|Note: This Lecture Supplement is based on chapters, co-authored with Nancy Cantor, that appeared in R.J. Sternberg, (Ed.), Handbook of Intelligence (2nd ed., 2000; 3rd ed., 2010. Links to published versions.|
The capacity to know oneself and to know others is as inalienable a part of the human condition as is the capacity to know objects or sounds, and it deserves to be investigated no less than these other"less charged" forms.Howard Gardner (1983, p. 243)
Frames of Mind
Intelligence, as defined in standard dictionaries, has two rather different meanings. In its most familiar meaning, intelligence has to do with the individual's ability to learn and reason. It is this meaning which underlies common psychometric notions such as intelligence testing, the intelligence quotient, and the like. In its less common meaning, intelligence has to do with a body of information and knowledge. This second meaning is implicated in the titles of certain government organizations, such as the Central Intelligence Agency in the United States, and its British counterparts MI-5 and MI-6. Similarly, both meanings are invoked by the concept of social intelligence. As originally coined by E.L. Thorndike (1920), the term referred tp the person's ability to understand and manage other people, and to engage in adaptive social interactions. More recently, however, Cantor and Kihlstrom (1987) redefined social intelligence to refer to the individual's fund of knowledge about the social world.
The psychometric view of social intelligence has its origins in E.L. Thorndike's (1920) division of intelligence into three facets, pertaining to the ability to understand and manage ideas (abstract intelligence), concrete objects (mechanical intelligence), and people (social intelligence). In his classic formulation:"By social intelligence is meant the ability to understand and manage men and women, boys and girls -- to act wisely in human relations" (p. 228). Similarly, Moss and Hunt (1927) defined social intelligence as the"ability to get along with others" (p. 108). Vernon (1933), provided the most wide-ranging definition of social intelligence as the person's"ability to get along with people in general, social technique or ease in society, knowledge of social matters, susceptibility to stimuli from other members of a group, as well as insight into the temporary moods or underlying personality traits of strangers" (p. 44).
Although the concept of social intelligence is generally attributed to Thorndike (1920), Landy (2006) notes that the term was actually first used by Dewey (1909), and then by Lull (1911). But neither of these writers ever did much with the idea, so credit goes properly to Thorndike.
The concept of social intelligence (though not by that name) was popularized by Dale Carnegie in his best-selling How to Win Friends and Influence People (1936). Reflecting the demise of Puritan and Quaker prescriptions for success, which emphasized diligent work and moral uprightness, Carnegie put the emphasis on psychology, and argued for the importance of personality. The key to success in 20th-century America was friendship and persuasion, and these social skills could be taught, and learned by anyone.
For an insightful biography of Carnegie, see Carnegie See Self-Help Messiah: Dale Carnegie and Success in Modern America by Steven Watts (2013). Watts notes that"With carnegie, the stress shifted from shaping one's inner moral character to shaping the impressions that one made upon other people -- what he described in his diary as 'the biggest problem I shall ever face: the management of Dale Carnegie'".
Carnegie also presaged the"positive psychology" and the"mindfulness movement" with his How to stop Worrying and Start Living (1948) -- but that's a story for another course.
By contrast, Wechsler (1939, 1958) gave scant attention to the concept. Wechsler did acknowledge that the Picture Arrangement subtest of the WAIS might serve as a measure of social intelligence, because it assesses the individual's ability to comprehend social situations (see also Rapaport, Gill, & Shafer, 1968; Campbell & McCord, 1996). In his view, however,"social intelligence is just general intelligence applied to social situations" (1958, p. 75). This dismissal is repeated in Matarazzo's (1972, p. 209) fifth edition of Wechsler's monograph, in which"social intelligence" dropped out as an index term.
intelligence seems easy enough, especially by analogy to
abstract intelligence. When it came to measuring
social intelligence, however, E.L. Thorndike (1920)
noted somewhat ruefully that"convenient
tests of social intelligence are hard to devise....
Social intelligence shows itself abundantly in the
nursery, on the playground, in barracks and factories
and salesroom (sic), but it eludes the formal
standardized conditions of the testing laboratory. It
requires human beings to respond to, time to adapt its
responses, and face, voice, gesture, and mien as tools" (p. 231).
Nevertheless, true to the goals of the psychometric
tradition, the abstract definitions of social
intelligence were quickly translated into standardized
laboratory instruments for measuring individual
differences in social intelligence (for additional
reviews, see Taylor, 1990; Taylor & Cadet, 1989;
Walker & Foley, 1973).
The first of these was the George
Washington Social Intelligence Test, (GWSIT; Hunt, 1928;
Moss, 1931; Moss, Hunt, Omwake, & Ronning, 1927; for
later editions, see Moss, Hunt, & Omwake, 1949; Moss,
Hunt, Omwake, & Woodward, 1955). Like the Stanford-Binet
Intelligence Test or Wechsler Adult Intelligence Scale, the
GWSIT was composed of a number of subtests, which can be
combined to yield an aggregate score. The subtests are:
Hunt (1928) originally validated the GWSIT through its correlations with adult occupational status, the number of extracurricular activities pursued by college students, and supervisor ratings of employees' ability to get along with people. However, some controversy ensued about whether social intelligence should be correlated with personality measures of sociability or extraversion (e.g., Strang, 1930; Thorndike & Stein, 1937). Most important, however, the GWSIT came under immediate criticism for its relatively high correlation with abstract intelligence. Thus, Hunt (1928) found that aggregate GWSIT score correlated r = .54 with aggregate score on the George Washington University Mental Alertness Test (GWMAT), an early IQ scale (see also Broom, 1928). A factor analysis by R.L. Thorndike (1936) indicated that the subtests of the GWSIT loaded highly on the same general factor as the subtests of the GWMAT. Woodrow (1939), analyzing the GWSIT with a much larger battery of cognitive tests, found no evidence for a unique factor of social intelligence. R.L. Thorndike and Stein (1937) concluded that the GWSIT"is so heavily loaded with ability to work with words and ideas, that differences in social intelligence tend to be swamped by differences in abstract intelligence" (p. 282).
The inability to discriminate between the social intelligence and IQ, coupled with difficulties in selecting external criteria against which the scale could be validated, led to declining interest in the GWSIT, and indeed in the whole concept of social intelligence as a distinct intellectual entity. Spearman's (1927) model of g afforded no special place for social intelligence, of course. Nor is social intelligence included, or even implied, in Thurstone's (1938) list of primary mental abilities.
After an initial burst of interest in the GWSIT, work on the assessment and correlates of social intelligence fell off sharply until the 1960s (Walker & Foley, 1973), when this line of research was revived within the context of Guilford's (1967) Structure of Intellect model. Guilford postulated a system of at least 120 separate intellectual abilities, based on all possible combinations of five categories of operations (cognition, memory, divergent production, convergent production, and evaluation), with four categories of content (figural, symbolic, semantic, and behavioral) and six categories of products (units, classes, relations, systems, transformations, and implications). Interestingly, Guilford considers his system to be an expansion of the tripartite classification of intelligence originally proposed by E.L. Thorndike. Thus, the symbolic and semantic content domains correspond to abstract intelligence, the figural domain to practical intelligence, and the behavioral domain to social intelligence.
Within Guilford's (1967) more differentiated system, social intelligence is represented as the 30 (5 operations x 6 products) abilities lying in the domain of behavioral operations. In contrast to its extensive work on semantic and figural content, Guilford's group addressed issues of behavioral content only very late in their program of research. Nevertheless, of the 30 facets of social intelligence predicted by the structure-of-intellect model, actual tests were devised for six cognitive abilities (O'Sullivan et al., 1965; Hoepfner & O'Sullivan, 1969) and six divergent production abilities (Hendricks, Guilford, & Hoepfner, 1969).
O'Sulivan et al. (1965) defined the category of behavioral cognition as representing the"ability to judge people" (p. 5) with respect to"feelings, motives, thoughts, intentions, attitudes, or other psychological dispositions which might affect an individual's social behavior" (O'Sullivan et al., p. 4). They made it clear that one's ability to judge individual people was not the same as his or her comprehension of people in general, or"stereotypic understanding" (p. 5), and bore no a priori relation to one's ability to understand oneself. Apparently, these two aspects of social cognition lie outside the standard structure-of-intellect model.
