sense of social connection. Participants playing a desert survival
task in one experiment reported feeling better understood in the
task when more anthropomorphic faces and voices appeared in the
interface (Burgoon et al., 2000). Such social bonds are likely to be
facilitated by increasing the extent to which a technological agent
is morphologically similar to selective human features (e.g., eyes),
a technique long used by toy makers to create strong bonds
between children and anthropomorphized toys such as the Furby
(Turkle, 2000). Among adults, human-gender stereotypes can be
activated by the presence of a male or female voice played through
a computer: Female voices lead people to rate computers as
friendlier than ones with male voices, but praise coming from a
male voice makes the computer appear more compelling and
persuasive than when the same praise comes from a female voice
(Nass, Moon, & Green, 1997).
In addition to increasing a sense of efficacy, anthropomorphiz-
ing nonhuman agents enables social connection to develop—
however anemic it may be compared with social connection with
actual humans—that increases liking for technological agents that
would otherwise evoke no sense of social connection (Nass, Is-
bister, & Lee, 2000; Nass, Moon, Fogg, Reeves, & Dryer, 1995).
Such anemic relationships may have some appeal if human social
connection is unavailable because of the increased control an
individual can have over an anthropomorphized agent compared
with another human, or simply because people settle for satisfac-
tory (rather than optimal) outcomes.
Note that these social connections enabled by humanlike tech-
nologies can have some unforeseen consequences for computer-
ized tasks designed to be devoid of social influence. Anonymous
computerized surveys, for instance, are designed to elicit more
honest responses from respondents to socially sensitive topics than
would normally be offered in the presence of another human. But
computerized voices increase socially desirable responding on
879
ANTHROPOMORPHISM
these questionnaires in much the same way that another actual
human questioner would, including responding in more socially
desirable ways to gender-attitude and personal drug use questions
(Couper, Tourangeau, & Steiger, 2001; Tourangeau, Couper, &
Steiger, 2003). Findings within this domain have revealed a great
deal of variability, however, and discrepancies remain. Our model
of anthropomorphism can again provide assistance by predicting
the conditions under which the influence of human cues and
particular social identity cues will be most powerful.
Inverting Anthropomorphism: Insights Into
Dehumanization
Finally, anthropomorphism is an act of humanization— of see-
ing humanlike characteristics and traits in nonhuman agents—and
our theory identifies factors that can increase as well as decrease
such humanization. Our theory might also yield some insights into
the inverse process of dehumanization— ceasing to attribute hu-
manlike characteristics to humans, treating them as if they were
nonhuman agents or objects (Haslam, 2006). Anthropomorphism
is of practical interest in most social spheres because it turns
nonhuman agents into moral agents who deserve to be treated with
respect and concern. Pollution takes on a different tone altogether
when it is “harming Mother Earth,” for instance, and it is no
surprise that such framing is common among environmentalist
groups who show the strongest concern for the environment.
Eating beef would become morally repugnant for many if bovine
neurologists were to determine—through rather large fMRI equip-
ment—that cattle were every bit as sentient and self-aware as (at
least some of) our family members and friends. Dehumanization is
troublesome for most because of the very inverse process of
stripping other humans of their moral agency and enabling them to
be treated as objects (e.g., Bandura, Underwood, & Fromson,
1975; Kelman, 1995; Pines & Solomon, 1977). The psychological
determinants that influence the perception of humanlike traits in
nonhumans should be the same that influence the perception of
these same traits— or the lack of these traits—in other humans as
well.
Indeed, existing research provides considerable support consis-
tent with the present theory. Consistent with predictions regarding
similarity derived from elicited agent knowledge, people are no-
toriously likely to dehumanize outgroups, who are naturally seen
and classified as being different from the ingroup (Demoulin et al.,
2002; Gaunt, Leyens, & Demoulin, 2002; L. T. Harris & Fiske,
2006; Leyens et al., 2000) or whose members behave bizarrely or
inexplicably, such as people with schizophrenia (Kramer & Buck,
1997). Consistent with predictions from effectance motivation,
those whom people seek to exterminate and therefore have no
motivation or desire to understand— or quite literally have no
chance of future interaction with—are the most likely to be dehu-
manized (Osofsky, Bandura, & Zimbardo, 2005). And consistent
with predictions from sociality motivation, strong social connec-
tions are necessarily had with ingroup members, and these strong
connections may diminish the baseline sociality motivation that
leads people to otherwise see humanlike traits in outgroup mem-
bers. Those who live in particularly highly connected communities
may be the least in need of social connection from outside group
members and possibly the most likely to fail to attribute humanlike
characteristics to more distant outgroup members (Marcu &
Chryssochoou, 2005; Pe´rez, Chulvi, & Rosario, 2001). Under-
standing the psychological determinants that make people likely to
see other agents as humanlike should also enable understanding
into when people are unlikely to do so.
