The Class–Compassion Gap How Socioeconomic Factors Influence Compassion

Paul K. Piff and Jake P. Moskowitz

Abstract

Who is more likely to experience compassion: someone who is rich or someone who is poor? In this
chapter, we review how psychological science can shed light on this question. We argue that social class
differences in objective material resources (e.g., income) and corresponding subjective perceptions of
rank produce self- versus other-oriented patterns of social cognition and behavior among upper- and
lower-class individuals, respectively. Extending this framework to the domain of compassion, empirical
studies find that individuals from lower social class backgrounds are more prone to feelings of
compassion and more likely to behave in ways that are compassionate, including sharing with, caring for,
and helping others, relative to individuals from higher social class backgrounds. We describe boundary
conditions and mitigating factors to the class–compassion gap, and conclude by outlining important
questions and lines of inquiry to guide future research.

Key Words: social class, socioeconomic status, SES, compassion, empathy, prosocial behavior,
generosity

In 2014, a group called OckTV posted a video online of what it described as a “social experiment”
(MoeAndET, 2014). In the video, different young men approached strangers in New York City who were
eating pizza and claimed to be hungry before asking them if they would share a slice. The men are repeatedly
turned down, sometimes quite hostilely, until finally one individual, who is ostensibly homeless, shares a slice.
The video’s message is simple: Whereas those who were better off declined to help the person in need, it was
the homeless person with so little to give who readily offered help.
The video has its limitations––for instance, the homeless man had earlier been given some pizza slices––but
it quickly went viral (it has been viewed over 30 million times as of this writing). The video generated
considerable controversy online as comments poured in from people offering their opinions. Some
commenters reflected on how they would respond if asked for food by a stranger. Others agreed with the
video’s conclusions and observed that individuals in more impoverished circumstances tend to be more
compassionate and more giving. Others countered that people work hard for what they have, and that those
with less tend to be less hardworking, reliant on others, even mean and harmful to society. The video had
clearly struck a nerve.
The interest in compassion, and whether it varies as a function of one’s socioeconomic position in society, is
not new. For centuries, scholars and lay people alike have debated the role that social class may play in shaping
the propensity to care for others (e.g., Marx, 1977/1867; Plato, 1987/380 b.c.e.). Distinct lines of reasoning
point in competing directions. On one hand, in many cultures there is a principle of noblesse oblige, stating that

409
those with more resources should assist those who have less (Pratto, Sidanius, Stallworth, & Malle, 1994),
and there are notable examples of wealthy individuals behaving in charitable ways. This indicates that higher
social class may enhance one’s propensity to care for others. On the other hand, there are myriad religious
teachings; folk stories, such as those surrounding Mammon (an evil deity representing riches and money); and
widely held stereotypes that extol the poor and admonish the rich (e.g., Fiske, Cuddy, Glick, & Xu, 2002; see
also Pew Research Center, 2012)––all of which indicate that higher social class may impair compassion.
Guided by mounting interest in how social class influences people’s thoughts, feelings, and actions,
psychological research is providing unique insights into this long-contested question: Do upper- and lower-
class individuals diverge with respect to compassion? We review this emerging research in the current chapter.
In the first section of the chapter, we summarize prior theorizing and research on social class to provide a
theoretical framework for understanding its influences upon compassion. In broad strokes, we argue that
elevated resources and rank among upper-class individuals produce an internal, self-oriented focus: social-
cognitive and behavioral tendencies characterized by greater attention to one’s internal states and goals. By
contrast, reduced resources and rank among lower-class individuals give rise to an external, other-oriented
focus: social-cognitive and behavioral tendencies characterized by heightened focus on the external social
environment and other individuals within it.
In the chapter’s second section, we extend this theoretical framework to the domain of compassion. Guided
by prior theorizing and research in this area, we conceptualize compassion as an affective state characterized
by concern for those who suffer or are vulnerable and a motivation to enhance the welfare of others (e.g.,
Goetz, Keltner, & Simon-Thomas, 2010; Oveis, Horberg, & Keltner, 2010). We review research suggesting
that basic class differences in how people orient to their social worlds manifest in divergent experiences of
compassion, such that individuals from lower social class backgrounds are more prone to feelings of
compassion than individuals from upper social class backgrounds. We also broaden our analysis beyond
compassion as an affective state to the behavioral domain, and describe research finding class differences in
compassionate actions that reflect a relative concern (or disregard) for the welfare of others, including
prosociality––tendencies to share, care, and assist. The third and final section of this chapter outlines
important questions and lines of inquiry to guide future research, and it explores possible boundary conditions
and mitigating factors (perspective-taking, contact, and utilitarianism).

The Emerging Psychology of Social Class

People fold into hierarchies of different kinds based on a number of different dimensions of social life,
including physical dominance, attractiveness, competence, prestige, power, and social class (or socioeconomic
status). Paralleling prior theorizing, we see social class as reflecting, in part, the objective material resources
(e.g., wealth) an individual possesses, which, in turn, evoke inferences and perceptions of one’s own subjective
social class rank vis-à-vis others (see also Kraus, Piff, & Keltner, 2011). As such, an individual’s social class
identity comprises both objective resources and corresponding perceptions of social-class rank. A person’s
material resources are most commonly indexed using “objective” characteristics, such as financial wealth or
income, educational attainment, or occupational prestige (Oakes & Rossi, 2003); whereas subjective
assessments ask individuals to self-identify as belonging to a particular social class category (e.g., “lower class,”
“middle class”) or to rank themselves relative to others in terms of income, education, and occupational

410
prestige (e.g., Adler, Epel, Castellazzo, & Ickovics, 2000; Kraus, Piff, Mendoza-Denton, Rheinschmidt, &
Keltner, 2012; Piff, 2014). Critically, subjective social-class rank is statistically related to objective resources
but contributes to the health and life outcomes of individuals independently of their objective resources. For
example, relative to objective measures of social class (e.g., education, income), subjective social-class rank
more strongly predicts self-rated health and physiological health outcomes such as body-fat distribution (e.g.,
Adler et al., 2000). These results underscore the importance of perceptions of social-class rank to the
experience of social class and its associated outcomes.
Disparities in resources and perceived rank among upper- and lower-class individuals lead them to develop
unique mindsets and perceive, interpret, and react to various situations differently. Abundant resources and
elevated rank afford individuals from upper-class backgrounds increased freedom and control (Kraus, Piff, &
Keltner, 2009), and their lives are relatively protected from external influences, threat, or unpredictability
(Johnson & Krueger, 2005). Upper-class individuals also experience increased geographic and upward social
mobility (Chetty, Hendren, Kline, Saez, & Turner, 2014), improved physical and mental health (Adler et al.,
2000), and more choices in their personal and professional lives (Stephens, Markus, & Townsend, 2007).
Moreover, upper-class environments are more likely to emphasize individuality and value personal
accomplishments and talents (Kohn, 1963; Lareau, 2002). This confluence of increased freedom and control,
greater independence, and reduced reliance on others shifts upper-class individuals in an individualistic
direction, giving rise to self-focused patterns of social cognition and behavior—heightened focus on the
individual, independent self (Kraus et al., 2012; Piff, Kraus, Côté, Cheng, & Keltner, 2010).
The environments and life outcomes of lower-class individuals contrast with those of upper-class
individuals, both objectively and in terms of subjective construal. Lower-class individuals experience greater
threats and environmental unpredictability (e.g., economic instability, more dangerous neighborhoods, poorly
funded schools; Evans, Gonnella, Marcynyszyn, Gentile, & Salpekar, 2005; Gallo & Matthews, 2003), and
they have limited resources to overcome these challenges. Given their more threatening environments, relative
lack of material resources, reduced rank, and decreased personal control, lower-class individuals engage in a
variety of adaptive social-cognitive responses. One such response is a heightened vigilance to the social
environment, which can help increase detection of potential threats and heighten attentiveness to others in the
social environment (Chen & Matthews, 2001; Kraus et al., 2009). A second strategy for coping with stressful
and threatening environments is to engage in affiliative behaviors that build cooperative networks of reciprocal
aid, to help withstand challenges and overcome obstacles (e.g., Bowlby, 1978; Piff, Stancato, Martinez, Kraus,
& Keltner, 2012; Taylor, 2006). According to this theorizing, lower-class individuals engage in tend-and-
befriend responses, which promote other-focused patterns of social cognition and behavior––greater focus on
the needs of others and more compassionate responses to suffering.
In sum, we propose that greater independence and reduced reliance on others lead upper-class individuals
to be more focused on their internal states and goals and less sensitive to the needs of others. By contrast, due
to their increased vigilance to the social environment and their greater investment in the development of
supportive, interdependent bonds, lower-class individuals exhibit increased attentiveness, awareness, and
concern toward others. There are, in turn, numerous empirical demonstrations of class differences in
attentiveness to others, which set the stage for our discussion of how social class influences compassion.

