118  Chapter 4   Self and identity

Who are you?

Take a look in your wallet. You will find cards that have your name on them, and probably a
rather gruesome photograph of yourself. What happens when you meet someone? Very early
on you discover each other’s name, and soon after that you establish such things as their occu-
pation, their attitudes and what they like to do. You also try to identify mutual acquaintances.
In more formal contexts, people sometimes display their identity by donning a uniform, whip-
ping out a flashy business card or wearing one of those often embarrassing name/role badges.
In the brave new world of the Internet, people can of course construct and nurture, courtesy
of Facebook and other social media, limitless more or less truthful selves and identities.
Your identity and your self-concept underpin your everyday life. Knowing who you are
allows you to know what you should think and do and how others might think of and treat
you; and knowing who others are allows you to predict what they think and what they do.
Knowing our identity regulates and structures how we interact with others, and in turn,
identities are grounded in social interaction and the structure of society.
Many scholars believe that it is reflexive thought – that is, the ability to think about our-
selves thinking – that separates us from almost all other animals. Reflexive thought means
that we can think about ourselves, about who we are, how we would like to be and how we
would like others to see us. Humans have a highly developed sense of self, and self and iden-
tity are fundamental parts of being human. We should not be surprised that social psycholo-
gists in particular have become intrigued with the self.
In this chapter, we explore the self – where it comes from, what it looks like and how it
Constructs influences thought and behaviour. Because self and identity are cognitive constructs that influ-
Abstract or theoretical ence social interaction and perception, and that are themselves influenced by society, the mate-
concepts or variables that
are not observable and are
rial in this chapter connects to virtually all other chapters in the text. The self is an enormously
used to explain or clarify a popular focus of research (e.g. Leary & Tangney, 2003; Sedikides & Brewer, 2001; Swann &
phenomenon. Bosson, 2010). Ashmore and Jussim (1997) reported 31,000 social psychological publications
on the self over a two-decade period to the mid-1990s. There is now an International Society
for Self and Identity and a scholarly journal imaginatively entitled Self and Identity.

Self and identity in historical context

The self is, historically, a relatively new idea (Baumeister, 1987). In medieval society, social
relations were fixed and stable and legitimised in religious terms. People’s lives and identities
were mapped out according to their position in the social order – by ascribed attributes such
as family membership, social rank, birth order and place of birth. In many ways, what you
saw was what you got, so the idea of a complex individual self lurking underneath it all was
superfluous and difficult to imagine.
All this started to change in the sixteenth century, and the change has gathered momen-
tum ever since. The forces for change included:
● Secularisation – the idea that fulfilment occurs in the afterlife was replaced by the idea
that you should actively pursue personal fulfilment in this life.
● Industrialisation – people were increasingly seen as units of production that moved from
place to place to work and thus had a portable personal identity that was not locked into
static social structures such as the extended family.
● Enlightenment – people felt that they could organise and construct different, better, iden-
tities and lives for themselves by overthrowing orthodox value systems and oppressive
regimes (e.g. the French and American revolutions of the late eighteenth century).
● Psychoanalysis – Freud’s theory of the human mind crystallised the notion that the self
was unfathomable because it lurked in the gloomy depths of the unconscious (see the
‘Psychodynamic self’ section).
 Self and identity in historical context   119

Psychoanalysis challenged the way we think about self and identity: it attributes behav-
iour to complex dynamics that are hidden deep within the person’s sense of who they are.
Earlier in the text (see Chapter 3; also see Chapter 5), we explored the theory of social rep-
resentations – a theory that invoked psychoanalysis as an example of how a novel idea or
analysis can entirely change the way that people think about their world (e.g. Moscovici,
1961; see Lorenzi-Cioldi & Clémence, 2001).
Together, these and other social, political and cultural changes caused people to think
about self and identity as complex and problematic. Theories of self and identity propa-
gated and flourished in this fertile soil.

Psychodynamic self
Freud (e.g. 1921) believed that unsocialised and selfish libidinal impulses (the id) are repressed
and kept in check by internalised societal norms (the superego), but that, from time to time
and in strange and peculiar ways, repressed impulses surface. Freud’s view of the self is one
in which you can only truly know yourself, or indeed others, when special procedures, such
as hypnosis or psychotherapy, are employed to reveal repressed thoughts. His ideas about
self, identity and personality are far-reaching in social psychology: for example, Adorno,
Frenkel-Brunswik, Levinson and Sanford’s (1950) influential authoritarian personality
­theory of prejudice is a psychodynamic theory (see Chapter 10).

Individual versus collective self

Freud, like many other psychologists, viewed the self as very personal and private – the high
point of individuality: something that uniquely describes an individual human being. When
someone says ‘I am . . .’ they are describing what makes them different from all other human
beings. But think about this for a moment. ‘I am British’, ‘I come from Bristol’, ‘I am a social
psychologist’ – these are all descriptions of myself, but they are also descriptions of many
other people’s selves (there are 64 million Britons, over 440,000 people currently living in
Bristol, and many thousands of social psychologists). So the self can also be a shared or col-
lective self – a ‘we’ or ‘us’.
Social psychologists have argued long and hard for more than a century over what to
make of this – is the self an individual or a collective phenomenon? The debate has created
polarised camps, with advocates of the individual self and advocates of the collective self
slogging it out in the literature. It is fair to say that the advocates of the individual self have
tended to prevail. This is largely because social psychologists have considered groups to be
made up of individuals who interact with one another rather than of individuals who have a
collective sense of shared identity. Individuals interacting in aggregates is the focus of social
psychology as a behavioural science, whereas groups as collectives is the focus of social sci-
ences, such as sociology and political science (see Chapters 1 and 11).
This perspective on groups, summed up by Floyd Allport’s legendary proclamation that
‘There is no psychology of groups which is not essentially and entirely a psychology of individu-
als’ (Allport, 1924, p. 4), has made it difficult for the collective self to thrive as a research topic.

Collective self
It was not always like this. In the early days of social psychology, things were very different
(see Farr, 1996; Hogg & Williams, 2000). Wilhelm Wundt was the founder of psychology as
an experimental science, and he proposed that social psychology was the study of:
those mental products which are created by a community of human life and are, therefore,
inexplicable in terms merely of individual consciousness since they presuppose the recipro-
cal action of many.
Wundt (1916, p. 3)
120  Chapter 4   Self and identity

Wundt’s social psychology dealt with collective phenomena, such as language, religion,
customs and myth, which, according to Wundt, could not be understood in terms of the
psychology of the isolated individual. Emile Durkheim (1898), one of the founding fathers
of sociology, was influenced by Wundt’s interest in collective life and also maintained that
collective phenomena could not be explained in terms of individual psychology.
The view that the self draws its properties from groups is shared by many other early
social psychologists: for example, early theorists of collective behaviour and the crowd (e.g.
LeBon, 1908; Tarde, 1901; Trotter, 1919; see also Chapter 11). Notably, William McDougall,
in his book The Group Mind (McDougall, 1920), argued that out of the interaction of indi-
viduals there arose a ‘group mind’, which had a reality and existence that was qualitatively
distinct from the isolated individuals making up the group. There was a collective self that
was grounded in group life. Although phrased in rather quaint old-fashioned language, this
idea has a direct line of descent to subsequent experimental social psychology which con-
firms that human interaction has emergent properties that endure and influence other peo-
ple: for example, Muzafer Sherif’s (1936) research on how norms emerge from interaction
and are internalised to influence behaviour, and some of Solomon Asch’s (1952) research on
conformity to norms.
Since the early 1980s there has been a revival of interest in the notion of a collective self,
largely initiated by European research on the emergence of social representations out of
social interaction (e.g. Farr & Moscovici, 1984; Lorenzi-Cioldi & Clémence, 2001; see
Chapters 3, 5, 7 and 8), and on the role of social identity in group processes and intergroup
behaviour (e.g. Tajfel & Turner, 1986; also see Hogg, 2006; Hogg & Abrams, 1988; dis-
cussed later in this chapter but covered fully in Chapter 11).