Concluding Thoughts: The Flexible Boundaries of Social
Cognition
Give me one minute—just one minute—inside the skin of this crea-
ture. Hook me up for just sixty seconds to the perceptual and con-
ceptual apparatus of this other being—and then I will know what
natural historians have sought through the ages. . . . [But] I am stuck
with a panalopy of ineluctably indirect methods.” (Gould, 1998, p.
376)
Natural historians are not alone in their desire to understand
what it is like to be another creature, and they suffer the very same
limitations as the rest of us when trying to achieve such under-
standing. Human sensory experience is distinctly embodied, mean-
ing that another agent’s actual experience is forever out of reach
and one can know the nature of another creature—including other
human creatures— only through a process of indirect inference.
When a person relies on egocentric or anthropocentric knowledge
to guide reasoning about nonhuman agents— be they religious,
mechanical, or animal agents— he or she is anthropomorphizing.
The theory we have outlined here provides a framework for
understanding this process of anthropomorphism in a way that
allows for predictions about when people will anthropomorphize
and when they will not, when nonhuman agents will be treated as
humanlike and when they will be treated as objects, and perhaps
also when humanlike agents will be treated as nonhuman. This
theory identifies three critical determinants of anthropomorphism
and makes predictions about dispositional, situational, develop-
mental, and cultural variability in the process of anthropomor-
phism. This theory does not provide a solution to Gould’s (1998)
frustration as an outside observer trying to peer into the mind of a
nonhuman agent. It does, however, provide insight into what an
outside observer trying to peer into the mind of a nonhuman agent
is likely to see.
References
Adler, J. (2005, August 29). In search of the spiritual. Newsweek, 146,
46 64.
Aggarwal, P., & McGill, A. L. (in press). Is that car smiling at me? Schema
congruity as a basis for the evaluation of anthropomorphized products.
Journal of Consumer Research.
Ainsworth, M. D., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns
of attachment: A psychological study of the strange situation. Hillsdale,
NJ: Erlbaum.
Allen, K., Blascovich, J., & Mendes, W. B. (2002). Cardiovascular reac-
tivity and the presence of pets, friends and spouses: The truth about cats
and dogs. Psychosomatic Medicine, 64, 727–739.
Ames, D. R. (2004). Inside the mind reader’s toolkit: Projection and
stereotyping in mental state inference. Journal of Personality and Social
Psychology, 87, 340 –353.
Amsterdam, B. K. (1972). Mirror self-image reactions before age two.
Developmental Psychobiology, 5, 297–305.
Asquith, P. J. (1986). Anthropomorphism and the Japanese and Western
traditions in primatology. In J. G. Else & P. C. Lee (Eds.), Primate
880
EPLEY, WAYTZ, AND CACIOPPO
ontogeny, cognition, and social behavior (pp. 61–71). Cambridge, En-
gland: Cambridge University Press.
Asquith, P. J. (1996). Japanese science and Western hegemonies: Prima-
tology and the limits set to questions. In L. Nader (Ed.), Naked science:
Anthropological inquiry into boundaries, power, and knowledge (pp.
239 –258). New York: Routledge.
Atran, S., Medin, D., Lynch, E., Vapnarsky, V., Ucan Ek’, E., & Sousa, P.
(2001). Folkbiology doesn’t come from folkpsychology: Evidence from
Yukatek Maya in cross-cultural perspective. Journal of Cognition and
Culture, 1, 3– 42.
Averill, J. R. (1973). Personal control over aversive stimuli and its rela-
tionship to stress. Psychological Bulletin, 80, 286 –303.
Axelrod, R., & Hamilton, W. (1981, March 27). The evolution of cooper-
ation. Science, 211, 1390 –1396.
Azhar, M. Z., & Varma, S. L. (1995). Religious psychotherapy in depres-
sive patients. Psychotherapy and Psychosomatics, 63, 165–168.
Bandura, A., Underwood, B., & Fromson, M. E. (1975). Disinhibition of
aggression through diffusion of responsibility and dehumanization of
victims. Journal of Research in Personality, 9, 253–269.
Bargh, J. A., Gollwitzer, P. M., Lee-Chai, A., Barndollar, K., & Tro¨tschel,
R. (2001). The automated will: Nonconscious activation and pursuit of
behavioral goals. Journal of Personality and Social Psychology, 81,
1014 –1027.
Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child
have a “theory of mind”? Cognition, 21, 37– 46.
Baron-Cohen, S., Tager-Flusberg, H., & Cohen, D. (1993). Understanding
other minds: Perspectives from autism. Oxford, England: Oxford Uni-
versity Press.
Barrett, D., Kurian, G., & Johnson, T. (Eds.). (2001). World Christian
encyclopedia (2nd Ed.). New York: Oxford University Press.
Barrett, J. L., & Johnson, A. H. (2003). The role of control in attributing
intentional agency to inanimate objects. Journal of Cognition and Cul-
ture, 3, 208 –217.
Barrett, J. L., & Keil, F. C. (1996). Conceptualizing a non-natural entity:
Anthropomorphism in God concepts. Cognitive Psychology, 31, 219
247.
Barrett, J. L., Newman, R. M., & Richert, R. A. (2003). When seeing is not
believing: Children’s understanding of humans’ and non-humans’ use of
background knowledge in interpreting visual displays. Journal of Cog-
nition and Culture, 3, 91–108.
Barsalou, L. W. (1983). Ad hoc categories. Memory & Cognition, 11,
211–227.
Bassili, J. N. (1976). Temporal and spatial contingencies in the perception
of social events. Journal of Personality and Social Psychology, 33,
680 685.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for
interpersonal attachments as a fundamental human motivation. Psycho-
logical Bulletin, 117, 497–529.
Berger, C. R., & Douglas, W. (1981). Studies in interpersonal epistemol-
ogy III: Anticipated interaction, self-monitoring, and observational con-
text selection. Communication Monographs, 48, 183–196.
Bering, J. M., & Bjorklund, D. F. (2004). The natural emergence of
reasoning about the afterlife as a developmental regularity. Developmen-
tal Psychology, 40, 217–233.
Berscheid, E., Graziano, W., Monson, T., & Dermer, M. (1976). Outcome
dependency: Attention, attribution, and attraction. Journal of Personal-
ity and Social Psychology, 34, 978–989.
Bertenthal, B. I. (1993). Perception of biomechanical motions by infants:
Intrinsic image and knowledge-based constraints. In C. Granrud (Ed.),
Carnegie Symposium on Cognition: Visual perception and cognition in
infancy (pp. 175–214). Hillsdale, NJ: Erlbaum.
Birgegard, A., & Granqvist, P. (2004). The correspondence between at-
tachment to parents and God: Three experiments using subliminal sep-
aration cues. Personality and Social Psychology Bulletin, 30, 1122–
1135.
Boomsma, D. I., Willemsen, G., Dolan, C. V., Hawkley, L. C., & Ca-
cioppo, J. T. (2005). Genetic and environmental contributions to lone-
liness in adults: The Netherlands Twin Register Study. Behavior Genet-
ics, 35, 745–752.
Bowlby, J. (1969). Attachment and loss (Vol. 1). New York: Basic Books.
Bowlby, J. (1973). Attachment and loss (Vol. 2). New York: Basic Books.
Brown, S. L., Nesse, R. M., House, J. S., & Utz, R. L. (2004). Religion and
emotional compensation: Results from a prospective study of widow-
hood. Personality and Social Psychology Bulletin, 30, 1165–1174.
Buccino, G., Binkofski, F., Fink, G. R., Fadiga, L., Fogassi, L., Gallese, V.,
et al. (2001). Action observation activates premotor and parietal areas in
a somatotopic manner: An fMRI study. European Journal of Neuro-
science, 13, 400 404.
Burger, J. M. (1992). Desire for control: Personality, social and clinical
perspectives. New York: Plenum.
Burger, J. M., & Cooper, H. M. (1979). The desirability of control.
Motivation and Emotion, 3, 381–393.
Burger, J. M., & Hemans, L. T. (1988). Desire for control and the use of
attribution processes. Journal of Personality, 56, 531–546.
Burgoon, J. K., Bonito, J. A., Bengtsson, B., Cederberg, C., Lundeberg, M.,
& Allspach, L. (2000). Interactivity in human-computer interaction: A
study of credibility, understanding, and influence. Computers in Human
Behavior, 16, 553–574.
Buss, D. M. (1991). Evolutionary personality psychology. Annual Review
of Psychology, 42, 459 491.
Cacioppo, J. T., Hawkley, L. C., Ernst, J. M., Burleson, M., Berntson,
G. G., Nouriani, B., & Spiegel, D. (2006). Loneliness within a nomo-
logical net: An evolutionary perspective. Journal of Research in Per-
sonality, 40, 1054 –1085.