In constructing their tests of behavioral cognition, O'Sullivan et al. (1965) assumed that"expressive
behavior, more particularly facial expressions, vocal
inflections, postures, and gestures, are the cues from
which intentional states are inferred"
(p. 6). While recognizing the value of assessing the
ability to decode these cues in real-life contexts with
real people serving as targets, economic constraints
forced the investigators to rely on photographs, cartoons,
drawings, and tape recordings (the cost of film was
prohibitive); verbal materials were avoided wherever
possible, presumably in order to avoid contamination of
social intelligence by verbal abilities. In the final
analysis, O'Sullivan et. al
developed at least three different tests within each
product domain, each test consisting of 30 or more
separate items -- by any standard, a monumental effort at
theory-guided test construction. The six cognitive
abilities defined by O'Sullivan
et al. were:
After devising these tests, O'Sullivan et al. (1965) conducted a normative study in which 306 high-school students received 23 different social intelligence tests representing the six hypothesized factors, along with 24 measures of 12 non-social ability factors. A principal factor analysis with orthogonal rotation yielded 22 factors, including the 12 non-social reference factors and 6 factors clearly interpretable as cognition of behavior. In general, the six behavioral factors were not contaminated by non-social semantic and spatial abilities. Thus, O'Sullivan et al. apparently succeeded in measuring expressly social abilities which were essentially independent of abstract cognitive ability. However, echoing earlier findings with the GWSIT, later studies found substantial correlations between IQ and scores on the individual Guilford subtests, as well as various composite social intelligence scores (Riggio, Messamer, & Throckmorton, 1991; Shanley, Walker, & Foley, 1971). Still Shanley et al. conceded that the correlations obtained were not strong enough to warrant the conclusion (e.g., Wechsler, 1958) that social intelligence is nothing more than general intelligence applied in the social domain.
In one of the last test-construction efforts by Guilford's group, Hendricks et. al (1969)
attempted to develop tests for coping with other people,
not just understanding them through their behavior -- what
they referred to as"basic
solution-finding skills in interpersonal relations" (p. 3). Because successful coping
involves the creative generation of many and diverse
behavioral ideas, these investigators labelled these
divergent-thinking abilities creative social
intelligence. The six divergent production abilities
defined by Hendricks et al. were:
As with the behavioral cognition abilities studied by O'Sullivan et al. (1965), the very nature of the behavioral domain raised serious technical problems for test development in the behavioral domain, especially with respect to contamination by verbal (semantic) abilities. Ideally, of course, divergent production would be measured in real-world settings, in terms of actual behavioral responses to real people. Failing that, testing could rely on nonverbal behaviors such as drawings, gestures, and vocalizations, but such tests could well be contaminated by individual differences in drawing, acting, or public-speaking ability that have nothing to do with social intelligence per se.
Still, Following the pattern of O'Sullivan et al., (1965), a battery of creative social intelligence tests, 22 for divergent production of behavioral products and another 16 representing 8 categories of cognition of behavior and divergent production in the semantic domain, was administered to 252 high-school students. As might be expected, scoring divergent productions proved considerably harder than scoring cognitions, as in the former case there is no one best answer, and the subject's responses must be evaluated by independent judges for quality as well as quantity. Principal-components analysis yielded 15 factors, with six factors clearly interpretable as divergent production in the behavioral domain. Again, the divergent-production abilities in the behavioral domain were essentially independent of both divergent semantic production and (convergent) cognition in the behavioral domain.
A later study by Chen and Michael (1993), employing more modern factor-analytic techniques, essentially confirmed these findings. In addition, Chen and Michael extracted a set of higher-order factors which largely conformed to the theoretical predictions of Guilford's (1981) revised structure-of-intellect model. A similar re-analysis of the O'Sullivan et al. (1965) has yet to be reported.
In summary, Guilford and his colleagues were successful in devising measures for two rather different domains of social intelligence: understanding the behavior of other people (cognition of behavioral content), and coping with the behavior of other people (divergent production of behavioral content). These component abilities were relatively independent of each other within the behavioral domain, and each was also relatively independent of the non-behavioral abilities, as predicted (and required) by the structure-of-intellect model.
Despite the huge amount of effort that the Guilford group invested in the measurement of social intelligence, it should be understood that the studies of O'Sullivan et al. (1965) and Hendricks et al. (1969) went only part of the way toward establishing the construct validity of social intelligence. Their studies described essentially established convergent and discriminant validity, by showing that ostensible tests of the various behavioral abilities hung together as predicted by the theory, and were not contaminated by other abilities outside the behavioral domain. As yet, there is little evidence for the ability of any of these tests to predict external criteria of social intelligence.
Tests of the
remaining three structure-of-intellect domains (memory,
convergent production, and evaluation) had not developed
by the time the Guilford program came to a close.
Hendricks et al. (1969) noted that"these constitute by far
the greatest number of unknowns in the [Structure of
Intellect] model" (p. 6).
However, O'Sullivan et al.
(1965) did sketch out how these abilities were defined.
Convergent production in the behavioral domain
was defined as"doing
the right thing at the right time"
(p. 5), and presumably might be tested by a knowledge of
etiquette. Behavioral memory was defined as the
ability to remember the social characteristics of people
(e.g., names, faces, and personality traits), while behavioral
evaluation was defined as the ability to judge the
appropriateness of behavior.
Following the Guilford
studies, a number of investigators continued the attempt
to define social intelligence and determine its relation
to general abstract intelligence. Most of these studies
explicitly employ the logic of the
multitrait-multimethod matrix (Campbell & Fiske,
1959), employing multiple measures of social and
nonsocial intelligence, and examining the convergent
validity of alternative measures within each domain, and
their discriminant validity across domains (e.g.,
Sechrest & Jackson, 1961).
For example, Keating (1978) measured social intelligence with a battery of instruments including Rest's (1975) Defining Issues Test, derived from Kohlberg's (1963) theory of moral development; Chapin's (1942) Social Insight Test, which asks the subject to resolve various social dilemmas; and Gough's (1966) Social Maturity Index, a self-report scale derived from the California Psychological Inventory measuring effective social functioning. Applying a multitrait-multimethod analysis, Keating found no evidence that social intelligence, so defined, was discriminable from academic intelligence. Thus, the average correlation between tests within each domain was actually lower than the corresponding average across domains. While a factor analysis produced two factors, each of these factors consisted of a mix of the two types of intelligence test. Finally, Keating found that the three measures of abstract intelligence were actually better predictors of Gough's (1966) Social Maturity Index than were the remaining two measures of social intelligence. However, it should be noted that Keating's putative measures of social intelligence are highly verbal in nature, so some contamination by abstract verbal and reasoning ability might be expected.
In response to Keating's (1978) study, Ford and Tisak (1983) conducted an even more substantial study involving over 600 high-school students. Four measures of verbal and mathematical ability were derived from school records of grades and standardized test scores. Social intelligence was measured by self-, peer-, and teacher-ratings of social competence, Hogan's (1969) empathy test, self-reports of social competence, and a judgment based on an individual interview. In contrast to Keating's (1968) results, Ford and Tisak found that the measures of academic and social intelligence loaded on different factors. Moreover, the three ratings of social competence and Hogan's empathy scale were more highly predictive of the interview ratings of social competence than were the academic measures. Ford and Tisak attributed these results to the selection of social intelligence measures according to a criterion of behavioral effectiveness in social situations, rather than cognitive understanding of them. Put another way: measures of verbal ability, including standard measures of IQ, are likely to correlate highly with verbal measures, but not nonverbal measures, of social intelligence.