Symbolic interactionist self

Another twist to the idea of the collective self is recognition that the self emerges and is
shaped by social interaction. Early psychologists such as William James (1890) distinguished
between self as stream of consciousness, ‘I’, and self as object of perception, ‘me’. In this
way, reflexive knowledge is possible because ‘I’ can be aware of ‘me’, and people can there-
fore know themselves. However, this is not to say that people’s self-knowledge is particularly
accurate. People tend to reconstruct who they are without being aware of having done it
(Greenwald, 1980), and in general, although people may be aware of who they are in terms
of their attitudes and preferences, they are rather bad at knowing how they arrived at that
knowledge (Nisbett & Wilson, 1977).
Symbolic interactionism Nevertheless, people do have a sense of ‘me’, and according to symbolic interactionism,
Theory of how the self the self arises out of human interaction (Mead, 1934; see also Blumer, 1969). G. H. Mead
emerges from human
interaction, which involves
believed that human interaction is largely symbolic. When we interact with people, it is
people trading symbols mainly in terms of words and non-verbal cues that are rich with meaning because they
(through language and symbolise much more than is superficially available in the behaviour itself (see Chapter 15).
gesture) that are usually Mead believed that society influences individuals through the way individuals think about
consensual and represent
abstract properties rather
themselves, and that self-conception arises and is continually modified through interaction
than concrete objects. between people. This interaction involves symbols that must have shared meaning if they
are to be communicated effectively. If you say to your friend ‘let’s eat out tonight’, you
both know what this means and that it opens up a variety of choices that each of you know
about.
Interacting effectively also rests on being able to take the role of the other person. This of
course entails ‘looking in from outside’ and seeing oneself as others do – as a social object,
‘me’, rather than a social subject, ‘I’ (cf. Batson, Early, & Salvarani, 1997). Because others
often view us as representatives of a category (e.g. a student), the ‘me’ is probably more
often seen as a collective ‘me’ – we might even think of it as ‘us’. The representations, or
views, that our society has of the world are traded through interacting symbolically with
 Self and identity in historical context   121

The looking glass self

According to G.H. Mead,
our self-concept derives
from seeing ourselves as
others see us. Is the
broken mirror a
metaphor for her
self-concept?

others. We are effective only if we can take the role of the other and thus see ourselves as
others (ultimately, society) do. In this way, we construct a self-concept that reflects the soci-
ety we live in; we are socially constituted.
Symbolic interactionism offers a quite sophisticated and complex model of how the self
is formed. And yet it generates a very straightforward prediction. Because forming our con-
cept of self comes from seeing ourselves as others see us (the idea of the looking-glass self), Looking-glass self
how we view ourselves should be closely shadowed by how others view us. Shrauger and The self derived from seeing
ourselves as others see us.
Schoeneman (1979) reviewed sixty-two studies to see if this was true. What they found was
that people did not tend to see themselves as others saw them but instead saw themselves as
they thought others saw them. For a more recent example of research on the looking-glass
self, see Box 4.1 and Figure 4.1.
One implication of the idea that people do not see themselves as others see them, but
instead see themselves as they think others see them, is that we do not actually take the role
of the other in constructing a sense of self. An alternative reading is that the communication
process in social interaction is noisy and inaccurate. It is influenced by a range of

Box 4.1 Research highlight

Public versus private self-presentation

Dianne Tice (1992) conducted an experiment where behaviour. (This was a ruse, since there was no one actu-
undergraduate students were asked to act as ‘stimulus per- ally monitoring the students.) In the next phase, they were
sons’ for postgraduate clinical psychology trainees. Their asked to rate themselves in terms of how responsive they
task was to use an intercom system to answer verbal ques- really were. They made their ratings on a 25-point scale
tions in a way that would reflect an aspect of their person- ranging from 1 (stable = not responsive) to 25
ality. Effectively, they were to describe themselves so that (responsive).
they would come across as either consistently emotionally Tice intended the public condition to be the one that
stable (implying not responsive) or emotionally responsive would engage the looking-glass self. As predicted, subse-
in different situations. quent descriptions of self were more radically altered
There were two experimental conditions: (a) a private under public conditions than private conditions (see
condition where the students believed no one was watch- Figure 4.1) – suggesting that the students did not see
ing them, and (b) a public condition where they believed a themselves as others saw them, but instead as they thought
clinical psychology trainee was closely monitoring their others saw them.
122  Chapter 4   Self and identity

20 Private
Public

Self-conception as emotionally responsive

15

Figure 4.1 Conceiving of oneself as emotionally

stable or emotionally responsive, as a function
of public or private self-presentation
● People were instructed to present themselves as 10
either less emotionally responsive (i.e. more stable)
or more emotionally responsive.
● Next, they rated their ‘true’ level of emotional
responsiveness on a 25-point scale, ranging from 5
a low score (less emotionally responsive) to a high
score (more emotionally responsive).
● When they believed that their earlier behaviour
had been public, their self-conception moved
in the direction of their action: closer to a score 0
Emotionally stable Emotionally responsive
of 1 for those who had been less emotionally
responsive, or closer to a score of 25 for those who Self-presentation
had been more emotionally responsive.
Source: Based on data from Tice (1992), Study 1.

self-construal motivations (motives to view others, and be viewed by them, in particular

ways) that conspire to construct an inaccurate image of others and what they think about
us. People are mostly unaware of what other people really think of them (Kenny & DePaulo,
1993), perhaps fortunately so. A sage person once said, ‘if you really want to hear how much
people like you, you’d better listen to what they say at your funeral!’
As we discover in this chapter, our concept of self is linked to how we go about enhancing
our self-image. People normally overestimate their good points, overestimate their control
over events and are unrealistically optimistic – Sedikides and Gregg (2007) call this the self-
enhancing triad.