Cacioppo, J. T., Hughes, M. E., Waite, L. J., Hawkley, L. C., & Thisted,
R. A. (2006). Loneliness as a specific risk factor for depressive symp-
toms: Cross sectional and longitudinal analyses. Psychology and Aging,
21, 140–151.
Cacioppo, J. T., Petty, R. E., Feinstein, J. A., & Jarvis, W. B. G. (1996).
Dispositional differences in cognitive motivation: The life and times of
individuals varying in need for cognition. Psychological Bulletin, 119,
197–253.
Cain, B. S. (1988). Divorce among elderly women: A growing social
phenomenon. Social Casework––Journal of Contemporary Social Work,
69, 563–568.
Carey, S. (1985). Conceptual change in childhood. Cambridge, MA: Brad-
ford Books, MIT Press.
Carr, L., Iacoboni, M., Dubeau, M. C., Mazziotta, J. C., & Lenzi, G. L.
(2003). Neural mechanisms of empathy in humans: A relay from neural
systems for imitation to limbic areas. Proceedings of the National
Academy of Sciences, USA, 100, 5497–5502.
Carruthers, P. (1996). Simulation and self-knowledge: A defense of theory-
theory. In P. Carruthers & P. K. Smith (Eds.), Theories of theories of
mind (pp. 22–38). Cambridge, England: Cambridge University Press.
Castelli, F., Frith, C., Happe´, F., & Frith, U. (2002). Autism and brain
mechanisms for the attribution of mental states to animated shapes.
Brain, 125, 1839 –1849.
Castelli, F., Happe´, F., Frith, U., & Frith, C. D. (2000). Movement and
mind: A functional imaging study of perception and interpretation of
complex intentional movement patterns. NeuroImage, 12, 314 –325.
Chandler, M. J. (1977). Social cognition: A selected review of current
research. In W. F. Overton & J. M. Gallagher (Eds.), Knowledge and
development: Vol. 1. Advances in research and theory (pp. 93–147).
New York: Plenum.
Cheney, D., & Seyfarth, R. (1990). How monkeys see the world. Chicago:
University of Chicago Press.
Cohen, D., & Gunz, A. (2002). As seen by the other . . .: Perspectives on
881
ANTHROPOMORPHISM
the self in the memories and emotional perceptions of Easterners and
Westerners. Psychological Science, 13, 55–59.
Cole, T., & Leets, L. (1999). Attachment styles and intimate television
viewing. Insecurely forming relationships in a parasocial way. Journal
of Social and Personal Relationships, 16, 495–511.
Couper, M. P., Tourangeau, R., & Steiger, D. M. (2001). Social presence
in Web surveys. In M. Beaudouin-Lafon, J. Beaudouin-Lafon, & J. K.
Robert (Eds.), Proceedings of the ACM CHI 2001 Human Factors in
Computing Systems Conference (pp. 412–417). New York: ACM Press.
D’Agostino, P. R., & Fincher-Kiefer, R. (1992). Need for cognition and
correspondence bias. Social Cognition, 10, 151–163.
Darwin, C. (2002). The expression of emotions in man and animals. New
York: Oxford University Press. (Original work published 1872)
Dasser, V., Ulbaek, I., & Premack, D. (1989, January 20). The perception
of intention. Science, 243, 365–367.
Dawes, R., & Mulford, M. (1996). The false consensus effect and over-
confidence: Flaws in judgment or flaws in how we study judgment?
Organizational Behavior and Human Decision Processes, 65, 201–211.
Demoulin, S., Cortes, B. P., Viki, T. G., Rodriguez, A. P., Rodriguez, R. T.,
Paladino, M. P., & Leyens, J. P. (2002). The role of in-group identifi-
cation in infra-humanization. Unpublished manuscript.
Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press.
Deutsch, W., & Pechmann, T. (1982). Social interaction and the develop-
ment of definite descriptions. Cognition, 11, 159 –184.
de Waal, F. B. M. (2003). Silent invasion: Imanishi’s primatology and
cultural bias in science. Animal Cognition, 6, 293–299.
DiSalvo, C., Gemperle, F., Forlizzi, J., & Kiesler, S. (2002, June). All
robots are not created equal: Design and the perception of humanness
in robot heads. Paper presented at the DIS2002 Conference Proceedings,
London, England.
Dittes, J. E. (1961). Impulsive closure as a reaction to failure-induced
threat. Journal of Abnormal and Social Psychology, 63, 562–569.