Similar findings were obtained by a number of other investigators (e.g., Brown & Anthony, 1990), including Marlowe (1986; Marlowe & Bedell, 1982), who assembled a large battery of personality measures ostensibly tapping various aspects of social intelligence. Factor analysis of these instruments yielded five dimensions of social intelligence: interest and concern for other people, social performance skills, empathic ability, emotional expressiveness and sensitivity to others' emotional expressions, and social anxiety and lack of social self-efficacy and self-esteem. Factor scores on these dimensions of social intelligence were essentially unrelated to measures of verbal and abstract intelligence.
In evaluating studies like Marlowe's (1986), however, it should be noted that the apparent independence of social and general intelligence may be at least partially an artifact of method variance. Unlike the GWSIT, and the batteries of cognitive and divergent-production measures devised by the Guilford group, Marlowe's ostensible measures of social intelligence are all self-report scales, whereas his measures of verbal and abstract intelligence were the usual sorts of objective performance tests. The difference in data collection methods alone may explain why the social and verbal/abstract dimensions lined up on different factors. In any event, the measurement of individual differences in social intelligence by means of self-report scales is a major departure from the tradition of intelligence testing, and it seems important to confirm Marlowe's findings using objective performance measures of the various facets of social intelligence.
For example, Frederickson, Carlson, & Ward (1984) employed an extensive behavioral assessment procedure, along with a battery of performance tests of scholastic aptitude and achievement and medical and non-medical problem solving. In addition, each subject conducted 10 interviews with simulated medical patients and nonmedical clients. Based on codings of their interview behavior, each subject received ratings of organization, warmth, and control. None of the measures of aptitude, achievement, or problem-solving behavior correlated substantially with any of the interview-based ratings of social intelligence. Lowman and Leeman (1988), employing a number of performance measures, obtained evidence for three dimensions of social intelligence: social needs and interests, social knowledge, and social ability. Interestingly, the correlations of all three dimensions with grade point average, a proxy for academic intelligence, were either null or negative.
On the other hand, Stricker and Rock (1990) administered a battery of performance measures of social intelligence, and found that subjects' accuracy in judging a person and a situation portrayed in a videotaped interview was correlated with verbal ability. Wong, Day, Maxwell, and Meara (1995) constructed measures of social perception (accuracy in decoding verbal and nonverbal behavior), social insight (accuracy in interpreting social behavior) and social knowledge (awareness of the rules of etiquette). Factor analysis showed that social perception and insight were closely related, neither of these dimensions was closely related to social knowledge, and none of the social abilities was related to traditional academic ability.
Expanding on the study by Wong et al., Jones and Day (1997) based their analysis on Cattell's (1971) distinction between fluid and crystallized intelligence. In the social domain, crystallized intelligence reflects the individual's accumulated fund of knowledge about the social world, including his or her vocabulary for representing social behaviors and situations; fluid intelligence, by contrast, reflects the individual's ability to quickly and accurately solve problems posed by novel social situations. Jones and Day assembled four measures of each kind of ability, including verbal and pictorial performance measures, self-ratings, and teacher ratings. They also had multiple measures of academic ability. Confirmatory factor analyses testing various specific models of the relations between social and academic intelligence indicated that crystallized social intelligence was discriminable from fluid social intelligence, but not from academic intelligence.
Clearly, more studies employing performance-based measures are needed before any definitive conclusions can be drawn about the relations among various aspects of social intelligence (convergent validity) and the relations between social intelligence and other intellectual abilities (discriminant validity).
Although social intelligence has
proved difficult for psychometricians to operationalize,
it does appear to play a major role in people's naive, intuitive concepts of
intelligence. Following up on earlier work by Rosch
(1978), Cantor (Cantor & Mischel, 1979; Cantor,
Smith, French, & Mezzich, 1980), and Neisser (1979),
Sternberg and his colleagues asked subjects to list the
behaviors which they considered characteristic of
intelligence, academic intelligence, everyday
intelligence, and unintelligence; two additional groups
of subjects rated each of 250 behaviors from the first
list in terms of how"characteristic" each was of the ideal person
possessing each of the three forms of intelligence
(Sternberg, Conway, Ketron, & Bernstein, 1981).
Factor analysis of ratings provided by laypeople yielded
a factor of"social
competence" in each context.
Prototypical behaviors reflecting social competence
Interestingly, a separate dimension of social competence did not consistently emerge in ratings made by a group of experts on intelligence. Rather, the experts' dimensions focused on verbal intelligence and problem-solving ability, with social competence expressly emerging only in the ratings of the ideal"practically intelligent" person. Perhaps these experts shared Wechsler's (1939) dismissive view of social intellience.
A similar study was performed by
Kosmitzki and John (1993). Based largely on prior
research by Orlik (1978), these investigators assembled
a list of 18 features which make up people's implicit concept of social
intelligence. When subjects were asked to rate how
necessary each feature was to their own personal
understanding of social intelligence, the following
dimensions emerged as most central to the prototype:
In another part of the study, subjects were asked to rate someone they liked on each of these attributes. After statistically controlling for differential likability of the traits, a factor analysis yielded a clear dimension of social intelligence, defined by the attributes listed above. The remaining two factors were named social influence and social memory.
A recent psychometric study of social intelligence used a methodology similar to that of Sternberg et al. (1981) and Kosmitzki & John (1993). Schneider, Ackerman, and Kanfer (1996) asked subjects to generate descriptions of socially competent behavior. These descriptors were then collated and reduced to form a Social Competence Questionnaire, in which subjects are asked to rate the extent to which each item described their typical social behavior. A factor analysis revealed seven dimensions of social competence: extraversion, warmth, social influence, social insight, social openness, social appropriateness, and social maladjustment. Composite scores on these dimensions were essentially uncorrelated with measures of quantitative and verbal/reasoning ability. On the basis of these findings, Schneider et al. concluded that"it is time to lay to rest any residual notions that social competence is a monolithic entity, or that it is just general intelligence applied to social situations" (p. 479). As with Marlowe's (1986) study, however, the reliance on self-report measures of social intelligence compromises this conclusion, which remains to be confirmed using objective performance measures of the various dimensions in the social domain.
Sternberg et al. (1981) has noted that in contrast to explicit theories of intelligence, which attempt to explain what intelligence is, implicit theories attempt to capture people's views of what the word intelligence means. Social intelligence played little role in Sternberg's early componential view of human intelligence (Sternberg, 1977, 1980; but see Sternberg, 1984), which was intended to focus on reasoning and problem-solving skills as represented by traditional intelligence tests. However, social intelligence is explicitly represented in Sternberg's more recent triarchic view of intelligence (Sternberg, 1984, 1985, 1988). According to the triarchic theory, intelligence is composed of analytical, creative, and practical abilities. Practical intelligence is defined in terms of problem-solving in everyday contexts, and explicitly includes social intelligence (Sternberg & Wagner, 1986). According to Sternberg, each type of intelligence reflects the operation of three different kinds of component processes: performance components, which solve problems in various domains; executive metacomponents, which plan and evaluate problem-solving; and knowledge-acquisition components, by which the first two components are learned. To complicate things further, Sternberg (1985, 1988) argues that the measurement of all forms of intelligence is sensitive to the context in which it is assessed. This may be especially the case for practical and social intelligence: for example, the correct answer to a question of social judgment may well be different if it is posed in a corporate (Wagner & Sternberg, 1985) or military (Legree, 1995) context.
For Sternberg, these abilities, and thus their underlying components, may well be somewhat independent of each other. There is no implication, for example, that a person who is strong on analytical intelligence will also be strong in creative and practical intelligence. In any event, the relation among various intellectual abilities is an empirical question. Answering this question, of course, requires that we have adequate instruments for assessing social intelligence -- tests which adequately sample the domain in question, in addition to being reliable and valid. At present, these instruments do not appear to exist. However, future investigators who wish to make the attempt might be well advised to begin with the intuitive concept of social intelligence held in the mind of the layperson. After all, social intelligence is a social construct, not just an academic one.
While the psychometric research just reviewed has focused -- though not quite exclusively -- on normal adults, there is also a longstanding interest in social intelligence among developmental psychologists (for a review, see Greenspan & Love, 1997) -- particularly among those psychologists concerned with the assessment, treatment, and rehabilitation of children (and adults) with developmental disorders such as mental retardation and autism.