Self-awareness
If the truth be known, you do not spend all your time thinking about yourself. Self-awareness
comes and goes for different reasons and has an array of consequences. Although I am sure
you can all think of people who appear to think only of themselves almost all the time (we
discuss narcissism later in this chapter)!
In their book A Theory of Objective Self-Awareness, Shelley Duval and Robert Wicklund
(1972) argued that self-awareness is a state in which you are aware of yourself as an object,
much as you might be aware of a tree or another person. When you are objectively self-aware,
you make comparisons between how you actually are and how you would like to be – an ideal,
a goal or some other standard. The outcome of this comparison is often a sense that you have
shortcomings, along with negative emotions associated with this recognition. People then try to
overcome their shortcomings by bringing the self closer into line with ideal standards. This can
be very difficult, leading people to give up trying and thus feel even worse about themselves.
Objective self-awareness is generated by anything that focuses your attention on yourself
as an object: for example, being in front of an audience (see Chapter 6) or catching your
image in a mirror. Indeed, a very popular method for raising self-awareness in laboratory
studies is actually to place participants in front of a mirror. Charles Carver and Michael
 Self-knowledge  123

Scheier (1981) introduced a qualification to self-awareness theory, in which they distin-

guished between two types of self that you can be aware of:
1 the private self – your private thoughts, feelings and attitudes;
2 the public self – how other people see you, your public image.
Private self-awareness leads you to try to match your behaviour to your internalised stand-
ards, whereas public self-awareness is oriented towards presenting yourself to others in a
positive light.
Being self-aware can be very uncomfortable. We all feel self-conscious from time to time
and are only too familiar with how it affects our behaviour – we feel anxious, we become
tongue-tied, or we make mistakes on tasks. We can even feel slightly paranoid (Fenigstein,
1984). However, sometimes being self-aware can be a terrific thing, particularly on those
occasions when we have accomplished a great feat. In early December 2003, having won the
Rugby World Cup, the England team paraded through London and ended up in Trafalgar
Square in front of three-quarters of a million people – standing in an open-topped bus, the
team looked freezing but certainly did not suffer from the crowd’s adulation.
Self-awareness can also make us feel good when the standards against which we compare
ourselves are not too exacting: for example, if we compare ourselves against standards
derived from ‘most other people’ or from people who are less fortunate than ourselves
(Taylor & Brown, 1988; Wills, 1981). Self-awareness can also improve introspection, inten-
sify emotions and improve performance of controlled effort-sensitive tasks that do not
require undue skill, such as checking over an essay you have written.
The opposite of being objectively self-aware is being in a state of reduced objective self-
awareness. Because elevated self-awareness can be stressful or aversive, people may try to
avoid this state by drinking alcohol, or by more extreme measures such as suicide (Baumeister,
1991). Reduced self-awareness has also been identified as a key component of deindividuation, Deindividuation
a state in which people are blocked from awareness of themselves as distinct individuals, fail to Process whereby people
lose their sense of socialised
monitor their actions and can behave impulsively. Reduced self-awareness may be implicated individual identity and
in the way that crowds behave and in other forms of social unrest. Read how this comes engage in unsocialised,
about in both small groups and crowd settings in Chapters 11 and 12. often antisocial, behaviours.
Self-awareness is about being aware of self. However, recent research suggests that the same
effect of trying to match one’s behaviour to standards can be obtained by unconsciously focus-
ing attention on self. Silvia and Phillips (2013) report two studies where self-awareness was
primed not by usual explicit mirror method but by subliminal first-name priming. The effects
were the same; suggesting that self-awareness can be primed subliminally and is therefore not
a deliberative awareness phenomenon but an automatic attention phenomenon. Silvia and
Phillips suggest perhaps talking more about self-focused attention than self-awareness.

Self-knowledge
When people are self-aware, what are they aware of? What do we know about ourselves, and
how do we gain a sense of who we are? Self-knowledge is constructed in much the same way
and through many of the same processes as we construct representations of other people.
We looked at some of these general processes when we discussed social thinking and attri-
bution in Chapters 2 and 3.
Schema
Self-schemas Cognitive structure that
represents knowledge about
Earlier (see Chapter 2) we saw how information about other people is stored in the form a concept or type of
stimulus, including its
of a schema. We cognitively store information about the self in a similar but more com- attributes and the relations
plex and varied way – as separate context-specific nodes where different contexts activate among those attributes.
124  Chapter 4   Self and identity

An ideal self
‘Wow! So how do
I look?’

different nodes and thus, effectively, different aspects of self (Breckler, Pratkanis, &
McCann, 1991; Higgins, Van Hook, & Dorfman, 1988). You are probably now itching to
ask, ‘So . . . where in the brain is the self?’ Well, research suggests that no single brain sys-
tem or area of the brain is, of itself, responsible for one’s sense of self. Instead, the experi-
ence of self emerges from widely distributed brain activity across the medial prefrontal
and medial precuneus cortex of the brain (e.g. Saxe, Moran, Scholz, & Gabrieli, 2006).
The self-concept is neither a singular, static, lump-like entity nor a simple averaged view
of the self – it is complex and multi-faceted, with a large number of discrete self-schemas
(Markus, 1977; Markus & Wurf, 1987). People tend to have clear conceptions of themselves
(i.e. self-schemas) on some dimensions but not others – i.e. they are schematic on some but
aschematic on others. People are self-schematic on dimensions that are important to them,
on which they think they are extreme and on which they are certain the opposite does not
hold. For example, if you think you are sophisticated, and being sophisticated is important
to you, then you are self-schematic on that dimension – it is part of your self-concept. If you
do not think you are sophisticated, and if this does not bother you, then being sophisticated
is not one of your self-schemas.
Most people have a complex self-concept with a relatively large number of discrete self-
schemas. Patricia Linville (1985, 1987; see ‘Many selves, multiple identities’ in this chapter)
has suggested that this variety helps to buffer people from the negative impact of life events by
making sure that there are always self-schemas from which they can derive a sense of satisfac-
tion. People can be quite strategic in how they use their self-schemas – Linville used a colour-
ful phrase to describe what we usually do: ‘don’t put all your eggs in one cognitive basket’.
Self-schemas that are rigidly compartmentalised have disadvantages (Showers, 1992). If
some self-schemas are very negative and some are very positive, events may cause extreme
mood swings according to whether a positive or negative self-schema is primed. Generally,
more integrated self-schemas are preferable. For example, if James believes that he is a won-
derful cook but an awful musician, he has compartmentalised self-schemas – contexts that
prime one or the other self-schema will produce very positive or very negative moods. Contrast
this with Sally, who believes she is a reasonably good cook but not a great musician. She has
self-schemas where the boundaries are less clear – context effects on mood will be less extreme.
 Self-knowledge  125

Self-schemas influence information processing and behaviour in much the same way as
schemas about other people (Markus & Sentis, 1982): self-schematic information is more
readily noticed, is overrepresented in cognition and is associated with longer processing
time. Self-schemas do not only describe how we are. Markus and Nurius (1986) have sug-
gested that we have an array of possible selves – future-oriented schemas of what we would
like to become, or what we fear we might become. For example, a postgraduate student may
have future selves as a university lecturer or a rock musician.
Another perspective is offered by Higgins’s (1987) self-discrepancy theory. Higgins sug- Self-discrepancy theory
gests that we have three types of self-schema: Higgins’s theory about the
consequences of making
1 actual self – how we currently are; actual – ideal and actual –
‘ought’ self-comparisons
2 ideal self – how we would like to be; that reveal
3 ‘ought’ self – how we think we should be. self-discrepancies.

The ideal self and the ought self are ‘self-guides’, but they mobilise different types of self-
related behaviours. The same goal – for example, prosperity – can be constructed as an ideal
(we strive to be prosperous) or an ‘ought’ (we strive to avoid not being prosperous).
Discrepancies between actual and ideal or ‘ought’ motivate change to reduce the discrep-
ancy – in this way we engage in self-regulation. (In Chapter 13 we discuss self-regulation in Self-regulation
the context of close relationships.) Furthermore, these self-discrepancies make us emotion- Strategies that we use to
match our behaviour to an
ally vulnerable. When we fail to resolve an actual–ideal discrepancy, we feel dejected (e.g. ideal or ‘ought’ standard.
disappointed, dissatisfied, sad); when we fail to resolve an actual–ought discrepancy, we feel
agitated (e.g. anxiety, threat, fear). Read how Higgins and his colleagues tested self-discrep-
ancy theory in Box 4.2 and Figure 4.2.