Dittrich, W. H., & Lea, S. E. G. (1994). Visual perception of intentional
motion. Perception, 23, 253–268.
Douglas, W. (1990). Uncertainty, information-seeking, and liking during
initial interaction. Western Journal of Speech Communication, 54, 66
81.
Epley, N. (2004). A tale of tuned decks? Anchoring as accessibility and
anchoring as adjustment. In D. J. Koehler & N. Harvey (Eds.), The
Blackwell handbook of judgment and decision making (pp. 240 –256).
Oxford, England: Blackwell Publishers.
Epley, N., Akalis, S., Waytz, A., & Cacioppo, J. T. (in press). Creating
social connection through inferential reproduction: Loneliness and per-
ceived agency in gadgets, gods, and greyhounds. Psychological Science.
Epley, N., & Gilovich, T. (2005). When effortful thinking influences
judgmental anchoring: Differential effects of forewarning and incentives
on self-generated and externally-provided anchors. Journal of Behav-
ioral Decision Making, 18, 199 –212.
Epley, N., & Gilovich, T. (2006). The anchoring and adjustment heuristic:
Why adjustments are insufficient. Psychological Science, 17, 311–318.
Epley, N., Keysar, B., Van Boven, L., & Gilovich, T. (2004). Perspective
taking as egocentric anchoring and adjustment. Journal of Personality
and Social Psychology, 87, 327–339.
Epley, N., Morewedge, C. K., & Keysar, B. (2004). Perspective taking in
children and adults: Equivalent egocentrism but differential correction.
Journal of Experimental Social Psychology, 40, 760 –768.
Epley, N., Waytz, A., Akalis, S., & Cacioppo, J. T. (in press). When we
need a human: Motivational determinants of anthropomorphism. Social
Cognition.
Farah, M. J., Wilson, K. D., Drain, M., & Tanaka, J. N. (1998). What is
“special” about face perception? Psychological Review, 105, 482– 498.
Fazio, R. H., & Olson, M. A. (2003). Implicit measures in social cognition
research: Their meaning and uses. Annual Review of Psychology, 54,
297–327.
Feldman, N. S., & Ruble, D., (1981). The development of person percep-
tion: Cognitive and social factors. In S. S. Brehm, S. M. Kassin, & F. X.
Gibbons (Eds.), Developmental social psychology: Theory and research
(pp. 191–210). New York: Oxford University Press.
Feuerbach, L. (2004). The essence of religion. Amherst, NY: Prometheus
Books. (Original work published 1873)
Flavell, J. H. (1973). Metacognitive aspects of problem-solving. In L. B.
Resnick (Ed.), The nature of intelligence (pp. 231–236). Hillsdale, NJ:
Erlbaum.
Flavell, J. H. (1986). The development of children’s knowledge about the
appearance-reality distinction. American Psychologist, 41, 418 425.
Fraley, R. C., Niedenthal, P. M., Marks, M., Brumbaugh, C., & Vicary, A.
(2006). Adult attachment and the perception of emotional expressions:
Probing the hyperactivating strategies underlying anxious attachment.
Journal of Personality, 74, 1163–1190.
Freud, S. (1961). The future of an illusion. New York: Norton. (Original
work published 1927)
Freud, S. (1989). Civilization and its discontents. New York: Norton.
(Original work published 1930)
Gardner, W. L., Pickett, C. L., Jefferis, V., & Knowles, M. (2005). On the
outside looking in: Loneliness and social monitoring. Personality and
Social Psychology Bulletin, 31, 1549 –1560.
Gardner, W. L., Pickett, C., & Knowles, M. L. (2005). “Social snacking”
and “social shielding”: The use of symbolic social bonds to maintain
belonging needs. In K. Williams, J. Forgas, & W. von Hippel (Eds.), The
social outcast: Ostracism, social exclusion, rejection, and bullying (pp.
227–242). New York: Psychology Press.
Gates, A. I. (1949). Edward L. Thorndike, 1874–1949. Psychological
Review, 56, 241–243.
Gaunt, R., Leyens, J. P., & Demoulin, S. (2002). Intergroup relations and
the attribution of emotions: Control over memory for secondary emo-
tions associated with the in-group and outgroup. Journal of Experimen-
tal Social Psychology, 38, 508 –514.
Gergely, G., Na´dasdy, Z., Csibra, G., & ´ro´, S. (1995). Taking the
intentional stance at 12 months of age. Cognition, 56, 165–193.
Gilbert, D. T. (1991). How mental systems believe. American Psycholo-
gist, 46, 107–119.