411
Social Class and Attentiveness to Others
An emerging literature in psychology documents divergent patterns of sociality among individuals from
different social class groups. Using a variety of methods and measures across student, online, nationally
representative, and international samples, studies find that upper-class individuals are more focused on and
prioritizing of the self, whereas lower-class individuals are more focused on the states and needs of others (for
a review, see Kraus et al., 2012; Piff, Stancato, & Horberg, 2016). Higher social class, assessed both
objectively (e.g., in terms of parental education) and subjectively (e.g., as one’s perceived socioeconomic rank
in society), is associated with increased narcissism and psychological entitlement––psychological traits
characterized by an increased sense of self-importance and deservingness vis-à-vis others (Cai, Kwan, &
Sedikides, 2012; Foster, Campbell, & Twenge, 2003; Piff, 2014). In a representative sample of close to 3,000
American adults, a composite of annual household income, assets, education, and occupational prestige was
negatively associated with agreeableness—a key factor of personality comprising traits reflecting compassion,
cooperation, and trust (Chapman, Fiscella, Kawachi, & Duberstein, 2010). In other work, lower-class
individuals (as indexed by level of education) reported having more close social relationships, increased levels
of socially engaged emotions (e.g., friendliness, guilt), and decreased levels of socially disengaged emotions
(e.g., pride, anger) compared to upper-class individuals (Na et al., 2010).
Individuals from lower-class backgrounds are more likely to spend time with family members, engage in the
caretaking of others, and have stronger extended-family ties (Argyle, 1994; Lareau, 2002). In one
investigation, lower-class students (assessed in terms of parental educational attainment) were nearly twice as
likely to report having interdependent motives for attending college (e.g., helping their families, giving back to
their communities) than higher-class students (Stephens, Fryberg, Markus, Johnson, & Covarrubias, 2012).
Studies of social interactions reveal similar class differences in behavior. In a study of videotaped interactions
with strangers, upper-class individuals (as indexed by parental education and household income) exhibited
reduced social engagement (e.g., less frequent eye contact, fewer head nods) and greater disengagement-
related behaviors, such as doodling on a questionnaire or fiddling with nearby objects, relative to their lower-
class peers (Kraus & Keltner, 2009).
Class differences in attentiveness to others are further evidenced by studies of empathy––which broadly
refers to processes through which people focus on and relate to the internal states of others (Zaki & Ochsner,
2012). One such process is empathic accuracy, or the ability to correctly infer the thoughts and emotions of
others (Ickes, Stinson, Bissonnette, & Garcia, 1990), which is integral to the experience of compassion.
Research finds that individuals with higher social class are less empathically accurate than their lower-class
counterparts. In one study, Kraus and colleagues (2010) found that, independent of the effects of gender and
agreeableness (factors related to empathic accuracy), participants with a four-year college degree performed
significantly worse on a test of their ability to accurately identify emotions in photographs of human faces
(Kraus, Côté, & Keltner, 2010). In another study, participants with lower subjective socioeconomic status
(SES) more accurately inferred the emotional states of their partners following a mock job interview, even
after controlling for numerous other covariates (e.g., the effects of actor/partner gender, ethnicity, and
agreeableness). In a follow-up experiment, participants made to feel higher in subjective SES––by comparing
themselves to someone at the very bottom of the socioeconomic ladder––were less effective at estimating the
emotional states of individuals based on images of the eye-region of the face (“Reading the Mind in the Eyes”

412
test; Baron-Cohen, Wheelwright, Hill, Raste, & Plumb, 2001), relative to participants made to feel lower in
subjective status––who had compared themselves to someone at the very top of the socioeconomic ladder. In
addition to an objective indicator of higher social class standing (education), mere perceptions of relatively
higher social class standing––even those triggered via moment-to-moment social comparison processes––seem
to dampen empathic accuracy (Kraus, Côté, & Keltner, 2010). That an objective resource measure and a
subjective social class rank-based manipulation similarly predicted empathic accuracy suggests that objective
social class and subjective social class rank both influence class-based psychological outcomes. More broadly,
these results highlight the importance of the social context in shifting the experience of subjective social-class
rank and class-based patterns of emotion––a point we will return to in our discussion.
Neural imaging studies further underscore the link between lower social class and increased empathy.
Research indicates that the mentalizing neural network––a set of brain regions that includes the dorsomedial
prefrontal cortex and posterior cingulate cortex and is implicated in tasks that require understanding the
mental states of others (Frith & Frith, 2006; Lieberman, 2010; Mitchell, 2009; Zaki, Weber, Bolger,
Ochsner, 2009)––is more active in lower-class individuals. In one investigation, students with lower subjective
SES experienced greater activation of the mentalizing neural network while reading a fictional first-person
narrative of a student describing their thoughts and feelings about the beginning of a new college quarter
(Muscatell et al., 2012). In a similar vein, Varnum and colleagues (2015) studied patterns of fronto-central P2
activation––a neural marker of empathy––among upper- and lower-class individuals in response to images of
others in pain (Varnum, Blais, Hampton, & Brewer, 2015). A composite of income, highest parental
education, and subjective SES was negatively related to fronto-central P2 activation, indicating that higher
social class was linked to diminished neural empathic responses toward the suffering of others.
The research we have reviewed highlights an intriguing pattern in the literature: lower social class is
associated with increased interdependence, greater other-focus, and heightened empathic responding.
Although “empathy,” which generally refers to the vicarious experiences of another’s emotions, is
differentiable from “compassion”––a specific emotion triggered by another’s suffering and that motivates
desires to help (Goetz et al., 2010)––these findings indicate that social class may underlie differential
experiences of compassion.