Of course, social intelligence has always played a role in the concept of mental retardation. This psychiatric diagnosis requires not only evidence of subnormal intellectual functioning (i.e., IQ < 70) but also demonstrated evidence of impairments in"communication, self-care, home living, social and interpersonal skills, use of community resources, self-direction, functional academic skills, work, leisure,health, and safety" (American Psychiatric Association, 1994, p. 46). In other words, the diagnosis of mental retardation involves deficits in social as well as academic intelligence. Furthermore, the wording of the diagnostic criteria implies that social and academic intelligence are not highly correlated -- it requires positive evidence of both forms of impairment, meaning that the presence of one cannot be inferred from the presence of the other.
While the conventional diagnostic criterion for mental retardation places primary emphasis on IQ and intellectual functioning Greenspan (1979) has argued that it should emphasize social and practical intelligence instead. To this end, Greenspan proposed a hierarchical model of social intelligence. In this model, social intelligence consists of three components: social sensitivity, reflected in role-taking and social inference; social insight, including social comprehension, psychological insight, and moral judgment; and social communication subsuming referential communication and social problem solving. Social intelligence, in turn, is only one component of adaptive intelligence (the others being conceptual intelligence and practical intelligence), which in turn joins physical competence and socioemotional adaptation (temperament and character) as the major dimensions of personal competence broadly construed. Greenspan did not propose specific tests for any of these components of social intelligence, but implied that they could be derived from experimental procedures used to study social cognition in general.
All this is well and good, but while the criterion for impaired intellectual functioning is clearly operationalized by an IQ threshold, there is as yet no standard by which impaired social functioning -- impaired social intelligence can be determined. The Vineland Social Maturity Scale (Doll, 1947) was an important step in this direction: this instrument, which yields aggregate scores of social age (analogous to mental age) and social quotient (by analogy to the intelligence quotient, calculated as social age divided by chronological age). However, it is a telling point that this instrument for evaluating social intelligence and other aspects of adaptive behavior was introduced almost a half century after the first IQ scale was introduced by Binet and Simon. The Vineland, which has been recently revised (Sparrow, Balla, & Cicchetti, 1984), but its adequacy as a measure of social intelligence is compromised by the fact that linguistic functions, motor skills, occupational skills, and self-care and self-direction are assessed as well as social relations. As an alternative, Taylor (1990) has proposed a semistructured Social Intelligence Interview covering such domains as social memory, moral development, recognition of and response to social cues, and social judgment. However, Taylor concedes that such an interview, being idiographically constructed to take account of the individual's particular social environment, cannot easily yield numerical scores by which individuals can be compared and ranked. More important than ranking individuals, from Taylor's point of view, is identify areas of high and low functioning within various environments experienced by the individual, and to determine the goodness of fit between the individual and the environments in which he or she lives. This latter goal, of course, is a primary thrust of the social intellience view of personality espoused by Cantor and Kihlstrom (1987).
Another group of developmental disabilities, autistic spectrum disorders, also invokes the concept of social intelligence. Kanner's (1943) classic description of autism portrays children who do not seem to be capable of engaging in normal social behavior or of maintaining normal social relationships, and the diagnostic criteria specified in DSM-IV emphasize deficits in social relations: impairments in nonverbal behavior, failures to develop peer relationships, lack of spontaneous sharing and other aspects of social reciprocity; impairments in communication, including an inability to initiate or sustain conversations or social imitative play; and stereotyped patterns of behavior, including inflexibility in various behavioral routines. All of these features suggest that autism is characterized not just by social withdrawal and language impairment, but a specific impairment in the abilities that underlie effective social interaction.
Specifically, it has been proposed by Leslie (1987) and Baron-Cohen (1995), among others, that autistic children and adults lack a"theory of mind" (Premack & Woodruff, 1978; see also Flavell, Green, & Flavell, 1995; Gopnik & Meltzoff, 1997; Wellman, 1990) by which they can attribute mental states to other people and reflect on their own mental life (for a summary review, see Klein & Kihlstrom, 1998). For example, Baron-Cohen, Leslie, and Frith (1985) suggested that the core deficit in autism is that the affected children are unable to appreciate that other people's beliefs, attitudes, and experiences might differ from their own. This hypothesis brought the problem of assessing social intelligence in disabled populations (including mental retardation and learning disability as well as autism; see Greenspan & Love, 1997) directly in contact with a literature on the development of social cognition in normal children which had been emerging since the 1970s (Flavell, 1974; Flavell & Ross, 1981; Shantz, 1975). In this way, scientific understanding of social cognition in general began to influence research and theory on individual differences in social cognition.
Still, the problem remains. Is the core deficit in autism one of social intelligence, as Baron-Cohen (1995) claims? In this respect, it is interesting to note, along with Gardner (1983), that autistic individuals can show an impaired ability to understand others' mental states, but spared abilities to deal cognitively with nonsocial objects and events, as well as to comprehend social situations where they are not required to understand another persons knowledge, belief, feelings, and desires. On the other hand, Bruner and Feldman (1993) have proposed that these deficits in social cognition are secondary to deficits in general cognitive functioning. Thus, although research on normal and abnormal development is more closely in contact with general social-cognitive theory than before, the fundamental questions endure: Is social cognition a separate faculty from nonsocial cognition? Is social intelligence anything different from general intelligence applied to the social domain?
Another trend contributing to revived interest in social intelligence was the upsurge of interest in moral reasoning following the publication of Kohlberg's Piagetian theory of moral reasoning (e.g., Kohlberg, 1963). As Turiel (Turiel, 2006) notes, Piaget himself had viewed moral reasoning within the wider context of the child's knowledge and judgment of social relationships. So, just as Thorndike raised the question of how social intelligence related to academic intelligence, so the Piaget-Kohlberg trend raised the question of how age differences in moral reasoning were related to social reasoning in general. One answer is that they do not relate much at all, because moral judgments are based on unconscious, intuitive processes that are based more on emotion than reason; in this view, the reasons we give for our judgments are little more than after-the-fact rationalizations (e.g., Greene, Sommerville, Nystrom, Darley, & Cohen, 2001; Haidt, 2001). Another approach is that moral reasoning, while obviously related to social reasoning and to reasoning in general, constitutes a separate domain of reasoning that might follow its own unique principles, developmental trajectory, and the like. This does not rule out a role for emotional processes, but it keeps social cognition at the center of the study of moral reasoning
According to social-cognitive domain theory (Turiel, Killen, & Helwig, 1987; Smetana, 2006), morality is only one of several aspects of the social world about which children and adults acquire knowledge, and about which they engage in reasoning, judgment, and decision-making. The "conventional" domain of social knowledge has to do with norms of social behavior that vary from one context to another. The"personal" domain has to do with our understanding of individual persons as psychological entities, including the attributions that we make for our own and others' behaviors, and our ability to infer meaning in social situations. The"moral" domain concerns universally applicable and obligatory concepts of harm, welfare, fairness, and rights. Most of the focus in social-cognitive domain theory has been on the moral domain, and on children's developing ability to understand moral concepts and render judgments of right and wrong. As a developmental theory, social-cognitive domain theory assumes that social-cognitive abilities are heterogeneous -- that children's (and adults') abilities to reason about the social world, and the trajectory of their development, may well differ from one domain to another. But for present purposes, social-cognitive domain theory offers an alternative description of the domains in which children and adults apply distinctively social intelligence.
As psychologists are fond of
saying, further research is needed to answer these
questions. However, we can hope that future research
on social intelligence will have a different character
than it has had in the past. One of the most salient,
and distressing, features of the history of
intelligence is how little contact there has been
between the instruments by which we assess individual
differences in intellectual ability and our
understanding of the processes which supply the
cognitive substrate of intellectual ability
(Sternberg, 1977). The IQ test, once touted as"psychology's most telling accomplishment
to date" (Herrnstein, 1974,
p. 62), is almost entirely atheoretical, having been
pragmatically constructed to model the sorts of things
which children do in school. So too with social
intelligence, which all too often has been
conceptualized informally, and assessed by means of a
jury-rigged assortment of tests (Walker & Foley,
1973). Perhaps new theoretical approaches, such as the
social-intelligence view of personality and the"theory
of mind" view of
development, will change this situation, so that
future reviews of this sort will be able to describe
assessments of social intelligence which are grounded
in an understanding of the general social-cognitive
processes out of which individual differences in
social behavior emerge.