Gilbert, D. T. (1998). Ordinary personology. In D. T. Gilbert, S. T. Fiske,
& G. Lindzey (Eds.), The handbook of social psychology (4th ed., pp.
89 –150). New York: McGraw Hill.
Gilbert, D. T., Gill, M. J., & Wilson, T. D. (2002). The future is now:
Temporal correction in affective forecasting. Organizational Behavior
and Human Decision Processes, 88, 403–444.
Gilbert, D. T., & Malone, P. S. (1995). The correspondence bias. Psycho-
logical Bulletin, 117, 21–38.
Gilovich, T., Vallone, R., & Tversky, A. (1985). The hot hand in basket-
ball: On the misperception of random sequences. Cognitive Psychology,
17, 295–314.
Glick, I. D., Weiss, R. S., & Parkes, C. M. (1974). The first year of
bereavement. New York: John Wiley.
Gopnik, A., & Meltzoff, A. N. (1994). Minds, bodies, and persons: Young
children’s understanding of the self and others as reflected in imitation
and ‘theory of mind’ research. Behavioral and Brain Sciences, 16, 1–14.
Gopnik, A., Meltzoff, A. N., & Kuhl, P. K. (2001). The scientist in the crib.
What early learning tells us about the mind. New York: Perennial.
(Original work published 1999)
Gould, S. J. (1998). Can we truly know sloth and rapacity? In Leonardo’s
Mountain of Clams and the Diet of Worms (pp. 375–391). New York:
Three Rivers Press.
Granqvist, P., & Hagekull, B. (2000). Religiosity, adult attachment, and
why “singles” are more religious. The International Journal for the
Psychology of Religion, 10, 111–123.
Gray, H. M., Gray, K., & Wegner, D. M. (2007, February 2). Dimensions
of mind perception. Science, 315, 619.
882
EPLEY, WAYTZ, AND CACIOPPO
Griffin, D., & Tversky, A. (1992). The weighing of evidence and the
determinants of confidence. Cognitive Psychology, 24, 411– 435.
Guthrie, S. E. (1993). Faces in the clouds: A new theory of religion. New
York: Oxford University Press.
Harlow, H. F. (1953). Mice, monkeys, men, and motives. Psychological
Review, 60, 23–32.
Harlow, H. F. (1958). The nature of love. American Psychologist, 13,
673– 685.
Harris, L. T., & Fiske, S. T. (2006). Dehumanizing the lowest of the low.
Psychological Science, 17, 847– 853.
Harris, P. (1992). From simulation to folk psychology: The case for
development. Mind and Language, 7, 120–144.
Harter, S. (1978). Effectance motivation reconsidered: Toward a develop-
mental model. Human Development, 21, 34 64.
Haslam, N. (2006). Dehumanization: An integrated review. Personality
and Social Psychology Review, 10, 252–264.
Hauser, M. D. (2000). Wild minds: What animals really think. New York:
Holt.
Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an
attachment process. Journal of Personality and Social Psychology, 52,
511–524.
Heal, J. (1986). Replication and functionalism. In J. Butterfield (Ed.),
Language, mind and logic (pp. 135–150). Cambridge, England: Cam-
bridge University Press.
Hebb, D. O. (1946). Emotion in man and animal: An analysis of the
intuitive processes of recognition. Psychological Review, 53, 88 –106.
Heberlein, A. S., & Adolphs, R. (2004). Impaired spontaneous anthropo-
morphizing despite intact perception and social knowledge. Proceedings
of the National Academy of Sciences, USA, 101, 7487–7491.
Heckhausen, H. (1991). Motivation and action. New York: Springer.
Heider, F., & Simmel, M. (1944). An experimental study of apparent
behavior. American Journal of Psychology, 57, 243–249.
Heltzer, R., & Vyse, S. A. (1994). Intermittent consequences and problem
solving: The experimental control of “superstitious” beliefs. Psycholog-
ical Record, 44, 155–169.
Heywood, V. H. (Ed.). (1995). Global Biodiversity Assessment. United
Nations Environment Programme. Cambridge, MA: Cambridge Univer-
sity Press.
Higgins, E. T. (1989). Knowledge accessibility and activation: Subjectivity
and suffering from unconscious sources. In J. S. Uleman & J. A. Bargh
(Eds.), Unintended thought (pp. 75–123). New York: Guilford Press.
Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability,
and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Social
psychology: Handbook of basic principles (pp. 133–168). New York:
Guilford Press.
Hofstede, G. (1980). Culture’s consequences: International differences in
work-related values. Beverly Hills, CA: Sage.