Social Class Differences in Compassion

In a major review, Goetz, Simon-Thomas, and Keltner (2010) define compassion as a “feeling that arises in
witnessing another’s suffering and that motivates a subsequent desire to help” (Goetz et al., 2010). According
to this conceptualization, compassion is an internal affective state, not an attitude or general behavioral
response, and although it is conceptually and experientially related to feelings of sympathy, empathy, or pity––
all of which can arise in response to the suffering of others––it is distinguishable from these insofar as it
entails a further desire to help improve the experience of the suffering individual (see also Gilbert, 2015).
Certain individuating processes that enhance attention to the individual self or are otherwise inimical to
interdependence exert a powerful influence on compassion and compassion-related states. For example,
interdependent self-construal and collectivism (i.e., viewing oneself as part of a whole) have been positively
related to experiences of compassion and sympathy in both American and Japanese samples (Dalsky, Gohm,
Noguchi, & Shiomura, 2008; Uchida & Kitayama, 2001). Moreover, individuals who are more self-focused,

413
narcissistic, or self-enhancing show reduced empathic tendencies and perspective taking (Watson, Grisham,
Trotter, & Biderman, 1984; Eysenck, 1981). Insofar as upper-class individuals are more self-focused, less
attentive to others, and less empathetic, they should also experience less compassion.
Research on social power is also relevant to our discussion of possible class differences in compassion.
Although social class and “power”––which refers to asymmetrical control over resources in social
relationships––are conceptually and empirically distinct (e.g., Kraus et al., 2012), higher social class, by being
associated with greater material resources, might increase people’s psychological feelings of powerfulness,
which studies have linked to reduced compassionate responses. In one study, participants with increased
feelings of power, as measured through the Sense of Power scale (e.g., “In my relationships with others, I can
get others to do what I want”; Anderson & Galinsky, 2006), reported feeling less compassion than their low-
power counterparts after listening to a lab partner describe a recent event that had “caused them a great deal of
emotional suffering and pain” (van Kleef et al., 2008). These findings link feelings of powerlessness to
increased compassionate responding and indicate that lower-class individuals (who may feel less powerful)
may be more compassionate than their upper-class counterparts.

Direct Evidence Linking Social Class to Experiences of Compassion

The research we have reviewed on social class, attentiveness to others, and empathy, as well as the findings
we have described on individual-level determinants of compassion (e.g., interdependence, power), indicate
that social class may influence responses to others’ suffering. Select studies to date have specifically examined
the association between social class and compassion, and their findings yield limited but suggestive evidence in
support of the claim that lower-class individuals are more prone to compassion than their upper-class
counterparts.
Stellar, Manzo, Kraus, and Keltner (2012) conducted one of the most pertinent investigations in this realm,
testing across multiple studies the specific association between social class and compassion. In an initial study,
participants were asked to indicate their current level of social class (e.g., “lower class,” “middle class”) and
complete the Dispositional Positive Emotion Scale (DPES), a well-validated measure of an individual’s
propensity (or trait-like tendency) to experience several distinct positive emotions, such as love, awe, pride,
and, critically, compassion (Shiota, Keltner, & John, 2006). In this measure, participants indicate their
agreement with a variety of statements that reference core appraisals and experiences of each target emotion.
Five items index compassion, capturing such things as attentiveness to others in need of help (“I often notice
people who need help”) and the motivation to act on their behalf (“When I see someone hurt or in need, I feel
a powerful urge to take care of them”). Individuals who categorized themselves as higher in social class
reported experiencing less compassion than individuals categorizing themselves as lower in social class, and
this trend remained significant when controlling for relevant covariates (gender, ethnicity, and spirituality).
Further analyses revealed no significant correlations between social class and any of the six other positive
emotions captured by DPES (joy, contentment, pride, love, amusement, and awe), suggesting that lower-class
individuals’ increased compassion was not attributable to a tendency to report increased positive emotion in
general.
In addition to assessing self-reported tendencies to experience compassion in daily life, Stellar and
colleagues (2012) measured people’s emotional reactions in a specific situation to an individual who is

414
suffering or otherwise vulnerable––conditions that are both central to the experience of compassion as well as
ideal for its study (Goetz et al., 2010). In one follow-up study, participants viewed two 2.5-minute videos: a
neutral-emotion video explaining the construction of a patio (which provided a baseline condition for
comparison purposes), and a compassion-inducing video depicting the experiences of children with cancer and
their families as they cope with the challenges of chemotherapy. After each video, participants were asked to
report how much they experienced feeling “compassion/sympathy,” as well as several other emotions (e.g.,
happy, inspired) while watching the video. The videos were separated by a five-minute resting period, and
heart rate was measured throughout using ECG, allowing the researchers to examine whether heart rate
deceleration––a physiological reaction associated with orienting toward and engaging with others (Eisenberg
et al., 1989)––would vary by social class alongside self-reported compassion. Paralleling the results of their
first study, but this time with an objective resource-based measure of social class (family income and parental
education), lower-class participants reported greater overall increases in compassion from the neutral video to
the compassion-inducing video than upper-class participants did. A similar pattern was observed in the heart
rate deceleration of participants: lower-class individuals exhibited greater levels of heart rate deceleration from
the neutral video to the compassion-inducing video than upper-class participants did.
In a final study by Stellar and colleagues that yielded complementary evidence (Stellar et al., 2012, Study 3),
participants engaged in a competitive mock job interview. The experimenter interviewed study participants in
pairs as they sat across from each other for a hypothetical psychology laboratory manager position.
Participants were told that the best interviewees would receive a cash prize. After the interview, participants
rated both their and their partners’ experience of a set of 20 positive and negative emotions, including
compassion, during the job interview. Consistent with the authors’ predictions, and paralleling the results
from their prior two studies, lower-class participants (assessed in terms of family income and parental
education) reported feeling more compassion for their partner, even when controlling for a number of other
factors, including gender, ethnicity, and partners’ social class. The study further revealed one mechanism
underlying class differences in compassion, and it harks back to our earlier discussion of class differences in
attentiveness to others. Specifically, lower-class individuals perceived greater distress in their partners during
the mock interview, an admittedly stressful and anxiety-provoking task, which statistically mediated their
increased compassion. In sum, increased compassion among lower-class individuals is attributable, in part, to
their increased attentiveness to others’ suffering.
A separate investigation by Côté, Piff, and Willer (2013) mirrored the findings of Stellar et al. (2012). The
central question in the work of Côté and colleagues was whether upper- and lower-class individuals would be
differentially utilitarian in their moral decision-making––specifically, are there social class differences in
people’s willingness to commit a smaller harm in the service of the greater good? Across several studies, the
researchers found that upper-class participants behaved in a more utilitarian fashion: they were more willing
to harm one person in order to maximize gains for the group.
Most pertinent to our review is why these class differences in utilitarian decision-making emerged in the
first place. The researchers hypothesized that upper-class individuals would make more utilitarian judgments
because they feel less compassion for those harmed by this type of judgment than lower-class individuals. In
one representative study, the researchers examined class differences in feelings of compassion for the losing
player of a virtual economic game (Côté et al., 2013, Study 2). Study participants were told that they had been

415
chosen to play the role of the “decider” and could choose to take money from a designated online player (“lose
member”) in order to maximize monetary gains for the other group members. Following the game,
participants reported how much compassion, sympathy, and other empathic emotions they felt for the losing
member––whose economic well-being would have been directly impacted. Consistent with their predictions,
and even when accounting for several other factors (gender, age, ethnicity, religiosity, and political
orientation), higher-income participants reported reduced feelings of compassion for the losing player, relative
to their lower-income counterparts, and these class differences in compassion accounted for the association
between social class and utilitarian judgment.
In a follow-up study, Côté et al. (2013) tested whether class differences in compassion are attributable more
generally to class differences in emotional reactivity—for example, it may be the case that lower-class
individuals are simply more reactive to any emotion-eliciting stimulus regardless of its content or valence. To
investigate this possibility, the researchers presented participants with sets of images that elicited either pride
(e.g., pictures of U.S. national landmarks), amusement (e.g., pictures of laughing monkeys), or compassion
(e.g., pictures of helplessness, vulnerability). After viewing the slides, participants reported their felt level of
each emotion, confirming that the manipulations were successful in eliciting their respective target emotion.
As expected, lower-income participants reported greater compassion in response to viewing images of others
suffering than upper-income participants, but no income differences emerged in response to images eliciting
pride or amusement––suggestive evidence that higher social class is specifically associated with reduced
reactivity to stimuli that elicit compassion.
These initial findings indicate that social class shapes people’s levels of compassion and are consistent with
the claim that individuals from lower-class backgrounds experience greater compassion than individuals from
upper-class backgrounds. Across both objective resource-based measures (e.g., income) and subjective rank-
based measures, lower-class individuals report experiencing more compassion in daily life, as well as respond
with greater compassion––as assessed via self-reports and compassion-related peripheral physiology––when
confronting the suffering of others. Moreover, these class differences in compassion are not reducible to
factors that may co-vary with social class (e.g., ethnicity, political orientation, religiosity), nor are they
attributable to more general class differences in emotional responding. The stronger compassionate
responding of lower-class individuals, we have theorized, may strengthen social connections that can
subsequently help them cope with their more threatening and resource-poor environments.