Reviewing the literature published up to 1983, Landy (2006) characterized the search for social intelligence as"long, frustrating, and fruitless". Certainly it has been long and frustrating. Decade by decade, Landy traces a record of"disappointing empirical results and substantial theoretical criticism" (p. 82). This record did not, however, diminish the enthusiasm of both basic and applied social psychologists for the concept of social intelligence. Landy's review essentially stopped at 1983, and for good reason -- for very soon events were to give social intelligence a new lease on life.
An exception to the general rule that
social intelligence plays little role in scientific
theories of intelligence is the theory of multiple
intelligences proposed by Gardner (1983, 1993,
1999; Walters & Gardner, 1984). Unlike
Spearman and other advocates of general intelligence,
Although Gardner's (1983) multiple intelligences are individual-differences constructs, in which some people, or some diagnostic groups, are assumed to have more of these abilities than others, Gardner does not rely on the traditional psychometric procedures -- scale construction, factor analysis, multitrait-multimethod matrices, external validity coefficients, etc. -- for documenting individual differences. Rather, his preferred method is a somewhat impressionistic analysis based on a convergence of signs provided by eight different lines of evidence -- chief among which are isolation by brain damage, such that one form of intelligence can be selectively impaired, leaving other forms relatively unimpaired; and exceptional cases, individuals who possess extraordinary levels of ability in one domain, against a background of normal or even impaired abilities in other domains (alternatively, a person may show extraordinarily low levels of ability in one domain, against a background of normal or exceptionally high levels of ability in others).
Although Gardner's (1983) multiple intelligences are individual-differences constructs, in which some people, or some groups, are assumed to have more of these abilities than others, Gardner does not rely on the traditional psychometric procedures -- scale construction, factor analysis, multitrait-multimethod matrices, external validity coefficients, etc. -- for documenting individual differences. Rather, his preferred method is a somewhat impressionistic analysis based on a convergence of signs provided by eight different lines of evidence.
Chief among these signs are isolation by brain damage, such that one form of intelligence can be selectively impaired, leaving other forms relatively unimpaired; and exceptional cases, individuals who possess extraordinary levels of ability in one domain, against a background of normal or even impaired abilities in other domains (alternatively, a person may show extraordinarily low levels of ability in one domain, against a background of normal or exceptionally high levels of ability in others). So, for example, Gardner (1983) argues from neurological case studies that damage to the prefrontal lobes of the cerebral cortex can selectively impair personal and social intelligence, leaving other abilities intact. The classic case of Phineas Gage may serve as an example (Macmillan, 1986). On the other hand, Luria's (1972) case of Zazetsky,"the man with a shattered world", sustained damage in the occipital and parietal lobes which severely impaired most of his intellectual capacities, but left his personal and social abilities relatively intact. Gardner also notes that while both Down syndrome and Alzheimer's disease have severe cognitive consequences but little impact on the person's ability to get along with other people, Pick's disease spares at least some cognitive abilities while severely impairing the person's ability to interact with others. In related work, Taylor and Cadet (1989) have proposed that three different brain systems provide the neurological substrate of social intelligence: a balanced or integrated cortical subsystem which relies on long-term memory to make complex social judgments; a frontal-dominant subsystem which organizes and generates social behaviors; and a limbic-dominant subsystem which rapidly produces emotional responses to events. However, it should be noted that, with the exception of emotion (for an authoritative summary, see LeDoux, 1996; see also Kihlstrom, Mulvaney, Tobias, & Tobis, 1998), research on the neurological underpinnings of social cognition and behavior is highly impressionistic and speculative (for a review of neuropsychological approaches to social cognition and social intelligence, see Klein & Kihlstrom, 1998).
With respect to exceptional individuals, Gardner offers Sigmund Freud and Marcel Proust as"prodigies" in the domain of intrapersonal intelligence, and Mahatma Gandhi and Lyndon Johnson as their counterparts in the domain of interpersonal intelligence. Each of these individuals, Gardner claims, displayed high levels of personal and social intelligence against a background of more"normal" abilities in other domains. On the negative side, Gardner notes that infantile autism (Kanner's syndrome, Williams' syndrome, etc.) severely impairs the individual's ability to understand other people and navigate the social world.
With respect to social intelligence, of course, the core operations are those which form the core of research on social cognition: person perception and impression formation, causal attribution, person memory, social categorization, impression management, and the like. The social cognition literature offers numerous paradigms for studying these operations, of course, and sometimes these experimental procedures have been translated into techniques for the analysis of individual differences (e.g., Kihlstrom & Nasby, 1981; Nasby & Kihlstrom, 1985). For example, Kaess and Witryol (1955) studied memory for names and faces; Sechrest and Jackson (1961) examined individual differences in the ability to predict other people's behavior in various kinds of situations; and Sternberg and his colleagues (Barnes & Sternberg, 1989; Sternberg & Smith, 1985) have assessed individual differences in the ability to decode nonverbal communications.
Whether the core operations involved in social cognition differ qualitatively from those involved in nonsocial cognition is, however, an open question. While perceiving emotion in a face may appear to differ qualitatively from mentally rotating an image of the letter R, a working assumption in most social cognition research is that the underlying mental processes are the same as those deployed in nonsocial cognition. Thus, for example, Cantor and Mischel (1979)'s research on prototypes in person perception was intended as a fairly direct translation of Rosch's (1978) pioneering work on fuzzy-set approaches to nonsocial categories. And while it is quite plausible to suggest that the perception of faces, those most social of stimuli, follows special rules and is mediated by a special brain area (e.g., Farah, 1996), recent experimental and neuroimaging evidence indicates that face recognition is simply an instance of a broader expertise for identifying objects at subordinate levels of categorization (Gauthier, 1998).
One potentially important difference between the social and nonsocial domains, of course, is that in social cognition the object (i.e., the person) represented in the observer's mind is intelligent and conscious. Thus, the person being perceived may try to control the impression formed by the perceiver through a variety of impression-management strategies (Goffman, 1959; Jones and Pittman, 1982). To complicate things further, the perceiver may well be aware of the possibility of strategic self-presentation, and thus adjust his or her perceptions accordingly, while the person being perceived may modulate his or her impression-management activities so as to minimize these corrections. Such interaction rituals (Goffman, 1967) are not likely to occur in nonsocial perception and cognition.
In addition to experimental and
psychometric evidence, Gardner (1983) also assumes that
qualitatively different forms of intelligence will show
developmental histories, in terms of different developmental
trajectories, from infancy through adolescence and
adulthood to old age -- and,
perhaps, different evolutionary pathways as well. Finally,
In addition to experimental and psychometric evidence, Gardner (1983) also assumes that qualitatively different forms of intelligence will show distinctive developmental histories. From an ontogenetic point of view, then, the hypothesis is that the acquisition and mastery of competencies in the social domain follows a different developmental trajectory, from infancy through adolescence and adulthood to old age, than other abilities. Similarly, from a phylogenetic point of view, the hypothesis would be that personal and interpersonal abilities trace different evolutionary pathways as well. Thus, Gardner (1983) cites Gallup's (1970, 1998; Gallup, Marino, & Eddy, 1997) finding that humans and chimpanzees, but not other primates (and not other mammals) pass the mirror-image test of self-recognition.