Regulatory focus theory

Self-discrepancy theory and the general notion of self-regulation have been elaborated into
regulatory focus theory (Higgins, 1997, 1998). Higgins proposes that people have two sepa- Regulatory focus theory
rate self-regulatory systems, termed promotion and prevention, which are concerned with A promotion focus causes
people to be approach-
the pursuit of different types of goals. oriented in constructing a
sense of self; a prevention
● The promotion system is concerned with the attainment of one’s hopes and aspirations
focus causes people to be
– one’s ideals. It generates sensitivity to the presence or absence of positive events. People more cautious and avoidant
in a promotion focus adopt approach strategic means to attain their goals. For example, in constructing a sense of self.
promotion-focused students would seek ways to improve their grades, find new challenges
and treat problems as interesting obstacles to overcome.
● The prevention system is concerned with the fulfilment of one’s duties and obligations –
one’s oughts. It generates sensitivity to the presence or absence of negative events. People
in a prevention focus use avoidance strategic means to attain their goals. For example,
prevention-focused students would avoid new situations or new people and concentrate
more on avoiding failure than on achieving the highest possible grade.
Some people are habitually more approach-focused and others more prevention-focused – it
is an individual difference that can arise during childhood (Higgins & Silberman, 1998). A pro-
motion focus can arise if children are habitually hugged and kissed for behaving in a desired
manner (a positive event) and love is withdrawn as a form of discipline (absence of a positive
event). A prevention focus can arise if children are encouraged to be alert to potential dangers
(absence of a negative event) and punished and shouted at when they behave undesirably (a
negative event). Against the background of individual differences, regulatory focus can also be
influenced by the immediate context, for example by structuring the situation so that people
focus on prevention or on promotion (Higgins, Roney, Crowe, & Hymes, 1994).
Research shows that people who are promotion-focused are especially likely to recall
information relating to the pursuit of success by others (Higgins & Tykocinski, 1992).
126  Chapter 4   Self and identity

Box 4.2 Research classic

Self-discrepancy theory: the impact of using self-guides

Tory Higgins and his colleagues measured self-discrepancy after a priming procedure. For their ‘ideal’ prime they were
by comparing the differences between attributes of the asked to discuss their own and their parents’ hopes for
actual self with those of either the ideal self or those of the them; for their ‘ought’ prime they discussed their own and
‘ought’ self (Higgins, Bond, Klein, & Strauman, 1986). their parents’ beliefs about their duties and obligations.
They administered questionnaires to identify students It was hypothesised that an actual–ideal discrepancy
who were either high in both kinds of discrepancies or else would lead to feeling dejected (but not agitated), whereas
low in both. Several weeks later, the same students partici- an actual–’ought’ discrepancy would lead to feeling agi-
pated in an experiment in which emotions that reflected tated (but not dejected). These predictions were sup-
dejection or agitation were measured, both before and ported, as the results in Figure 4.2 show.

6 Dejection
Agitation

4
Change in emotion

Figure 4.2 Priming the ideal self can lead to 2

dejection, whereas priming the ‘ought’ self
can lead to agitation
People with a high actual–ideal and actual–ought
self-discrepancy experienced: 0
● an increase in dejection but not agitation
emotions after being primed to focus on their
ideal self, and
● an increase in agitation but not dejection –2
Ideal prime Ought prime
emotions after being primed to focus on their
Type of self-prime
‘ought’ self.
Source: Based on Higgins, Bond, Klein and Strauman (1986),
Experiment 2.

Lockwood and her associates found that people who are promotion-focused look for inspi-
ration to positive role models who emphasise strategies for achieving success (Lockwood,
Jordan, & Kunda, 2002). Such people also show elevated motivation and persistence on
tasks that are framed in terms of gains and non-gains (Shah, Higgins, & Friedman, 1998).
People who are prevention-focused behave quite differently – they recall information relating
to the avoidance of failure by others, are most inspired by negative role models who high-
light strategies for avoiding failure, and exhibit motivation and persistence on tasks that are
framed in terms of losses and non-losses.
Regulatory focus theory has also been explored in the context of intergroup relations and
how people feel about and behave towards their ingroup and relevant outgroups (e.g. Jonas,
Sassenberg, & Scheepers, 2010; see Chapter 11). For example, studies have shown that in
intergroup contexts, a measured or manipulated promotion focus strengthens positive emo-
tion-related bias and behavioural tendencies towards the ingroup, while a prevention focus
strengthens more negative emotion-related bias and behavioural tendencies against the out-
group (Shah, Brazy, & Higgins, 2004).
 Self-knowledge  127

Inferences from our behaviour

One of the most obvious ways to learn about who you are is to examine your private thoughts
and feelings about the world – knowing what you think and feel about the world is a very
useful clue to the sort of person you are.
However, when these internal cues are weak, we may make inferences about ourselves
from what we do – our behaviour. This idea underpins Daryl Bem’s self-perception theory Self-perception theory
(Bem, 1967, 1972). Bem argues that we make attributions not only for others’ behaviour Bem’s idea that we gain
knowledge of ourselves only
(see Chapter 3) but also for our own, and that there is no essential difference between by making self-attributions:
self-attributions and other-attributions. Furthermore, just as we form an impression of for example, we infer our
someone else’s personality by making internal dispositional attributions for their behav- own attitudes from our own
iour, so we form a concept of who we are not by introspection but by being able to attrib- behaviour.

ute our own behaviour internally. So, for example, I know that I enjoy eating curry
because, if given the opportunity, I eat curry of my own free will and in preference to
other foods, and not everyone likes curry – I am able to make an internal attribution for
my behaviour.
How we perceive ourselves can also be based on simply imagining ourselves behaving in a
particular way (Anderson & Godfrey, 1987). For example, sports psychologist Caroline van
Gyn and her colleagues divided runners into two groups; one group practised power training
on exercise bikes, the other did not. Half of each group used imagery (i.e. also imagined
themselves sprint training), whereas the others did not. Of course, the sweaty business of
power training itself improved subsequent performance; but, remarkably, those who imag-
ined themselves sprint training did better than those who did not. The researchers concluded
that imagery had affected self-conception, which in turn produced performance that was
consistent with that self-conception (Van Gyn, Wenger, & Gaul, 1990).
Self-attributions have implications for motivation. If someone is induced to perform a
task by either enormous rewards or fearsome penalties, task performance is attributed exter-
nally and thus motivation to perform is reduced. If there are minimal or no external factors
to which performance can be attributed, we cannot easily avoid attributing performance
internally to enjoyment or commitment, so motivation increases. This has been called the
overjustification effect (see Figure 4.3), for which there is now substantial evidence (Deci & Overjustification effect
Ryan, 1985). In the absence of obvious
external determinants of our
For example, Mark Lepper and his colleagues had nursery-school children draw pictures.
behaviour, we assume that
Some of the children drew of their own free will, while the rest were induced to draw with we freely choose the
the promise of a reward, which they were subsequently given. A few days later, the children behaviour because we
were unobtrusively observed playing; the children who had previously been rewarded for enjoy it.
drawing spent half as much time drawing as did the other group. Those who had received no
extrinsic reward seemed to have greater intrinsic interest in drawing (Lepper, Greene, &
Nisbett, 1973).
A review by John Condry (1977) concludes that introducing external rewards may back-
fire by reducing motivation and enjoyment of a task that was previously intrinsically moti-
vated. The educational implications are obvious. Parents love to tell their children stories,
and they encourage the young ones to enjoy stories by learning to read themselves. However,
if reading is accompanied by rewards, the children’s intrinsic joy is put at risk. So, is it pos-
sible for rewards to play any useful role? The answer is yes. The trick is to reduce reliance on
rewards that are task-contingent and make more use of those that are performance-­
contingent. Even a task that people find boring can be enlivened when they shift their atten-
tion to features of their performance (Sansone, Weir, Harpster, & Morgan, 1992). Consider
how you look for ways to maintain interest in a monotonous physical fitness programme,
especially when you have to exercise alone. You could, of course, listen to music or watch
television. However, a performance-contingent strategy is to set targets using measures such
as ‘distance’ covered on an exercycle, checking your heart rate and how many calories you
expended.
128  Chapter 4   Self and identity