How Social Class Influences Compassionate Action: Sharing, Caring, and Assisting
As part of the growing scientific interest in understanding the psychological effects of social class,
investigators have explored how social class influences not only people’s thoughts and feelings, but also their
behavioral responses toward others in the social environment (for a review, see Kraus et al., 2012; Piff et al.,
2016). Investigations in this realm have documented numerous class differences in prosocial behavior––
tendencies to share, care, and assist (Keltner, Kogan, Piff, & Saturn, 2014)––both in the context of a specific
behavioral response initiated by the perception of need, as well as in terms of more diffuse or cooperative
forms of helping (e.g., helping an anonymous target who is not identified as suffering or being in a state of
dire need).
In describing research on class differences in compassionate actions that reflect a relative concern (or

416
disregard) for the concerns of others, we do not mean to suggest that compassion and prosocial behavior are
one and the same; indeed, there is reason to differentiate them. First and foremost, we have conceptualized
compassion as a specific subjective feeling state that is triggered by the suffering of others and that motivates
desires to help, which parallels prior treatments of the construct (e.g., Gilbert, 2015; Goetz et al., 2010). As
such, compassion is distinct from a general benevolent response to others regardless of suffering. Second,
although feelings of compassion can trigger subsequent helping behavior, compassion and helping are not
reducible to one another: compassion does not always translate to helping behavior––for example, when the
costs of helping are deemed too high (Keltner et al., 2014; see also Cameron & Payne, 2011; and Cameron
D., this volume); and helping behavior can be motivated by various states besides compassion, including
gratitude (e.g., Bartlett & DeSteno, 2006), moral elevation (Schnall, Roper, & Fessler, 2010), and awe (Piff,
Dietze, Feinberg, Stancato, & Keltner, 2015).
Notwithstanding these important distinctions, there is a strong theoretical and empirical overlap between
compassion as an affective state and compassionate acts, such as helping and generosity (Gilbert, 2015),
underscoring the relevance of research on class differences in prosociality to our review of class and
compassion. Importantly, the motivation to alleviate suffering is a defining feature of compassion (Goetz et
al., 2010). Although this motivation can be expressed in a variety of acts, among the most well-documented is
that of helping behavior. Consistent with this notion, research finds that individuals with increased propensity
to feelings of compassion display greater levels of prosociality. For example, in a nationwide representative
sample of adults, individuals who reported experiencing greater compassion behaved in a more generous
fashion toward an anonymous study partner, even after accounting for the effects of other positive prosocial
emotions (e.g., love; Piff et al., 2015; see also Batson, 1998; Batson & Shaw, 1991). To the extent that lower-
class individuals are more attentive to the needs of others than their upper-class counterparts, as we have
argued and as prior research documents, they should not only be more likely to feel compassion, but also to
behave in ways that are compassionate, by prioritizing the needs and concerns of others.
Nationwide surveys of charitable giving in America frequently find what is referred to as a “giving gap”:
lower-income households often give a larger percentage of their annual incomes to charity than upper-income
households do (Gipple & Gose, 2012; Greve, 2009; James & Sharpe, 2007; Johnston, 2005; Independent
Sector, 2001). For example, in a large-scale analysis conducted by the Chronicle of Philanthropy of IRS
charitable giving records, individuals with yearly incomes of $50,000–$75,000 donated an average of 7.6% of
their discretionary income to charity, whereas those with yearly incomes above $100,000 contributed 4.2% of
their discretionary income to charity (Gipple & Gose, 2012; however, for a different perspective, see
Korndörfer, Egloff, & Schmukle, 2015). These patterns suggest, somewhat provocatively, that those with less
may at times be more giving.
Laboratory studies in which the opportunity to respond prosocially was an observed reaction to another
individual, often in conditions of anonymity, yield corroborating results. In one study using the Dictator
Game, a well-validated measure of generosity in which participants split points between themselves and an
anonymous other (e.g., Forsythe, Horowitz, Savin, & Sefton, 1994), individuals lower in subjective SES
donated more credits––which would later be exchanged for cash––to their partner than did individuals higher
in subjective SES (Piff et al., 2010, Study 1). In another experiment, individuals made to feel relatively lower
in social class rank, by comparing themselves to someone at the very top of the socioeconomic ladder,

417
endorsed increased charitable donations, relative to individuals primed to feel relatively higher in social class
rank, by comparing themselves to someone at the very bottom of the social ladder (Piff et al., 2010, Study 2).
This study simultaneously revealed an independent effect of objective social class: lower-income individuals
were also more charitable than upper-income individuals, even when adjusting for the effects of manipulated
social class rank. These results indicate that alongside objective social class, subjective perceptions of relative
social class rank––even as triggered via fleeting social comparisons––can alter compassionate behavior.
There is some initial evidence that class-related differences in giving may generalize to diverse measures of
social class and generosity (e.g., Kirkpatrick, Delton, de Wit, & Robertson, 2015), and perhaps to even
different age groups and cultural contexts (e.g., Chen, Zhu, & Chen, 2013; Kirkpatrick, Delton, de Wit, &
Robertson, 2015; Miller, Kahle, & Hastings, 2015). In one study in this realm, researchers observed the
giving patterns of preschool children whose families varied in terms of wealth. Children from lower-income
families acted more altruistically than children from wealthier families, donating a greater number of prize
tokens to anonymous sick children (Miller et al., 2015). Despite this preliminary evidence, the question of
how social class differences in prosociality manifest across different populations and cultural contexts is an
important, and largely open, extension of the research we have reviewed.
Whereas lower-class individuals may be more prosocial, research finds that upper-class individuals are more
likely to behave in ways that harm others or are otherwise unethical for self-gain (for a review, see Piff et al.,
2016). In one relevant line of work (Piff, Stancato, Côté, Mendoza-Denton, & Keltner, 2012), individuals
higher in subjective SES were more likely to endorse various unethical behaviors, such as accepting bribes or
deceiving others, and were more likely to cheat in a game in order to increase their chances of winning a cash
prize. In two field studies that were also a part of this investigation, drivers with more expensive vehicles (a
real-world proxy for wealth) were more likely to cut off other drivers at a four-way intersection, and to fail to
yield for a pedestrian––actually a study confederate––waiting to cross at a crosswalk, actions that are not only
unlawful but that also reflect a disregard for the concerns of others (see also Blanco et al., 2008; Dubois,
Rucker, & Galinsky, 2015; Johns & Slemrod, 2010; Konigsberg, MacGregor, Johnson, Massey, & Daubman,
2013; Lyons et al., 2012; Wang & Murnighan, 2014).
What are the mechanisms underlying these class differences in prosocial behavior? Piff and colleagues
(2010, Study 4) conducted an experiment that underscores compassion as one potent driver of class-related
differences in prosociality. Piff and colleagues theorized that class differences in sensitivity to the welfare of
others and feelings of compassion might explain why lower-class individuals engage in more prosocial
behavior than upper-class individuals do. They proceeded to test this question in the context of giving
participants the opportunity to help a person in distress.
Study participants were randomly assigned to one of two experimental conditions. In the control condition,
participants viewed a video of two actors talking quietly to one another in a courtroom; this video was
designed to elicit a neutral emotional response. In the compassion-induction condition, participants watched a
video on child poverty, which elicited compassion by showing images of suffering and vulnerability. Shortly
after the video, a supposed lab partner––actually a confederate in the experiment––rushed into the lab,
apologized for being late, and asked whether she could still participate, all the while displaying classic
nonverbal signs of distress (e.g., widening of the eyes, brow movements up and in; Eisenberg et al., 1989;
Gross & Levenson, 1993). The partner was seated in an adjacent room, and then, ostensibly to make up for