Finally, Gardner argues that each form of intelligence is encoded in a unique symbol system by which the ability in question can be manipulated and transmitted by a culture. For some of his proposed intelligences, the existence of the symbol system is fairly obvious: written language, mathematical symbols, and musical notation are clear examples. As evidence suggestive of special personal symbol systems, Gardner cites Geertz's (1975) ethnographic work in Java, Bali, and Morocco, which revealed considerable cultural diversity in the means by which people maintain a sense of self and the rules which govern their social relations -- personal and social intelligence which is acquired through socialization. Certainly, the English language contains a large vocabulary of words -- 17,953 by one count (Allport & Odbert, 1937) -- which can represent people's cognitive, emotional, and motivational states, behavioral dispositions, and other psychosocial characteristics. And within Western culture, structures like the classic fourfold classification of temperament (melancholic, phlegmatic, choleric, and sanguine; Kant, 1798/1978) and the Big Five personality dimensions (neuroticism, extraversion, agreeableness, conscientiousness, and openness to experience; John, 1990) are commonly employed to capture and communicate the gist of another person's personality.
Gardner did not offer any new tests of social intelligence, nor did he provide compelling evidence that his multiple intelligences were really qualitatively different from each other. But in the context of a growing interest in cognitive neuroscience, and a growing inclination among psychologists to take neurobiological data as the gold standard of what is psychologically"real", claims for a neuropsychological dissociation between interpersonal intelligence and other forms of intelligence (e.g., that damage to the prefrontal cortex can selectively impair intrapersonal and interpersonal intelligence while leaving other abilities intact) gave new life to the notion that social intelligence can be distinguished from linguistic, logical-mathematical, and spatial intelligence.
The idea of
social intelligence also got a boost from arguments in
favor of individual differences in emotional
intelligence, defined as"the ability to monitor
one's own and others' feelings, to discriminate among
them, and to use this information to guide one's thinking and action" (Salovey & Mayer, 1990, p.
189; see also Mayer, Roberts, & Barsade, 2008;
Mayer, Salovey, & Caruso, 2008; Salovey &
Grewal, 2005). Emotional
intelligence subsumes four component abilities: the
ability to perceive emotions in oneself and others; to
use emotions in the service of thinking and problem
solving; to understand emotions and the relations among
them; and to manage emotions in oneself and others. Emotional
intelligence and social intelligence are not the same
thing: there is nothing particularly social about snake
phobia, and there are many aspects of social cognition
where emotion plays little or no role. But, as the
listing of the component abilities indicates, emotion is
frequently evoked in a social context, so emotional
intelligence and social intelligence do share a sort of
family resemblance. Together with David Caruso, Mayer and
Salovey hvae published the Mayer-Salovey-Caruso Test of Emotional Intelligence.
(Salovey is now president of Yale University.)
The idea of emotional intelligence was popularized by Daniel Goleman in a series of books (e.g., Goleman, 1995), and quickly caught on in both academic and applied psychology. A search of the PsycInfo database reveals that, before 1990, only 3 items had the phrase"emotional intelligence" in their title or abstract, compared to 253 for"social intelligence". For the decade 1990-1999, emotional intelligence had 77 such items, compared to 97 for social intelligence. But for the decade 2000-2009, emotional intelligence garnered 1,838 items (this is not a misprint), compared to 289 for social intelligence. Whereas Thorndike (1920) postulated social intelligence as the third member of a triad of intelligences, along with mechanical and abstract intelligence, it seems possible that, as suggested by Mayer, "Emotional intelligence could be... the replacement member of the triumvirate where social intelligence failed" (quoted in Goleman, 2006, p. 330).
This explosion of interest in emotional intelligence probably has much to do with what might be called the"affective counterrevolution" in psychology -- the feeling that, since the cognitive revolution of the 1950s and 1960s, psychology had gone overboard in emphasizing epistemology, and needed to pay more attention to feelings and desires. Certainly there is little reason to think that emotional intelligence is a clearer concept than social intelligence, or any easier to measure (Murphy, 2006). Whatever the reason, the upsurge of interest in emotional intelligence seems to have carried social intelligence along with it, so that we can look forward to a revival of research interest in this topic.
All the more so, perhaps, now that
Goleman (Goleman, 2006) has done for social intelligence
what he did earlier for emotional intelligence. The premise of
Goleman's book is that
rewarding social relationships are the key to happiness
and health (roughly half of the book reviews research on
the social psychology of health), and that the key to
rewarding social relationships is social intelligence. Therefore, we
need new tools for the assessment of individual
differences in social intelligence, but -- more to the point -- we need educational programs
that will enable people to learn how to increase their
emotional intelligence, and therefore to be happier and
healthier, as well as wiser. Whereas
Goleman provocatively characterizes
previous work on social intelligence as a"scientific
backwater" (p. 330) in need of
total rethinking. Taking
a key from Gardner (Gardner, 1999; Walters &
Gardner, 1984), who relied more on neuropsychology than
on psychometrics, as well as the doctrine of modularity
as it has developed in contemporary cognitive and social
neuroscience (Fodor, 1983; Kihlstrom, 2009b), Goleman
hypothesizes that social intelligence is mediated by an
extensive network of neural modules, each dedicated to a
particular aspect of social interaction. But more than
neuroscientific findings have the potential to
reinvigorate the social and behavioral sciences", just as "the
basic assumptions of economics...
have been challenged by the emerging 'neuroeconomics', which studies the brain
during decision-making" (p.
this prediction will come true. At the same
time, however, it is a matter of historical fact that
the real revolution in economics -- the advances that
garnered the Nobel Prizes -- flowed from observational
field studies (e.g., Simon, 1947; Simon, 1955) and
paper-and-pencil questionnaires (Kahneman, 2003;
Tversky & Kahneman, 1974). But even if
cognitive and social neuroscience do not prove to be
the saviors of social intelligence (or of cognitive
and social psychology in general), Goleman's list of social-intelligence
abilities is as good a place as any to start
developing a new generation of instruments for
assessing social intelligence.
as defined in standard dictionaries, has two rather
In its most familiar meaning, intelligence has to
do with the individual's
ability to learn and reason. It is this
meaning which underlies common psychometric notions such
testing, the intelligence
quotient, and the like. As originally
coined by E.L. Thorndike (1920), and pursued in the
studies reviewed so far, social intelligence
referred to the person's
ability to understand and manage other people, and to
engage in adaptive social interactions. In its less
common meaning, intelligence has to do with a body of
information and knowledge.
This second meaning is implicated in the titles
of certain government organizations, such as the Central
Intelligence Agency in the
On the other hand,
Cantor and Kihlstrom have offered an alternative"knowledge
view" of social intelligence
which refers simply to the individual's fund of knowledge about the
social world (Cantor & Kihlstrom, 1987, 1989;
Kihlstrom & Cantor, 1989, 2000, 2010; see also
Cantor & Kihlstrom, 1987, 1989; Cantor &
Fleeson, 1994; Cantor & Harlow, 1994; Kihlstrom
& Cantor, 1989; see also Cantor & Kihlstrom,
1982; Cantor & Zirkel, 1990; Snyder & Cantor,
1998). In contrast to the psychometric
approaches reviewed above, the social intelligence
view of personality does not conceptualize
social intelligence as a trait, or group of traits, on
which individuals can be compared and ranked on a
dimension from low to high. Rather, the
social-intelligence view of personality begins with
the assumption that social behavior is intelligent
-- that it is mediated by cognitive processes of
perception, memory, reasoning, and problem-solving,
rather than being mediated by innate reflexes,
conditioned responses, evolved genetic programs, and
the like. Accordingly, the social intelligence view
construes individual differences in social behavior --
the public manifestations of personality -- to be the
product of individual differences in the knowledge
which individuals bring to bear on their social
interactions. Differences in social knowledge cause
differences in social behavior, but it does not make
sense to construct measures of social IQ. The
important variable is not how much social
intelligence the person has, but rather what
social intelligence he or she possesses.
The social intelligence view of personality has its origins in the social-cognitive tradition of personality theory, in which construal and reasoning processes are central to issues of social adaptation. Thus, Kelly (1955) characterized people as naive scientists generating hypotheses about future interpersonal events based on a set of personal constructs concerning self, others, and the world at large. These constructs were idiographic with respect to both content and organization. Individuals might be ranked in terms of the complexity of their personal construct systems, but the important issue for Kelly was knowing what the individual's personal constructs were. Beyond complexity, the idiosyncratic nature of personal construct systems precluded much nomothetic comparison.