Conditions of task performance Attribution/reason Example Consequences

for performing task

• Intrinsic interest • Enjoy the task

salient • Greater efficiency
• Rewards signify Internal/ Working • Confront greater
competence intrinsic because challenges
• Rewards are contingent interest one enjoys it • Better performance
on good performance • Self-motivated
performance

• Rewards are salient • Work hard

• Rewards do not signify • Less task enjoyment
competence External/ Working • Externally motivated
• Rewards are not extrinsic because it performance
contingent on good rewards pays well • Avoid challenges
performance • Less efficient
• Rewards constitute • Poorer performance
efforts at control

Figure 4.3 The overjustification effect

One’s motivation to perform a task can be reduced, and performance of the task impaired, if there are obvious external causes for task
performance – an overjustification effect that is reversed if performance can be internally attributed.

Social comparison and self-knowledge

Are you intelligent? How do you know? Although we can learn about ourselves through
introspection and self-perception, we can also learn about ourselves by comparing ourselves
Social comparison with other people. This simple truth lies at the core of Festinger’s (1954) social comparison
(theory) theory, which describes how people learn about themselves through comparisons with oth-
Comparing our behaviours
and opinions with those of
ers (see also Suls & Wheeler, 2000; Wheeler, 1991). People need to be confident about the
others in order to establish validity of their perceptions, attitudes, feelings and behaviour, and because there is rarely an
the correct or socially objective measure of validity, people ground their cognitions, feelings and behaviour in those
approved way of thinking of other people. In particular, they seek out similar others to validate their perceptions and
and behaving.
attitudes, which can, to some extent, be read as meaning that people anchor their attitudes
and self-concept in the groups to which they feel they belong.
When it comes to performance, we try to compare ourselves with people who are slightly
worse than us – we make downward social comparisons which deliver an evaluatively posi-
tive self-concept (Wills, 1981). Often, however, our choices are limited: for example, younger
siblings in families often have no option but to compare themselves with their more compe-
tent older brothers and sisters. Indeed, upward comparison may sometimes have a harmful
effect on self-esteem (Wood, 1989).
Self-evaluation How can we avoid this? According to Abraham Tesser’s (1988) self-evaluation mainte-
maintenance model nance model, we try to downplay our similarity to the other person or withdraw from our
People who are constrained
to make esteem-damaging
relationship with that person. Medvec and her colleagues conducted an intriguing study
upward comparisons can along these lines (Medvec, Madley, & Gilovich, 1995). They coded the facial expressions of
underplay or deny similarity medal winners at the 1992 Olympic Games in Barcelona and found that the bronze medal-
to the target, or they can lists expressed noticeably more satisfaction than the silver medallists! Medvec and colleagues
withdraw from their
relationship with the target.
argued that silver medallists were constrained to make unfavourable upward comparisons
with gold medallists, whereas bronze medallists could make self-enhancing downward com-
parisons with the rest of the field, who received no medal at all.
 Many selves, multiple identities   129

Downward comparisons also occur between groups. Groups try to compare themselves
with inferior groups in order to feel that ‘we’ are better than ‘them’. Intergroup relations are
largely a social comparison-based struggle for evaluative superiority of one’s own group
over relevant outgroups (see Hogg, 2000c; Hogg & Gaffney, 2014; Turner, 1975). Because we
tend to describe and evaluate ourselves in terms of groups we belong to, this process
enhances self-evaluation and self-esteem (Tajfel & Turner, 1986 – see Chapter 11).
Sport provides a perfect context in which the outcome of this process can be seen. Few
Portuguese will not have felt enormously positive when their team beat France, the host
nation, in the finals of the 2016 UEFA European Championship. Bob Cialdini and his col-
leagues have referred to this phenomenon as ‘basking in reflected glory’, or BIRGing (Cialdini BIRGing
et al., 1976). To illustrate the effect, they conducted experiments in which they raised or Basking in Reflected Glory;
that is, name-dropping to
lowered self-esteem via feedback on a general knowledge test; and student participants were
link yourself with desirable
then, seemingly incidentally, asked about the outcome of a recent football game. Participants people or groups and thus
who had had their self-esteem lowered tended to associate themselves with winning and not improve other people’s
with losing teams – they tended to refer to the teams as ‘we’ in the former case and as ‘they’ impression of you.
in the latter.

Many selves, multiple identities

It is probably inaccurate to characterise the self as a single undifferentiated entity. In his
book The Concept of Self, Kenneth Gergen (1971) depicts the self-concept as containing a
repertoire of relatively discrete and often quite varied identities, each associated with a dis-
tinct body of knowledge. These identities have their origins in the array of different social
relationships that form, or have formed, the anchoring points for our lives, ranging from
close personal relationships with friends and family, through relationships and roles defined
by work groups and professions, to relationships defined by ethnicity, race, nationality and
religion.
As we noted earlier, we differ in self-complexity (Linville, 1985). Some of us have a more
diverse and extensive set of selves than do others – people with many independent aspects of
self have higher self-complexity than people with only a few, relatively similar, aspects of
self. The notion of self-complexity is given a slightly different emphasis by Marilynn Brewer
and her colleagues (Brewer & Pierce, 2005; Roccas & Brewer, 2002), who focus on self that
is defined in group terms (social identity) and the relationship among identities rather than
the number of identities people have. People have a complex social identity if they have dis-
crete social identities that do not share many attributes, and a simple social identity if they
have overlapping social identities that share many attributes.
Grant and Hogg (2012) have recently suggested and shown empirically that the effect,
particularly on group identification and group behaviours, of the number of identities one
has and their overlap may be better explained in terms of the general property of social iden-
tity prominence – how subjectively prominent, overall and in a specific situation, a particu-
lar identity is in one’s self-concept.

Social identity
Types of self and identity That part of the self-concept
that derives from our
membership in social
Social identity theorists (Tajfel & Turner, 1986) have argued that there are two broad classes
groups.
of identity that define different types of self:
Personal identity
1 social identity, which defines self in terms of group memberships; The self defined in terms of
unique personal attributes
2 personal identity, which defines self in terms of idiosyncratic traits and close personal or unique interpersonal
relationships. relationships.
130  Chapter 4   Self and identity

Table 4.1 Self and self-attributes as a function of level of identity (social versus personal) and type
of attributes (identity versus relationship)
Identity attributes Relationship attributes
Social identity Collective self Collective relational self
Attributes shared with others that Attributes that define how the self as
differentiate the individual from a an ingroup member relates to specific
specific outgroup, or from others as ingroup or outgroup
outgroups in general. members.
Personal identity Individual self Individual relational self
Attributes unique to self that Attributes that define how the self as a
differentiate the individual from unique individual relates to others as
specific individuals, or from other individuals.
individuals in general.