418
lost time due to the partner’s late arrival, participants were asked to help divide a list of tasks, each with a
specified time requirement, between themselves and their partner. The total duration of the chosen tasks
served as a behavioral measure of helping: the more minutes participants volunteered to help their distressed
partner in the study, the more helpful they were being.
Piff and colleagues predicted that if lower-class individuals are, by default, more prone to feelings of
compassion and subsequent prosocial behavior, then they should exhibit high helping in both the compassion
and neutral conditions. On the other hand, upper-class individuals, who by default exhibit reduced
compassion and prosocial behavior, should exhibit high helping in the compassion condition but not in the
neutral condition. In other words, if class differences in prosociality are rooted, in part, in class differences in
compassion, then inducing in upper-class participants increased feelings of compassion should cause them to
be just as helpful as their lower-class counterparts.
Consistent with predictions, lower-income participants in the neutral condition were found to help their
partners to a greater extent than higher-income participants, mirroring past findings showing reduced
prosociality among higher-class individuals. However, experimentally inducing compassion via a brief video
about child poverty caused these class differences in helping behavior to disappear: inducing compassion
caused upper-income participants to exhibit levels of helping behavior that matched their lower-income
peers’. These findings help underscore two important points. First, they indicate that class-based differences
in prosociality may be attributable, at least in part, to class differences in baseline levels of compassion.
Second, these results indicate that class-based divergences in compassionate behavior are not immutable, but
rather sensitive to even brief reminders of the needs of others and “nudges” of compassion.
This research indicates that class differences in compassionate behavior parallel the class differences in
feelings of compassion we described earlier. Relative to upper-class individuals, lower-class individuals exhibit
increased tendencies to share with, care for, and help others, tendencies that arise, in part, out of their
increased attentiveness to others and baseline levels of compassion. Importantly, when upper-class individuals
are induced to feel compassion (for instance, via reminders of the suffering of others), they may become as
prosocial as lower-class individuals. Class differences in compassionate action, it would seem, do not arise
from a diminished capacity but rather from a reduced default tendency toward compassion among upper-class
individuals. What, then, determines when and where compassion on the part of the upper class will emerge?
What factors may help promote it? We turn to these questions in our final section.

Open Questions and Future Directions in Research on Social Class and Compassion
In this chapter, we have described an emerging pattern in the study of compassion––upper-class individuals
display less compassion than their lower-class counterparts do (Côté et al., 2013; Piff et al., 2010; Stellar et al.,
2012). Though provocative, the research we have described is also preliminary, and many important questions
remain. For example, do these effects extend to extremes of wealth and poverty and other instantiations of
compassion? And might class differences in compassion vary as a function of who is suffering––for example,
whether the person is rich or poor?
One central question concerns the origins of class differences in compassion, and in particular what aspects
of social class underlie these divergences. We have argued that “social class” comprises both objective resources
and subjective social-class rank, and that together these two components produce an internal, self-oriented

419
focus among upper-class individuals, and an external, other-oriented focus among lower-class individuals––
broad social-cognitive divergences that extend to class differences in compassion. We have reviewed research
indicating that subjective perceptions of relative class rank, and even context-specific social comparisons that
shift these perceptions, are associated with compassionate responding in ways that parallel more objective,
resource-based measures of social class. These findings suggest that class differences in compassion are very
likely to be multiply determined, originating from a confluence of cultural, environmental, and social-
cognitive factors tied to objective characteristics of social class as well as subjective perceptions of social class
rank.
It will be important to build on these findings by examining what aspects of the class-compassion gap are
relatively enduring or stable, versus shaped by the social context. For example, structural features of upper-
class environments, such as relative independence from others and increased autonomy, may provide fewer
opportunities for social interaction and reduced experiences with tending to the needs of others, both of which
may attenuate compassion. Compassion may also be less normative in upper-class contexts, which could
decrease the social value of compassion among individuals who identify as upper class (e.g., Lareau, 2002).
Future research should delineate to what extent relatively enduring features of higher social class, including
cultural values, identity, and autonomy, shape chronically lower feelings of compassion (e.g., Snibbe &
Markus, 2005; Stephens et al., 2007).
Compassion is also influenced by perceptions of social-class rank, which, as we have discussed, can shift as a
function of the social context and situation-specific perceptions of relative standing. This cautions researchers
against drawing firm conclusions about class differences in compassion as essential or categorical; rather, class
patterns of compassion are heavily influenced by contextual and situational factors. Perceptions of higher
social class rank may be more chronically activated among individuals with increased objective resources (e.g.,
greater wealth, higher education; Kraus et al., 2012), which could shape habituated responses to the social
environment. At the same time, social contexts––and social comparisons––that induce perceptions of
relatively lower social class rank can increase compassionate responding, even among individuals who are
objectively wealthy (e.g., Piff et al., 2010).
Future investigations should seek to better understand the pathways through which perceptions of relatively
decreased social class rank might independently increase compassion. Social rank is a fundamental means by
which people navigate the social realm, and it determines the individual’s privileged access to valued resources
(e.g., reproductive partners; Keltner, van Kleef, Chen, & Kraus, 2008). We argue that perceptions of one’s
own relatively lower-class rank (even those that arise in situ) trigger a heightened vigilance of the social
context and an other-focused social orientation, which are well-documented and adaptive strategies of lower-
rank individuals navigating more unstable environments (e.g., Kraus et al., 2009; Piff et al., 2010), both of
which may increase compassion. We posit that the effects of perceived social class rank are parallel to, but
distinguishable from, those of other rank-based processes that seem to similarly shift patterns of self- versus
other-focus, such as power (e.g., Guinote & Vescio, 2010) or winning in a competition (Schurr & Ritov,
2016)––an intriguing direction for research.
The mutability of class differences in compassion is further underscored by research we have described
finding that compassionate, prosocial action among upper-class individuals is responsive to psychological
intervention, such as brief reminders of the needs of others (e.g., Piff et al., 2010). These findings cast further