While Kelly's theory was somewhat iconoclastic, similar developments occurred in the evolution of social learning theories of personality. The initial formulation of social learning theory (Miller and Dollard, 1941), a combination of Freudian psychoanalysis and Hullian learning theory, held that personality was largely learned behavior, and that understanding personality required understanding the social conditions under which it was acquired. However, the slow rise of cognitive theories of learning (e.g., Tolman, 1932) soon lent a cognitive flavor to social learning theory itself. Thus, habit and drive played little role in Rotter's (1954) cognitive social learning theory. In contrast to earlier behaviorist conceptions of organismal responses to environmental stimuli controlled by objective contingencies of reinforcement (e.g., Skinner, 1953; Staats & Staats, 1963), Rotter argued that the individual's behavior reflected choices which followed from their goals in a particular situation, and their expectations of the outcomes of their behavior. Similarly, Bandura (Bandura & Walters, 1963; Bandura, 1973) argued for the acquisition of social knowledge through precept and example rather than the direct experience of rewards and punishment, and later (1986) distinguished between the outcome expectancies emphasized by Rotter (1954) and expectancies of self-efficacy -- the individual's judgment of belief concerning his or her ability to carry out the actions required to achieve control over the events in a situation. Self-efficacy provides the cognitive basis for motivation, but it should be understood that judgments of self-efficacy are highly context-specific. Although Rotter (1966) proposed an individual-difference measure of internal vs. external locus of control, it would never occur to Bandura to propose a nomothetic instrument for measuring individual differences in generalized self-efficacy expectations. The important consideration is not whether an individual is relatively high or low in self-perceptions of competence, but rather whether the person feels competent to perform a particular behavior in some particular situation.
The immediate predecessor to the social-intelligence view of personality is Mischel's (1968, 1973) cognitive social-learning reconceptualization of personality. Although sometimes couched in behaviorist language, Mischel's (1968) provocative critique of the trait approach to personality was explicitly cognitive in nature:"[O]ne must know the... meaning that the stimulus has acquired for the subject. Assessing the acquired meaning of stimuli is the core of social behavior assessment" (p. 190). Thus, understanding individual differences in social behavior requires understanding individual differences in the meaning given to behavior, its outcome, and the situation in which it takes place.
This emphasis on the subjective meaning of the situation marked Mischel's early theory as cognitive in nature. Since that time, Mischel (1973) has broadened his conceptualization of personality to include a wide variety of different constructs, some derived from the earlier work of Kelly, Rotter, and Bandura, and others reflecting the importation into personality theory of concepts originating in the laboratory study of human cognitive processes. All are construed as modifiable individual differences, products of cognitive development and social learning, that determine how features of the situation will be perceived and interpreted. Thus, they contribute to the construction of the meaning of the stimulus situation -- in other words, to the cognitive construction of the situation itself -- to which the person ultimately responds.
From Mischel's (1973) point of view, the most important product of cognitive development and social learning is the individual's repertoire of cognitive and behavioral construction competencies -- the ability to engage in a wide variety of skilled, adaptive behaviors, including both overt action and covert mental activities. These construction competencies are as close as Mischel gets to the psychometric notion of social (or, for that matter, nonsocial) notion of intelligence.
The importance of perception and
interpretation of events in Mischel's
system calls for a second set of person variables,
having to do with encoding strategies governing
selective attention and personal constructs -- Kelly-like
categories which filter people's
perceptions, memories, and expectations. Then, of
course, following Rotter and Bandura, Mischel also
stresses the role of stimulus-outcome,
behavior-outcome, and self-efficacy expectancies
concerning the outcomes of environmental events and
personal behaviors, as well as self-efficacy
expectancies. Also in line with Rotter's theory, Mischel notes that
behavior will be governed by the subjective values
associated with various outcomes. A final set of
relevant variables consists of self-regulatory
systems and plans, self-imposed goals and
consequences which govern behavior in the absence (or
in spite) of social monitors and external constraints.
From a cognitive point of view, Mischel's"cognitive-social learning person variables" all represent the person's knowledge and expertise -- intelligence -- concerning him- or herself and the surrounding social world. Following Winograd (1975) and Anderson (1976), this social intelligence (Cantor & Kihlstrom, 1987) is classified into two broad categories: declarative knowledge, consisting of abstract concepts and specific memories, and procedural knowledge, consisting of the rules, skills, and strategies by which the person manipulates and transforms declarative knowledge, and translates knowledge into action. The individual's fund of declarative knowledge, in turn, can be broken down further into context-free semantic memory about the world in general and episodic memory for the events and experiences, each associated with a unique spatiotemporal context, which make up the person's autobiographical record (Tulving, 1983). Similarly, procedural knowledge can be subclassified in terms of cognitive and motor skills. These concepts, personal memories, interpretive rules, and action plans are the cognitive structures of personality. Together, they constitute the expertise which guides an individual's approach to solving the problems of social life.
The cognitive architecture of social intelligence will be familiar from the literature on social cognition (for overviews, see Cantor & Kihlstrom, 1982; Fiske & Taylor, 1991; Kihlstrom & Hastie, 1997) -- a literature which, interestingly, had its beginnings in early psychometric efforts to measure individual differences in social intelligence. Thus, for Vernon (1933) one of the characteristics of a socially intelligent person was that he or she was a good judge of personality -- a proposition which naturally led to inquires into how people form impressions of personality, (Asch, 1946) or engage in person perception (Bruner & Tagiuri, 1954), as well as to the implicit theories of personality (Bruner & Tagiuri, 1954; Cronbach, 1955) which lies at the base of such impressions and perceptions. Specifically, Cronbach argued that one's implicit theory of personality consisted of his or her knowledge of"the generalized Other" (1955, p. xx): a mental list of the important dimensions of personality, and estimates of the mean and variance of each dimension within the population, as well as estimate of the covariances among the several dimensions. Cronbach argued that this intuitive knowledge might be widely shared, and acquired as a consequence of socialization and acculturation processes; but he also assumed that there would be individual and cultural differences in this knowledge, leading to individual and group differences in social behavior. Studies of impression formation, implicit personality theory, and, later, causal attributions (e.g., Kelley, 1967), social categories (Cantor & Mischel, 1979; Cantor, Mischel, & Schwartz, 1982b), and scripts (Schank & Abelson, 1977), and person memories (Hastie, Ostrom, Ebbesen, Wyer, Hamilton, & Carlston, 1980) provided the foundation for the social-intelligence analysis of personality structures and processes.
Following Kelly (1955) and Mischel (1973), Cantor and Kihlstrom (1987) accorded social concepts a central status as cognitive structures of personality. If the purpose of perception is action, and if every act of perception is an act of categorization (Bruner, 1957), the particular categories which organize the people's perception of themselves, others, interpersonal behavior, and the social world in which behavior takes place assume paramount importance in a cognitive analysis of personality. Some of these concepts concern the world of other people and the places we encounter them: knowledge of personality types (e.g., achievers and altruists; Cantor & Mischel, 1979) and social groups (e.g., women and WASPS; Hamilton, 1981), and situations (e.g., blind dates and job interviews; Cantor, Mischel, & Schwartz, 1982a). Other concepts concern the personal world: knowledge of the kind of person we are, both in general and in particular classes of situations (e.g., an achiever at work but an altruist at home; Kihlstrom & Klein, 1994; Kihlstrom, Marchese, & Klein, 1996), and our theories of how we got that way (e.g., an adult child of alcoholics or a survivor of child sexual abuse; Ross, 1989). Based on studies of categorization in nonsocial domains (e.g., Rosch, 1978; Ross & Spalding, 1994), social concepts may be viewed as structured as fuzzy sets around summary prototypes, perhaps along with representative exemplars which epitomize the category, and related to each other through tangled hierarchies reflecting conceptual relations. Some of these conceptual relations may be universal, and others may be highly consensual within the individual's culture; but, as Kelly (1955) argued, some may be quite idiosyncratic. Regardless of whether they are shared with others, the individual's conceptual knowledge about the social world forms a major portion of his or her declarative social intelligence.