Now check the first ‘What do you think’ question.

Brewer and Gardner (1996) asked the question ‘Who is this “We”?’ and distinguished
three forms of self:
1 Individual self – based on personal traits that differentiate the self from all others.
2 Relational self – based on connections and role relationships with significant others.
3 Collective self – based on group membership that differentiates ‘us’ from ‘them’.
More recently it has been proposed that there are four types of identity (Brewer, 2001;
Chen, Boucher, & Tapias, 2006):
1 Person-based social identities – emphasising the internalisation of group properties by
individual group members as part of their self-concept.
2 Relational social identities – defining the self in relation to specific other people with
whom one interacts in a group context – corresponding to Brewer and Gardner’s (1996)
relational identity and to Markus and Kitayama’s (1991) ‘interdependent self’.
3 Group-based social identities – equivalent to social identity as defined above.
4 Collective identities – referring to a process whereby group members not only share self-
defining attributes but also engage in social action to forge an image of what the group
stands for and how it is represented and viewed by others.
The relational self is interesting. Although in one sense it is an interpersonal form of self,
it can also be considered a particular type of collective self. For example, East Asian cultures
define groups in terms of networks of relationships (Yuki, 2003), and women place greater
importance than men on their relationships with others in their groups (Seeley, Gardner,
Pennington, & Gabriel, 2003; see also Baumeister & Sommer, 1997; Cross & Madson,
1997). East Asians and women are often considered to be more collectivist than Western
Europeans and men, respectively.
Table 4.1 shows one way in which different types of self and self-attributes could be clas-
sified according to level of identity (social versus personal) and type of attributes (identity
defining versus relationship defining).

Contextual sensitivity of self and identity

Evidence for multiple selves comes from research where contextual factors are varied to dis-
cover that people describe themselves and behave differently. For example, Russell Fazio and
his colleagues were able to get participants to describe themselves in very different ways by
 Many selves, multiple identities   131

asking them loaded questions that made them search through their stock of self-knowledge
for information that presented the self in a different light (Fazio, Effrein, & Falender, 1981).
Other researchers have found, time and time again, that experimental procedures that
focus on group membership lead people to act very differently from procedures that focus on
individuality and interpersonal relationships. Consider ‘minimal group’ studies in which
participants are either (a) identified as individuals or (b) explicitly categorized, randomly or
by some minimal or trivial criterion as group members (Tajfel, 1970; see Diehl, 1990, and
Chapter 11). A consistent finding is that being categorized makes people discriminate
against an outgroup, conform to ingroup norms, express attitudes and feelings that favour
the ingroup, and indicate a sense of belonging and loyalty to the ingroup. Furthermore,
these effects of minimal group categorization are generally very fast and automatic (Otten
& Wentura, 1999).
The idea that we have many selves, and that contextual factors bring different selves into
play, has a number of ramifications. Social constructionists have suggested that the self is
entirely situation-dependent. An extreme form of this position argues that we do not carry
self-knowledge around in our heads as cognitive representations at all; rather, we construct
disposable selves through talk (e.g. Potter & Wetherell, 1987; see the discussion of discourse
analysis in Chapters 1 and 15). A less extreme version has been proposed by Penny Oakes
(e.g. Oakes, Haslam, & Reynolds, 1999), who does not emphasise the role of talk but still
maintains that self-conception is highly context-dependent. A middle way is to argue that
people do have cognitive representations of the self that they carry in their heads as organis-
ing principles for perception, categorization and action, but that these representations are
temporarily or more enduringly modified by situational factors (e.g. Abrams & Hogg, 2001;
Turner, Reynolds, Haslam, & Veenstra, 2006).

In search of self-coherence
That we have many selves needs to be placed in perspective. Although we may have a diver-
sity of relatively discrete selves, we also have a quest: to find and maintain a reasonably
integrated picture of who we are. Self-conceptual coherence provides us with a continuing
theme for our lives – an ‘autobiography’ that weaves our various identities and selves
together into a whole person. People who have highly fragmented selves (e.g. some people
with schizophrenia, amnesia or Alzheimer’s disease) find it extraordinarily difficult to func-
tion effectively.
People use many strategies to construct a coherent sense of self (Baumeister, 1998). Here
are some that you may have used yourself:
● Restrict your life to a limited set of contexts. Because different selves come into play as
contexts keep changing, you will protect yourself from self-conceptual clashes.
● Keep revising and integrating your ‘autobiography’ to accommodate new identities.
Along the way, get rid of any worrisome inconsistencies. In effect, you are rewriting your
history to make it work to your advantage (Greenwald, 1980).
● Attribute changes in the self externally to changing circumstances, rather than internally
to fundamental changes in who you are. This is an application of the actor–observer Actor–observer effect
effect (Jones and Nisbett, 1972; see also Chapter 3). Tendency to attribute our
own behaviours externally
We can also develop a self-schema that embodies a core set of attributes that we feel dis- and others’ behaviours
internally.
tinguishes us from all other people – that makes us unique (Markus, 1977; see ‘Self-schemas’
discussed earlier in this chapter). We then tend to recognise these attributes disproportion-
ately in all our selves, providing thematic consistency that delivers a sense of a stable and
unitary self (Cantor & Kihlstrom, 1987).
In summary, people find ways to construct their lives such that their self-conceptions
appear steady and coherent.
132  Chapter 4   Self and identity

Social identity theory

Social identity theory Because social identity theory is a theory of both self and identity, and group and inter-
Theory of group group behaviour, we say a little about it here, but we discuss it fully in Chapter 11 (see
membership and intergroup
relations based on self-
Abrams & Hogg, 2010; Hogg, 2006; Hogg & Abrams, 1988, 2003).
categorization, social Social identity theory has its origins in research by Henri Tajfel on social categorization,
comparison and the intergroup relations, social comparison, and prejudice and stereotyping (e.g. Tajfel, 1969,
construction of a shared 1974) – often called the social identity theory of intergroup relations (Tajfel & Turner,
self-definition in terms of
ingroup-defining properties.
1986). Later developments by John Turner and his associates specified the role of social cat-
egorization of self and others to broaden the theory to understand group behaviour more
generally (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) – called the social identity
Self-categorization theory ­theory of the group, or self-categorization theory.
Turner and associates’
theory of how the process
of categorizing oneself as a Personal identity and social identity
group member produces
social identity and group As noted above, social identity theorists propose the existence of two broad classes of iden-
and intergroup behaviours. tity that define different types of self: (1) social identity, which defines the self in terms of
group memberships (e.g. one’s ethnicity), and (2) personal identity, which defines the self in
terms of idiosyncratic personal relationships and traits (e.g. one’s relationship with one’s
romantic partner, or being witty). We have as many social identities as there are groups that
we feel we belong to, and as many personal identities as there are interpersonal relationships
we are involved in and clusters of idiosyncratic attributes that we believe we possess.
Social identity, our main focus in this section, is associated with group and intergroup behav-
iours such as ethnocentrism, ingroup bias, group solidarity, intergroup discrimination, con-
formity, normative behaviour, stereotyping and prejudice. Social identity can be a very important
aspect of our self-concept. For example, Citrin, Wong and Duff (2001) describe a study in which
46 per cent of Americans reported that they felt being an American, a social identity, was the
most important thing in their life. In contrast, personal identity is associated with positive and
negative close interpersonal relationships and with idiosyncratic personal behaviour.