420
doubt on the notion that class differences in compassion reflect a reduced capacity for compassion among the
upper class. Rather, upper- and lower-class individuals may have differential motivations surrounding
compassion. One possibility is that upper-class individuals regulate or even avoid compassion, because they
expect that compassion inhibits self-relevant goals, whereas lower-class individuals may place more value on
compassion because they believe it advances their goals. Research on social power lends credence to this
motivational perspective, finding that high-power individuals may engage in emotion-regulation in response
to others’ suffering, thereby reducing their overall feelings of compassion (van Kleef et al., 2008). In the
future, studies should examine whether upper- and lower-class individuals hold differing motivations
surrounding compassion, and whether altering these motivations––for instance, by priming the relative
benefits versus costs of compassion––can lead to downstream shifts in compassionate responding.
One notable implication of the patterns we have described is that the people perhaps best positioned to care
for others—due to their elevated rank and resources—are seemingly the least likely to do so. Upper-class
individuals’ reduced compassion for the plight of others may exacerbate class divides in a socioeconomic
climate already rife with inequality (e.g., Saez & Zucman, 2014). What, then, can be done to alter this trend?
We discuss three promising factors—perspective-taking, contact, and utilitarianism—that point toward
important future directions and possible interventions.
One route to compassion involves transforming self-focus into perspective-taking. As we have seen, upper-
class individuals are more self-focused and less attentive to others in the social environment. Reduced
tendencies to see the world through others’ eyes may signify that upper-class individuals, at the level of basic
perception, are less cognizant of the suffering of others, even when confronted directly by it. Social
psychological research has shown that taking another’s perspective––for example, by imagining how she is
thinking and feeling––enhances empathy and compassion (e.g., Coke, Batson, & McDavis, 1978; Eisenberg
& Miller, 1987; Underwood & Moore, 1982). To the extent that upper-class individuals hold the view that
perspective-taking confers a disadvantage, a possibility we alluded to before, one effective strategy may be to
tie perspective-taking to preexisting values of self-interest, thereby harnessing the very self-interest that would
otherwise mitigate perspective-taking and compassion. Future investigations could identify effective
interventions to increase chronic perspective-taking among upper-class individuals, such as through explicit
training programs (e.g., compassion training; Jazaieri et al., 2016).
A related second avenue to increasing upper-class compassion involves inter-group contact. People tend to
surround themselves with those of similar class backgrounds; consequently, affluent individuals living in
comfortable environments may be less frequently confronted with the suffering of others. This could
undermine their feelings of personal efficacy when encountering suffering––in a sense, not knowing how to
respond. Moreover, affluent individuals may be more likely to categorize individuals who are suffering as
dissimilar to themselves (or as belonging to an out-group), perceptions that can hamper compassionate
responding (e.g., Cikara, Bruneau, Van Bavel, & Saxe, 2014; Tarrant, Dazeley, & Cotton, 2009). One way to
short-circuit these mitigating factors is through prolonged, interdependent contact with socioeconomically
diverse others (e.g., Aronson & Patnoe, 2011). Increased personal contact with those of lower social classes
might help upper-class individuals build skills for effectively responding to the suffering of others, enhance
their feelings of closeness and understanding, and broaden their definition of the in-group to include those
who are less socioeconomically advantaged. An interesting direction for research in this vein would be to test

421
whether experimentally induced cross-class contact can increase compassion across social class boundaries
(e.g., Page-Gould, Mendoza-Denton, & Tropp, 2008).
A final route to compassion involves harnessing the utilitarian proclivities of upper-class individuals. We
have described studies finding that when making certain moral decisions, upper-class individuals are
significantly more likely to prioritize the needs of the many over the few; they strive to maximize the greatest
good for the greatest number of people (Côté et al., 2013). Given that previous research has tested class
differences in compassion toward specific individuals who are suffering, it may be that upper-class individuals
dampen their feelings of compassion in these instances because they deem them less utilitarian. Following this
lead, upper-class individuals may be most inclined to respond with compassion to issues that they believe
negatively affect a large number of people, such as instances of collective or mass suffering. Future
investigations could examine whether framing instances of suffering in more collective ways––for example, by
underscoring how one individual’s suffering implicates many others––can boost compassion among upper-
class individuals. In sum, upper-class compassion may be facilitated through processes that link compassion to
utilitarianism.

Conclusion
People have made the case that our culture today is experiencing a compassion deficit. Over the past 50
years, people have become more individualistic, more self-focused, less connected to others, and less
empathetic (e.g., Twenge, Campbell, & Freeman, 2012), all trends that could signify societal declines in
compassion. It is perhaps partly for this reason that, in recent years, scientific interest in compassion, and the
factors that promote it, has surged. In this chapter, we have reviewed findings indicating that a person’s
propensity toward compassion is tied to her/his social class, such that increased social class is frequently
associated with less compassionate responding. We have outlined some of the possible psychological factors
underlying these associations, and described certain psychological, situational, and structural factors that could
mitigate them. Future research should build on these initial insights to better our understanding of the class-
compassion gap and, in turn, how it may be bridged.

References
Adler, N. E., Epel, E. S., Castellazzo, G., & Ickovics, J. R. (2000). Relationship of subjective and objective social status with psychological and
physiological functioning: Preliminary data in healthy, white women. Health Psychology, 19(6), 586–592.
Anderson, C., & Galinsky, A. D. (2006). Power, optimism, and risk-taking. European Journal of Social Psychology, 36(4), 511–536.
Argyle, M. (1994). The Psychology of Social Class. London: Routledge.
Aronson, E., & Patnoe, S. (2011). Cooperation in the Classroom: The Jigsaw Classroom. London: Pinter and Martin.
Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y., & Plumb, I. (2001). The “Reading the Mind in the Eyes” test, revised version: A study
with normal adults, and adults with Asperger syndrome or high-functioning autism. Journal of Child Psychology and Psychiatry, 42(2), 241–
251.
Bartlett, M. Y., & DeSteno, D. (2006). Gratitude and prosocial behavior: Helping when it costs you. Psychological Science, 17(4), 319–325.
Batson, C. D. (1998). Altruism and prosocial behavior. In S. T. Fiske, D. T. Gilbert, & G. Lindzey (Eds.), Handbook of Social Psychology (pp.
282–316). Boston, MA: McGraw-Hill.
Batson, C. D., & Shaw, L. L. (1991). Evidence for altruism: Toward a pluralism of prosocial motives. Psychological Inquiry, 2(2), 107–122.
Blanco, C., Grant, J., Petry, N. M., Simpson, H. B., Alegria, A., Liu, S., & Hasin, D. (2008). Prevalence and correlates of shoplifting in the
United States: Results from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). American Journal of
Psychiatry, 165(7), 905–913.
Bowlby, J. (1978). Attachment theory and its therapeutic implications. Adolescent Psychiatry, 6, 5–33.