Another important set of declarative social knowledge structures represents the individual's autobiographical memory (Conway, 1990; Rubin, 1996; Thompson, 1996, 1998). In the context of social intelligence, autobiographical memory includes a narrative of the person's own actions and experiences, but it also includes what he or she has learned through experience about the actions and experiences of specific other people (Hastie et al., 1980), and the events that have transpired in particular situations. While social concepts comprise more or less abstract and context-free semantic memory, autobiographical memory is episodic memory -- each piece of the narrative is tied to a specific location in space and time (Tulving, 1983). In addition, every piece of conscious autobiographical memory is linked to a mental representation of the self as the agent or patient of some action, or the stimulus or experiencer of some state (Kihlstrom, 1997). As part of this connection to the self, each fragment of autobiographical memory is, at least in principle also connected to knowledge about the person's emotional and motivational states at the time of the event in question. Thus, autobiographical memory is rich in content and complicated in structure -- so rich and complicated that it is no wonder that most cognitive psychologists fall back on laboratory tasks involving memory for words and pictures.
On the procedural side, a substantial portion of the social intelligence repertoire consists of interpretive rules for making sense of social experience: for inducing social categories and deducing category membership, making attributions of causality, inferring other people's behavioral dispositions and emotional states, forming judgments of likability and responsibility, resolving cognitive dissonance, encoding and retrieving memories of our own and other people's behavior, predicting future events, and testing hypotheses about our social judgments. Some of these procedures are algorithmic in nature, while others are heuristic shortcuts (Nisbett & Ross, 1980). Some are enacted deliberately, while others are enacted automatically, without much attention and cognitive effort on our part (for summaries, see Bargh, 1994; Kihlstrom, 1987, 1996a, 1996b; Wegner & Bargh, 1998). But they are all part of the procedural repertoire of social intelligence.
Given this summary, it should be
clear that, from the point of view of the social
intelligence theory of personality, the assessment of
social intelligence has quite a different character
than it does from the psychometric point of view. From
a psychometric point of view, the questions posed have
answers that are right or wrong: Are smart people also
friendly" How do you know
when a person is happy or sad? Is it proper to laugh
at a funeral? In this way, it is possible, at least in
principle, to evaluate the accuracy of the person
social knowledge, and the effectiveness of his or her
social behaviors. However, as noted at the outset, the
social intelligence approach to personality abjures
such rankings of people. Rather than asking how
socially intelligent a person is, compared to
some norm, the social intelligence view of personality
asks what social intelligence a person has,
which he or she can use to guide his or her
interpersonal behavior. In fact, the social
intelligence approach to personality is less
interested in assessing the individual's repertoire of social
intelligence, than in seeking to understand the
general cognitive structures and processes out of
which individuality is constructed, how these develop
over the life course of the individual, and how they
play a role in ongoing social interactions. For this
reason, Cantor and Kihlstrom (1987, 1989; Kihlstrom
& Cantor, 1989) have not proposed any
individual-difference measures by which the person's social intelligence can be
Although the social intelligence view of personality diverges from the psychometric approach to social intelligence on the matter of assessment, it agrees with some contemporary psychometric views that intelligence is context-specific. Thus, in Sternberg's (1985, 1988) triarchic theory, social intelligence is part of a larger repertoire of knowledge by which the person attempts to solve the practical problems encountered in the physical and social world. According to Cantor and Kihlstrom (1987), social intelligence is specifically geared to solving the problems of social life, and in particular managing the life tasks, current concerns (Klinger 1977) or personal projects (Little, 1989) which the person selects for him- or herself, or which other people impose on him or her from outside. Put another way, one's social intelligence cannot be evaluated in the abstract, but only with respect to the domains and contexts in which it is exhibited and the life tasks it is designed to serve. And even in this case,"adequacy" cannot be judged from the viewpoint of the external observer, but rather from the point of view of the subject whose life tasks are in play.
Life tasks provide an integrative unit of analysis for the analysis of the interaction between the person and the situation. They may be explicit or implicit, abstract or circumscribed, universal or unique, enduring or stage-specific, rare or commonplace, ill-defined or well-defined problems. Whatever their features, they give meaning to the individual's life, and serve to organize his or her daily activities. They are defined from the subjective point of view of the individual: they are the tasks which the person perceives him- or herself as"working on and devoting energy to solving during a specified period in life (Cantor & Kihlstrom, 1987, p. 168). First and foremost, life tasks are articulated by the individual as self-relevant, time-consuming, and meaningful. They provide a kind of organizing scheme for the individual's activities, and they are embedded in the individual's ongoing daily life. And they are responsive to the demands, structure, and constraints of the social environment in which the person lives. Life tasks are imposed on people, and the ways in which they are approached may be constrained by sociocultural factors. However, unlike the stage-structured views of Erikson (1950) and his popularizers (e.g., Levinson, 1978; Sheehy, 1976), the social-intelligence view of personality does not propose that everyone at a particular age is engaged in the same sorts of life tasks. Instead, periods of transition, where the person is entering into new institutions, are precisely those times where individual differences in life tasks become most apparent.
For example, Cantor and her associates have chosen the transition from high school to college as a particularly informative period to investigate life tasks (Cantor, Acker, and Cook-Flanagan, 1992; Cantor & Fleeson, 1991, 1994; Cantor & Harlow, 1994; Cantor & Langston, 1989; Cantor & Malley, 1991; Cantor, Norem, Langston, Zirkel, Fleeson, & Cook-Flanagan, 1991; Cantor, Norem, Niedenthal, Langston, & Brower, 1987; Zirkel & Cantor, 1990). Freshman year is more than just convenient for academic researchers to study: The transition from high school to college and adulthood is a critical developmental milestone, where many individuals leave home for the first time to establish various independent habits and lifestyles. And although the decision to attend college may have been made for them (or may not have been a decision at all, but just a fact of life), students still have a great deal of leeway to decide for themselves that they are going to do with the opportunity -- what life tasks will occupy them for the next four years. Accordingly, when college students are asked to list their life tasks, they list social life tasks (e.g., making friends or being on my own) as often as they list academic ones (e.g., getting good grades or carving a future direction). And while the majority of students' life tasks could be slotted into a relatively small number of common categories, their individual construals of these tasks were quite unique, and led to equally unique strategies for action.
The intelligent nature of life-task pursuit is clearly illustrated by the strategies deployed in its service. People often begin to comprehend the problem at hand by simulating a set of plausible outcomes, relating them to previous experiences stored in autobiographical memory. They also formulate specific plans for action, and monitor their progress toward the goal, taking special note of environmental factors which stand in the way, and determining whether the actual outcome meets their original expectations. Much of the cognitive activity in life-task problem solving involves forming causal attributions about outcomes, and in surveying autobiographical memory for hints about how things might have gone differently. Particularly compelling evidence of the intelligent nature of life task pursuit comes when, inevitably, plans go awry or some unforeseen event frustrates progress. Then, the person will map out a new path toward the goal, or even choose a new goal compatible with a superordinate life task. Intelligence frees us from reflex, tropism, and instinct, in social life as in nonsocial domains.
It is possible that the concept of social intelligence has outlived its usefulness, and will be supplanted by emotional intelligence. Alternatively, it is possible that neuroscientific analyses will give new life to the study of social intelligence, as they promise to do in other areas of psychology. On the other hand, perhaps we should abandon the"ability" model of social intelligence completely, along with its psychometric emphasis on developing instruments for the measuring of individual differences in social competencies of various sorts -- tests intended to rank people, and on which some people must score high and others must score low. Instead of focusing on how people compare, perhaps we should focus on what people know, and how they bring their social intelligence to bear on their interactions with other people, on the tasks life has set for them, and on the tasks they have set for themselves. In this way, we would honor the primary idea of the cognitive view of social interaction, which is that interpersonal behavior is intelligent, based on what the individual knows and believes -- no matter how smart or stupid it may appear to be to other people.