Processes of social identity salience

In any given situation, our sense of self and associated perceptions, feelings, attitudes and
behaviour rests on whether social or personal identity, and which specific social or personal
identity, is the psychologically salient basis of self-conception. The principle that governs social
identity salience hinges on the process of social categorization (Oakes, 1987) and on people’s
motivation to make sense of and reduce uncertainty about themselves and others (Hogg, 2012),
and to feel relatively positive about themselves (e.g. Abrams & Hogg, 1988) – see Figure 4.4.
People use limited perceptual cues (what someone looks like, how they speak, what atti-
tudes they express, how they behave) to categorise other people. Generally, we first ‘try out’
categorizations that are readily accessible to us because we often use them, because they are
important to us or perhaps because they are glaringly obvious in the situation. The catego-
rization brings into play all the additional schematic information we have about the cate-
Prototype
Cognitive representation of
gory. This information is cognitively stored as a prototype, which describes and prescribes
the typical/ideal defining the attributes of the category in the form of a fuzzy set of more or less related attributes,
features of a category. rather than a precise checklist of attributes.
Meta-contrast principle
Category prototypes accentuate similarities within groups, but they also accentuate dif-
The prototype of a group is ferences between groups – they obey what is called the meta-contrast principle. As such,
that position within the group prototypes usually do not identify average or typical members or attributes, but ideal
group that has the largest members or attributes. The content of a group prototype may also vary from situation to
ratio of ‘differences to
ingroup positions’ to
situation. For example, Britishness will probably be slightly different in a situation where
‘differences to outgroup one is interacting with other ‘Brits’ than a situation where one is interacting with Americans.
positions’. Category attributes in memory interact with situational factors to generate the
 Social identity theory   133

Social identity
salience
The wearing of the kilt is
a mark of nationalism,
commitment to the
cause and a resolve to
act in unison in times
of stress.

Social categorization ‘X’ is situationally

accessible and chronically accessible

Does categorization ‘X’ have good structural fit?

Does it account for relevant similarities and If no – try a new
differences between people in the context? categorization

If yes – does categorization ‘X’ have good

normative fit? Does it make sense of people’s If no – try a new
behaviour in the context? categorization

If yes – does categorization ‘X’ satisfy If no – try a new

uncertainty reduction in that context? categorization

If yes – does categorization ‘X’ satisfy If no – try a new

self-enhancement motives in that context? categorization

Figure 4.4 Social

identity theory’s
model of the
If yes – categorization ‘X’ is the psychologically sequence through
salient basis for self-conception in that context. which a particular self-
conception becomes
psychologically salient
in a specific context
134  Chapter 4   Self and identity

situation-specific prototype. Category attributes stored in memory act as an anchor that

ensures the integrity of the core identity and imposes limits on the amount and type of influ-
ence situation can have on the prototype (Van Bavel & Cunningham, 2010).
Ultimately, if the categorization fits, in that it accounts for similarities and differences
between people satisfactorily (called structural fit), and it makes good sense of why people
are behaving in particular ways (called normative fit), then the categorization becomes psy-
chologically salient as the basis of categorizing self and others.

Consequences of social identity salience

When a categorization becomes psychologically salient, people’s perception of themselves and
others becomes depersonalised. This means that people no longer consider themselves or oth-
ers as unique multidimensional persons but as more or less complete embodiments of the cat-
egory prototype – they are viewed through the relatively narrow lens of a group membership
that is defined by the specific ingroup or outgroup prototype. Swann and colleagues have sug-
gested that when this process is extreme, identity fusion arises such that one’s personal identity
becomes fused with the group and thus with social identity (Swann, Jetten, Gomez, Whitehouse,
& Bastian, 2012); and then, because there is no prototype-based differentiation of self within
the group, behavior can become extreme (Swann, Gómez, Seyle, Morales, & Huici, 2009).
In addition to the transformation of self-conception into social identity, people also think,
feel, believe and behave in terms of the relevant prototype. The process produces the range
of behaviour we characteristically associate with people in groups and with the way groups
treat each other, a theme that recurs throughout this text.
The actual nature of the behaviour (what people think and do) depends on the specific
content of the relevant prototype, and on people’s beliefs about the status of their group in
society and about the nature of the relations between groups (Tajfel & Turner, 1986; see
Ellemers, 1993; Hogg & Abrams, 1988). Group status is important because groups define
social identity and social identity defines our self-concept; thus, the evaluative implications
of a specific group (the status, prestige and regard in which it is held) reflect the esteem in
which others hold us, and they influence the esteem in which we hold ourselves, our self-
esteem (Crocker & Major, 1994; see the discussion of social stigma in Chapter 10).
So, people strive for membership in prestigious groups, or they strive to protect or enhance
the prestige and esteem of their existing group. How they do this is influenced by their
understanding of the nature of the status relations between their group and a specific
­outgroup – is it permeable, is it stable, is it legitimate? If the group’s evaluation in society is
generally unfavourable and you feel you can pass into a more prestigious group, you might
try to leave the group entirely; however, this can be very difficult, because in reality the psy-
chological boundaries between groups can be impermeable or impassable. For example,
various immigrant groups in Britain may find it difficult to ‘pass’ as British because they
simply do not look British or they are readily ‘given away’ by subtle clues in their accent. If
‘passing’ is not possible, people can try to make sure that the attributes that do define their
group are positive ones, or they can focus attention on less prestigious groups, in compari-
son with which they will look rather good.
Groups can sometimes recognise that the entire basis on which their group is considered low
status is illegitimate, unfair and unstable. If this recognition is tied to feasible strategies for
change, then groups will compete directly with one another to gain the upper hand in the status
stakes – a competition that can range from rhetoric and democratic process to terrorism and war.

Self-motives
Because selves and identities are critical reference points for leading a well-adapted life, peo-
ple are enthusiastically motivated to secure self-knowledge. Entire industries are based on
this search for knowledge, ranging from personality tests to dubious practices such as
 Self-motives  135

astrology and palmistry. However, people do not go about this search in a dispassionate way;
they have preferences for what they would like to know about themselves and can be dis-
mayed when the quest unearths things they did not expect or did not want to find.
Social psychologists have identified three classes of motive that interact to influence self-
construction and the search for self-knowledge:
● self-assessment motivates pursuit of valid information about self;
● self-verification motivates pursuit of information that is consistent with our own self-image;
● self-enhancement motivates pursuit of information that makes us look good.

Self-assessment and self-verification

The first motive is a simple desire to have accurate and valid information about oneself –
there is a self-assessment motive (e.g. Trope, 1986). People strive to find out the truth about Self-assessment
themselves, regardless of how unfavourable or disappointing the truth may be. The motivation to seek out
new information about
But people also like to engage in a quest for confirmation – to confirm what they already ourselves in order to find
know about themselves they seek out self-consistent information through a self-verification out what sort of person we
process (e.g. Swann, 1987). So, for example, people who have a negative self-image will actu- really are.
ally seek out negative information to confirm the worst. Although the ‘self’ in self-­verification Self-verification
was originally viewed as the idiosyncratic personal self, research shows that self-verification Seeking out information
can also occur at the group level. People seek information and behave in ways aimed at that verifies and confirms
­verifying their social identity (Chen, Chen, & Shaw, 2004). what we already know
about ourselves.