422
Cai, H., Kwan, V. S., & Sedikides, C. (2012). A sociocultural approach to narcissism: The case of modern China. European Journal of
Personality, 26(5), 529–535.
Cameron, C. D., & Payne, B. K. (2011). Escaping affect: How motivated emotion regulation creates insensitivity to mass suffering. Journal of
Personality and Social Psychology, 100(1), 1–15.
Chapman, B. P., Fiscella, K., Kawachi, I., & Duberstein, P. R. (2010). Personality, socioeconomic status, and all-cause mortality in the United
States. American Journal of Epidemiology, 171(1), 83–92.
Chen, E., & Matthews, K. A. (2001). Cognitive appraisal biases: An approach to understanding the relation between socioeconomic status and
cardiovascular reactivity in children. Annals of Behavioral Medicine, 23(2), 101–111.
Chen, Y., Zhu, L., & Chen, Z. (2013). Family income affects children’s altruistic behavior in the Dictator Game. PLoS ONE, 8(11), e80419.
Chetty, R., Hendren, N., Kline, P., Saez, E., & Turner, N. (2014). Is the United States still a land of opportunity? Recent trends in
intergenerational mobility. National Bureau of Economic Research (NBER Working Paper No. 19844).
Cikara, M., Bruneau, E., Van Bavel, J. J., & Saxe, R. (2014). Their pain gives us pleasure: How intergroup dynamics shape empathic failures
and counter-empathic responses. Journal of Experimental Social Psychology, 55, 110–125.
Coke, J. S., Batson, C. D., & McDavis, K. (1978). Empathic mediation of helping: A two-stage model. Journal of Personality and Social
Psychology, 36(7), 752–766.
Côté, S., Piff, P. K., & Willer, R. (2013). For whom do the ends justify the means? Social class and utilitarian moral judgment. Journal of
Personality and Social Psychology, 104(3), 490–503.
Dalsky, D., Gohm, C. L., Noguchi, K., & Shiomura, K. (2008). Mutual self-enhancement in Japan and the United States. Journal of Cross-
Cultural Psychology, 39(2), 215–223.
Dubois, D., Rucker, D. D., & Galinsky, A. D. (2015). Social class, power, and selfishness: When and why upper and lower class individuals
behave unethically. Journal of Personality and Social Psychology, 108(3), 436–449.
Eisenberg, N., Fabes, R. A., Miller, P. A., Fultz, J., Shell, R., Mathy, R. M., & Reno, R. R. (1989). Relation of sympathy and personal distress
to prosocial behavior: A multimethod study. Journal of Personality and Social Psychology, 57(1), 55–66.
Eisenberg, N., & Miller, P. A. (1987). The relation of empathy to prosocial and related behaviors. Psychological Bulletin, 101(1), 91–119.
Evans, G. W., Gonnella, C., Marcynyszyn, L. A., Gentile, L., & Salpekar, N. (2005). The role of chaos in poverty and children’s
socioemotional adjustment. Psychological Science, 16(7), 560–565.
Eysenck, S. B. (1981). Impulsiveness and antisocial behaviour in children. Current Psychological Research, 1(1), 31–37.
Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow
from perceived status and competition. Journal of Personality and Social Psychology, 82(6), 878–902.
Forsythe, R., Horowitz, J. L., Savin, N. E., & Sefton, M. (1994). Fairness in simple bargaining experiments. Games and Economic Behavior,
6(3), 347–369.
Foster, J. D., Campbell, W. K., & Twenge, J. M. (2003). Individual differences in narcissism: Inflated self-views across the lifespan and around
the world. Journal of Research in Personality, 37(6), 469–486.
Frith, C. D., & Frith, U. (2006). The neural basis of mentalizing. Neuron, 50(4), 531–534.
Gallo, L. C., & Matthews, K. A. (2003). Understanding the association between socioeconomic status and physical health: Do negative
emotions play a role? Psychological Bulletin, 129(1), 10–51.
Gilbert, P. (2015). The evolution and social dynamics of compassion. Social and Personality Psychology Compass, 9(6), 239–254.
Gipple, E., & Gose, B. (2012, August 19). America’s generosity divide. The Chronicle of Philanthropy. Retrieved February 26, 2016, from
https://philanthropy.com/article/America-s-Generosity-Divide/156175
Goetz, J. L., Keltner, D., & Simon-Thomas, E. (2010). Compassion: An evolutionary analysis and empirical review. Psychological Bulletin,
136(3), 351–374.
Greve, F. (2009, May 23). America’s poor are its most generous. The Seattle Times, Nation & World. Retrieved February 26, 2016, from
http://www.seattletimes.com/nation-world/americas-poor-are-its-most-generous-donors/
Gross, J. L., & Levenson, R. W. (1993). Emotional suppression: Physiology, self-report, and expressive behavior. Journal of Personality and
Social Psychology, 64(6), 970–986.
Guinote, A., & Vescio, T. (2010). The Social Psychology of Power. New York: Guilford Press.
Ickes, W., Stinson, L., Bissonnette, V., & Garcia, S. (1990). Naturalistic social cognition: Empathic accuracy in mixed-sex dyads. Journal of
Personality and Social Psychology, 59(4), 730–742.
Independent Sector. (2001). Giving and volunteering in the United States. Washington, DC. Retrieved February 26, 2016, from
http://www.cpanda.org/pdfs/gv/GV01Report.pdf
James, R. N., III, & Sharpe, D. L. (2007). The nature and causes of the U-shaped charitable giving profile. Nonprofit and Voluntary Sector
Quarterly, 36(2), 218–238.

423
Jazaieri, H., Lee, I. A., McGonigal, K., Jinpa, T., Doty, J. R., Gross, J. J., & Goldin, P. R. (2016). A wandering mind is a less caring mind:
Daily experience sampling during compassion meditation training. The Journal of Positive Psychology, 11(1), 37–50.
Johns, A., & Slemrod, J. (2010). The distribution of income tax noncompliance. National Tax Journal, 63(3), 397–418.
Johnson, W., & Krueger, R. F. (2005). Genetic effects on physical health: Lower at higher income levels. Behavior Genetics, 35(5), 579–590.
Johnston, D. C. (2005, December 19). Study shows the superrich are not the most generous. The New York Times. Retrieved February 26, 2016,
from http://www.nytimes.com/2005/12/19/us/study-shows-the-superrich-are-not-the-most-generous.html
Keltner, D., Kogan, A., Piff, P. K., & Saturn, S. R. (2014). The sociocultural appraisals, values, and emotions (SAVE) framework of
prosociality: Core processes from gene to meme. Annual Review of Psychology, 65, 425–460.
Keltner, D., Van Kleef, G. A., Chen, S., & Kraus, M. W. (2008). A reciprocal influence model of social power: Emerging principles and lines
of inquiry. Advances in Experimental Social Psychology, 40, 151–192.
Kirkpatrick, M., Delton, A. W., de Wit, H., & Robertson, T. E. (2015). Prosocial effects of MDMA: A measure of generosity. Journal of
Psychopharmacology, 29(6), 661–668.
Kohn, M. L. (1963). Social class and parent-child relationships: An interpretation. American Journal of Sociology, 68(4), 471–480.
Konigsberg, L., MacGregor, S., Johnson, M., Massey, A., & Daubman, K. (2013). Higher SES predicts unethical behavior in men. Retrieved
February 26, 2016, from http://www.psychfiledrawer.org/replication.php?attempt=MTUw
Korndörfer, M., Egloff, B., & Schmukle, S. C. (2015). A large scale test of the effect of social class on prosocial behavior. PLoS ONE, 10(7),
e0133193.
Kraus, M. W., Côté, S., & Keltner, D. (2010). Social class, contextualism, and empathic accuracy. Psychological Science, 21(11), 1716–1723.
Kraus, M. W., & Keltner, D. (2009). Signs of socioeconomic status: A thin-slicing approach. Psychological Science, 20(1), 99–106.
Kraus, M. W., Piff, P. K., & Keltner, D. (2009). Social class, sense of control, and social explanation. Journal of Personality and Social Psychology,
97(6), 992–1004.
Kraus, M. W., Piff, P. K., & Keltner, D. (2011). Social class as culture the convergence of resources and rank in the social realm. Current
Directions in Psychological Science, 20(4), 246–250.
Kraus, M. W., Piff, P. K., Mendoza-Denton, R., Rheinschmidt, M. L., & Keltner, D. (2012). Social class, solipsism, and contextualism: How
the rich are different from the poor. Psychological Review, 119(3), 546–572.
Lareau, A. (2002). Invisible inequality: Social class and childrearing in black families and white families. American Sociological Review, 67(5),
747–776.
Lieberman, M. D. (2010). Social cognitive neuroscience. In S. T. Fiske, D. T. Gilbert, & G. Lindzey (Eds.), Handbook of Social Psychology (5th
ed., pp. 143–193). New York: McGraw-Hill.
Lyons, J., Born, K., Pearsen, J., Fox, M., Collins, C., Trosch, E., & Daubman, K. (2012). Higher SES predicts unethical decision making.
Retrieved February 26, 2016, from http://www.psychfiledrawer.org/replication.php?attempt=MTMx
Marx, K. (1977). Capital: Critique of Political Economy (Vol. 1). New York: Vintage. (Original work published 1867).
Miller, J., Kahle, S., & Hastings, P. (2015). Roots and benefits of costly giving: Young children’s altruism is related to having less family wealth
and more autonomic flexibility. Psychological Science, 26(7), 1038–1045.
Mitchell, J. P. (2009). Inferences about mental states. Philosophical Transactions of the Royal Society B, 364(1521), 1309–1316.
MoeAndET. (2014, October 22). Asking Strangers for Food! (Social Experiment) [video file]. Retrieved February 26, 2016, from
https://www.youtube.com/watch?v=1fC6RP6tZ1Q
Muscatell, K. A., Morelli, S. A., Falk, E. B., Way, B. M., Pfeifer, J. H., Galinsky, A. D., … Eisenberger, N. I. (2012). Social status modulates
neural activity in the mentalizing network. NeuroImage, 60(3), 1771–1777.
Na, J., Grossmann, I., Varnum, M. E., Kitayama, S., Gonzalez, R., & Nisbett, R. E. (2010). Cultural differences are not always reducible to
individual differences. Proceedings of the National Academy of Sciences, 107(14), 6192–6197.
Oakes, J. M., & Rossi, P. H. (2003). The measurement of SES in health research: Current practice and steps toward a new approach. Social
Science & Medicine, 56(4), 769–784.
Oveis, C., Horberg, E. J., & Keltner, D. (2010). Compassion, pride, and social intuitions of self-other similarity. Journal of Personality and Social
Psychology, 98(4), 618–630.
Page-Gould, E., Mendoza-Denton, R., & Tropp, L. R. (2008). With a little help from my cross-group friend: Reducing anxiety in intergroup
contexts through cross-group friendship. Journal of Personality and Social Psychology, 95(5), 1080–1094.
Pew Research Center. (2012, August 27). Yes, the Rich Are Different. Retrieved February 26, 2016, from
http://www.pewsocialtrends.org/2012/08/27/yes-the-rich-are-different/
Piff, P. K., Dietze, P., Feinberg, M., Stancato, D. M., & Keltner, D. (2015). Awe, the small self, and prosocial behavior. Journal of Personality
and Social Psychology, 108(6), 883–899.
Piff, P. K., Kraus, M. W., Côté, S., Cheng, B. H., & Keltner, D. (2010). Having less, giving more: The influence of social class on prosocial