Hofstede, G. (2001). Culture’s consequences; Comparing values, behav-
iors, institutions, and organizations across nations (2nd ed.). Thousand
Oaks, CA: Sage.
Holland, J. H., Holyoak, K. J., Nisbett, R. E., & Thagard, P. R. (1986).
Induction: Processes of inference, learning, and discovery. Cambridge,
MA: MIT Press/Bradford Books.
Hume, D. (1956). The natural history of religion. Stanford, CA: Stanford
University Press. (Original work published 1757)
Iacoboni, M., Lieberman, M. D., Knowlton, B. J., Molnar-Szakacs, I.,
Moritz, M., Throop, C. J., & Fiske, A. P. (2004). Watching social
interactions produces dorsomedial prefrontal and medial parietal BOLD
fMRI signal increases compared to a resting baseline. NeuroImage, 21,
1167–1173.
Inagaki, K., & Hatano, G. (1987). Young children’s spontaneous person-
ification as analogy. Child Development, 58, 1013–1020.
Ito, T. A., & Cacioppo, J. T. (2000). Electrophysiological evidence of
implicit and explicit categorization processes. Journal of Experimental
Social Psychology, 36, 660 676.
James, W. (1950). The principles of psychology. New York: Dover. (Orig-
inal work published 1890)
Johnson, D. P., & Mullins, L. C. (1987). Growing old and lonely in
different societies: Toward a comparative perspective. Journal of Cross-
Cultural Gerontology, 2, 257–275.
Johnson, S. C. (2003). Detecting agents. Philosophical Transactions of the
Royal Society B, 358, 549 –559.
Johnson, S. C., Booth, A. E., & O’Hearn, K. (2001). Inferring the goals of
a non-human agent. Cognitive Development, 16, 637– 656.
Johnson, S. C., Slaughter, V., & Carey, S. (1998). Whose gaze will infants
follow? The elicitation of gaze following in 12-month-olds. Develop-
mental Science, 1, 233–238.
Kellerman, K., & Reynolds, R. (1990). When ignorance is bliss: The role
of motivation to reduce uncertainty in uncertainty reduction theory.
Human Communication Research, 17, 5–75.
Kelman, H. C. (1995). The social context of torture: Policy process and
authority structure. In R. D. Crelinsten & A. P. Schmid (Eds.), The
politics of pain: Torturers and their masters (pp. 19–34). Boulder, CO:
Westview Press.
Keysar, B., & Barr, D. J. (2002). Self-anchoring in conversation: Why
language users don’t do what they “should.” In T. Gilovich, D. Griffin,
& D. Kahneman (Eds.), Heuristics and biases: The psychology of
intuitive judgment (pp. 150 –166). Cambridge, England: Cambridge Uni-
versity Press.
Kiesler, S., & Goetz, J. (2002). Mental models and cooperation with
robotic assistants. In Proceedings of CHI’02 on Human Factors in
Computing Systems (pp. 576 –577). Minneapolis, MN: ACM Press.
Kirkpatrick, L. A., & Shaver, P. R. (1990). Attachment theory and religion:
Childhood attachments, religious beliefs, and conversion. Journal for
the Scientific Study of Religion, 29, 315–334.
Kling, A., Lancaster, J., & Benitone, J. (1970). Amygdalectomy in the
free-ranging vervet (Cercopithecus aethiops). Journal of Psychiatric
Research, 7, 191–199.
Knight, N., Sousa, P., Barrett, J. L., & Atran, S. (2004). Children’s
attributions of beliefs to humans and God: Cross cultural evidence.
Cognitive Science, 28, 117–126.
Kramer, A., & Buck, L. A. (1997). Encountering people labeled “schizo-
phrenic.” Journal of Humanistic Psychology, 37, 12–29.
Kruger, J. (1999). Lake Wobegon be gone! The “below-average effect” and
the egocentric nature of comparative ability judgments. Journal of
Personality and Social Psychology, 77, 221–232.
Kruglanski, A. (1990). Motivations for judging and knowing: Implications
for causal attribution. In E. T. Higgins & R. M. Sorrentino (Eds.),
Handbook of motivation and cognition: Foundations of social behavior
(Vol. 2, pp. 333–368). New York: Guilford Press.
Kruglanski, A. W., & Freund, T. (1983). The freezing and unfreezing of lay
inferences: Effects on impressional primacy, ethnic stereotyping, and
numerical anchoring. Journal of Experimental Social Psychology, 19,
448 468.