Self-enhancement
Self-enhancement
Above all else, we like to learn good things about ourselves – we seek new favourable knowl-
The motivation to develop
edge about ourselves as well as revise pre-existing but unfavourable views of ourselves. We and promote a favourable
are guided by a self-enhancement motive (e.g. Kunda, 1990). This motive to promote image of self.

Self-affirmation theory
‘Way to go, man!’
136  Chapter 4   Self and identity

self-positivity has a mirror motive, self-protection, which fends off self-negativity. Research
suggests that self-enhancement functions operate routinely and relatively globally, but that
self-protection functions are usually occasioned only by an event or series of events that
threatens a specific self-related interest (Alicke & Sedikides, 2009).
One manifestation of the self-enhancement motive is described by self-affirmation
Self-affirmation theory theory (Sherman & Cohen, 2006; Steele, 1988). People strive publicly to affirm positive
The theory that people aspects of who they are; this can be done blatantly by boasting or more subtly through
reduce the impact of threat
to their self-concept by
rationalisation or dropping hints. The urge to self-affirm is particularly strong when an
focusing on and affirming aspect of one’s self-esteem has been damaged. So, for example, if someone claims you are a
their competence in some lousy artist, you might retort that while that might be true, you are an excellent dancer. Self-
other area. affirmation rests on people’s need to maintain a global image of themselves as being compe-
tent, good, coherent, unitary, stable, capable of free choice, capable of controlling important
outcomes and so on. Ultimately, we like to be viewed as moral beings – and so we engage in
a range of behaviours aimed at establishing and even asserting our moral credentials
(Merritt, Effron, Fein, Savitsky, Tuller, & Monin, 2012; Monin & Miller, 2001). Box 4.3
describes research by Claude Steele (1975) in which self-affirmation processes were studied
in the context of religious adherence.
Which motive is more fundamental and more likely to prevail in the pursuit of self-­
knowledge – self-assessment, self-verification or self-enhancement? In a series of six experi-
ments, Constantine Sedikides (1993) pitted the three motives against one another. He used a
self-reflection task where participants can ask themselves more or less diagnostic questions
focusing on different aspects of themselves – the asking of more diagnostic questions indi-
cates greater self-reflection, and the focus of self-reflection differs depending on what
­self-motive is operating:
● Self-assessment – greater self-reflection on peripheral than central traits of self, whether
the attribute is desirable or not, indicates a drive to find out more about self (people
already have knowledge about traits that are central for them).
● Self-verification – greater self-reflection on central than on peripheral traits, whether the
attribute is positive or not, indicates a drive to confirm what one already knows about
oneself.

Box 4.3 Research classic

Self-affirmation in Salt Lake City

Claude Steele (1975) reported a study in Salt Lake City in knowledge that, as members of their community, they
which Mormon women who were at home during the day were either:
were telephoned by a female researcher posing as a com-
● un-cooperative with community projects (a direct
munity member. The researcher asked the women if they
threat to a core component of their self-concept), or
would be willing to list everything in their kitchen to assist
● un-concerned about driver safety and care (a threat to a
the development of a community food cooperative; those
relatively irrelevant component of their self-concept), or
who agreed would be called back the following week.
● cooperative with community projects (positive rein-
Because community cooperation is a very strong ethic
forcement of their self-concept).
among Mormons, about 50 per cent of women agreed to
this time-consuming request. Consistent with self-affirmation theory, the two threats
In addition to this baseline condition, there were three greatly increased the probability that women would sub-
other conditions in the study arising from a previous call, sequently agree to help the food cooperative – about 95
two days earlier, by an entirely unrelated researcher pos- per cent of women agreed to help. Among women who
ing as a pollster. In the course of this previous call, the had been given positive reinforcement of their self-­
pollster mentioned in passing that it was common concept, 65 per cent agreed to help the cooperative.
 Self-esteem  137

Box 4.4 Your life

Techniques to enhance or protect positive aspects of the self

You may have noticed how people (perhaps you!) are ● They try to dismiss interpersonal criticism as being
inclined to boost themselves. Think about all the ways you motivated by prejudice (e.g. Crocker & Major, 1989).
might do this . . . then read on. Here are some of the tricks ● They perform a biased search of self-knowledge to sup-
that people get up to – do they seem familiar to you? port a favourable self-image (e.g. Kunda & Sanitoso,
1989).
● They take credit for their successes but deny blame for
● They place a favourable spin on the meaning of ambig-
their failures (e.g. Zuckerman, 1979); this is one of the
uous traits that define self (e.g. Dunning, Meyerowitz, &
self-serving biases (see Chapter 3).
Holzberg, 1989).
● They forget failure feedback more readily than success
● They persuade themselves that their flaws are widely
or praise (e.g. Mischel, Ebbesen, & Zeiss, 1976).
shared human attributes but that their qualities are rare
● They accept praise uncritically but receive criticism
and distinctive (e.g. Campbell, 1986).
sceptically (e.g. Kunda, 1990).

● Self-enhancement – greater self-reflection on positive than on negative aspects of self,

whether the attribute is central or not, indicates a drive to learn positive things about self
(see Box 4.4).
Sedikides found that self-enhancement was strongest, with self-verification a distant sec-
ond and self-assessment an even more remote third. The desire to think well of ourselves
reigns supreme; it dominates both the pursuit of accurate self-knowledge and the pursuit of
information that confirms self-knowledge. (Does this apply to you? See the second ‘What do
you think?’ focus question.)
Because self-enhancement is so important, people have developed a formidable repertoire
of strategies and techniques to pursue it. People engage in elaborate self-deceptions to
enhance or protect the positivity of their self-concepts (Baumeister, 1998). It has even been
suggested that the name-letter effect, where people prefer letters that occur in their own
name over those that do not, reflects self-esteem and can actually be used as an indirect
measure of self-esteem (Hoorens, 2014).

Self-esteem
Why are people so strongly motivated to think well of themselves – to self-enhance? Research
suggests that people generally have a rosy sense of self – they see, or try to see, themselves
through ‘rose-tinted spectacles’. For example, people who are threatened or distracted often
display what Del Paulhus and Karen Levitt (1987) called automatic egotism – a widely favour-
able self-image. In their review of a link between illusions and a sense of well-being, Shelley
Taylor and Jonathon Brown (1988) concluded that people normally overestimate their good
points, overestimate their control over events and are unrealistically optimistic. Sedikides and
Gregg (2003) call these three characteristics of human thought the self-enhancing triad.
For example, a study conducted in an American setting found that very low-achieving stu-
dents (in the bottom 12 per cent) thought they were relatively high achievers (in the top 38 per
cent) (Kruger & Dunning, 1999). According to Patricia Cross (1977), your lecturers show
positivity bias too, with 94 per cent convinced that their teaching ability is above average! The
tendency to overestimate our good points is well documented in research (Brown, 2012;
Guenther & Alicke, 2010; Williams & Gilovich, 2012) and is referred to as the above-average-
effect. See Box 4.5 and Figure 4.5 for an applied example of this bias among young drivers.