424
 behavior. Journal of Personality and Social Psychology, 99(5), 771–784.
Piff, P. K., Stancato, D. M., Côté, S., Mendoza-Denton, R., & Keltner, D. (2012). Higher social class predicts increased unethical behavior.
Proceedings of the National Academy of Sciences, 109(11), 4086–4091.
Piff, P. K., Stancato, D. M., & Horberg, E. J. (2016). Wealth and wrongdoing: Social class differences in ethical reasoning and behavior. In J.
W. Van Prooijen & P. A. M. van Lange (Eds.), Cheating, Corruption, and Concealment: The Roots of Dishonest Behavior. Cambridge, UK:
Cambridge University Press.
Piff, P. K., Stancato, D. M., Martinez, A. G., Kraus, M. W., & Keltner, D. (2012). Class, chaos, and the construction of community. Journal of
Personality and Social Psychology, 103(6), 949–962.
Piff, P. K. (2014). Wealth and the inflated self: Class, entitlement, and narcissism. Personality and Social Psychology Bulletin, 40(1), 34–43.
Plato. (1987). The Republic. New York: Penguin. (Original work published 380 b.c.e.)
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and
political attitudes. Journal of Personality and Social Psychology, 67(4), 741–763.
Saez, E., & Zucman, G. (2014). Wealth inequality in the United States since 1913: Evidence from capitalized income tax data. National Bureau
of Economic Research (NBER Working Paper No. 20625).
Schnall, S., Roper, J., & Fessler, D. M. (2010). Elevation leads to altruistic behavior. Psychological Science, 21(3), 315–320.
Schurr, A., & Ritov, I. (2016). Winning a competition predicts dishonest behavior. Proceedings of the National Academy of Sciences, 113(7), 1754–
1759.
Shiota, M. N., Keltner, D., & John, O. J. (2006). Positive emotion dispositions differentially associated with Big Five personality and
attachment style. Journal of Positive Psychology, 1(2), 61–71.
Snibbe, A. C., & Markus, H. R. (2005). You can’t always get what you want: Educational attainment, agency, and choice. Journal of Personality
and Social Psychology, 88(4), 703–720.
Stellar, J. E., Manzo, V. M., Kraus, M. W., & Keltner, D. (2012). Class and compassion: Socioeconomic factors predict responses to suffering.
Emotion, 12(3), 449–459.
Stephens, N. M., Fryberg, S. A., Markus, H. R., Johnson, C. S., & Covarrubias, R. (2012). Unseen disadvantage: How American universities’
focus on independence undermines the academic performance of first-generation college students. Journal of Personality and Social Psychology,
102(6), 1178–1197.
Stephens, N. M., Markus, H. R., & Townsend, S. M. (2007). Choice as an act of meaning: The case of social class. Journal of Personality and
Social Psychology, 93(5), 814–830.
Tarrant, M., Dazeley, S., & Cotton, T. (2009). Social categorization and empathy for outgroup members. British Journal of Social Psychology,
48(3), 427–446.
Taylor, S. E. (2006). Tend and befriend: Biobehavioral bases of affiliation under stress. Current Directions in Psychological Science, 15, 273–277.
Twenge, J. M., Campbell, W. K., & Freeman, E. C. (2012). Generational differences in young adults’ life goals, concern for others, and civic
orientation, 1966–2009. Journal of Personality and Social Psychology, 102(5), 1045–1062.
Uchida, Y., & Kitayama, S. (2001). Development and validation of a sympathy scale. Japanese Journal of Psychology, 72(4), 275–282.
Underwood, B., & Moore, B. (1982). Perspective-taking and altruism. Psychological Bulletin, 91(1), 143–173.
van Kleef, G. A., Oveis, C., van der Löwe, I., LuoKogan, A., Goetz, J., & Keltner, D. (2008). Power, distress, and compassion: Turning a blind
eye to the suffering of others. Psychological Science, 19(12), 1315–1322.
Varnum, M. E. W., Blais, C., Hampton, R. S., & Brewer, G. A. (2015). Social class affects neural empathic responses. Culture and Brain, 3(2),
122–130.
Wang, L., & Murnighan, J. K. (2014). Money, emotions, and ethics across individuals and countries. Journal of Business Ethics, 125(1), 163–
176.
Watson, P. J., Grisham, S. O., Trotter, M. V., & Biderman, M. D. (1984). Narcissism and empathy: Validity evidence for the Narcissistic
Personality Inventory. Journal of Personality Assessment, 48(3), 301–305.
Zaki, J., & Ochsner, K. N. (2012). The neuroscience of empathy: Progress, pitfalls and promise. Nature Neuroscience, 15(5), 675–680.
Zaki, J., Weber, J., Bolger, N., & Ochsner, K. (2009). The neural bases of empathic accuracy. Proceedings of the National Academy of Sciences,
106(27), 11382–11387.

425