Kruglanski, A. W., & Mayseless, O. (1988). Contextual effects in hypoth-
esis testing: The role of competing alternatives and epistemic motiva-
tions. Social Cognition, 6, 1–21.
Kruglanski, A. W., & Webster, D. M. (1991). Group members’ reactions to
opinion deviates and conformists at varying degrees of proximity to
decision deadline and of environmental noise. Journal of Personality
and Social Psychology, 61, 212–225.
Kruglanski, A. W., & Webster, D. M. (1996). Motivated closing of the
mind: “Seizing” and “freezing.” Psychological Review, 103, 263–283.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: Uni-
versity of Chicago Press.
Lazarus, R. S. (1966). Psychological stress and the coping process. New
York: McGraw-Hill.
883
ANTHROPOMORPHISM
Lesher, J. H. (1992). Xenophanes of Colophon: Fragments. Toronto:
University of Toronto Press.
Leslie, A. M. (1982). The perception of causality in infants. Perception, 11,
173–186.
Leslie, A. M., & Keeble, S. (1987). Do six-month-olds perceive causality?
Cognition, 25, 265–288.
Leyens, J. P., Cortes, B. P., Demoulin, S., Dovidio, J., Fiske, S. T., Gaunt,
R., et al. (2003). Emotional prejudice, essentialism, and nationalism.
European Journal of Social Psychology, 33, 703–717.
Leyens, J. P., Paladino, P. M., Rodriguez-Torres, R., Vaes, J., Demoulin,
S., Rodriguez-Perez, A., & Gaunt, R. (2000). The emotional side of
prejudice: The attribution of secondary emotions to ingroups and out-
groups. Personality and Social Psychology Review, 4, 186–197.
Liu, T. J., & Steele, C. M. (1986). Attributional analysis as self-affirmation.
Journal of Personality and Social Psychology, 51, 531–540.
Loukatos, D. (1976). Personification of capes and rocks in the Hellenic
seas. In A. Bharati (Ed.), The realm of the extra-human (pp. 467–474).
Paris: Mouton Publishers.
MacDonald, G., & Leary, M. R. (2005). Why does social exclusion hurt?
The relationship between social and physical pain. Psychological Bul-
letin, 131, 202–223.
Maner, J. K., DeWall, C. N., Baumeister, R. F., & Schaller, M. (2007).
Does social exclusion motivate withdrawal or reconnection? Resolving
the “porcupine problem.” Journal of Personality and Social Psychology,
92, 42–55.
Maner, J. K., Kenrick, D. T., Becker, D. V., Robertson, T. E., Hofer, B.,
Neuberg, S. L., et al. (2005). Functional projection: How fundamental
social motives can bias interpersonal perception. Journal of Personality
and Social Psychology, 88, 63–78.
Marcu, A., & Chryssochoou, X. (2005). Exclusion of ethnic groups from
the realm of humanity: Prejudice against the Gypsies in Britain and in
Romania. Psicologia Politica, 30, 41–56.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications
for cognition, emotion, and motivation. Psychological Review, 98, 224
253.
Martel, B. (2006, January 17). Storms payback from God, Nagin says. The
Washington Post, p. A04.
Marx, K. (1959). Economic and philosophic manuscripts of 1844. Mos-
cow: Progress Publishers. (Original work published 1844)
Maslow, A. (1954). Motivation and personality. New York: Harper.
Maslow, A. (1968). Toward a psychology of being (2nd ed.). New York:
Van Nostrand Reinhold.
McFadden, S. H. (1995). Religion and well-being in aging persons in an
aging society. Journal of Social Issues, 51, 161–175.
McIntosh, D. N., Silver, R. C., & Wortman, C. B. (1993). Religion’s role
in adjustment to a negative life event: Coping with the loss of a child.
Journal of Personality and Social Psychology, 65, 812– 821.
Medin, D. L., & Atran, S. (2004). The native mind: Biological categori-
zation, reasoning, and decision making in development and across
cultures. Psychological Review, 111, 960 –983.
Meltzoff, A. N., & Brooks, R. (2001). “Like me” as a building block for
understanding other minds: Bodily acts, attention, and intention. In B. F.
Malle, L. J. Moses, & D. A. Baldwin (Eds.), Intentions and intention-
ality: Foundations of social cognition (pp. 171–191). Cambridge, MA:
MIT Press.
Michael, S. T., Crowther, M. R., Schmid, B., & Allen, R. S. (2003).
Widowhood and spirituality: Coping responses to bereavement. Journal
of Women and Aging, 15, 145–165.
Michotte, A. (1963).