Click here for general information about exams in this course.
Link to the slides used in the in-class
The Final Examination is structured as a two-hour exam, which you will have the regulation three hours to complete. The first hour (maybe a little more) will be noncumulative, covering new material (lectures and readings) since the Midterm Examination. The second hour (maybe a little less) will be cumulative, covering the course as a whole.
There will be a review session in our regular classroom and time during RRR Week This narrative review takes the place of a oral review by me, but at the review session I will take questions on the lectures and readings. Questions can also be posted to the Comments and Queries forum on the course website. Questions posted by noon on the Friday of RRR Week will be answered by the end of the business day.
Because material from the Midterm Examination will be covered
in the cumulative portion of the Final Examination, this
review incorporates by reference the narrative review for the
Social cognition tries to understand how we process information about other people (and the situations in which we encounter them), but it also tries to understand how we process information about ourselves.
You should understand William James' distinctions among the material self, the social self, and the spiritual self.
Defining the self has been problematic throughout the history of psychology (as illustrated by Allport's "puzzling problem", but it's possible, as a kind of heuristic approach, to define "the self" simply as one's mental representation of oneself. Just as we have mental representations of other people's traits, attitudes, and the like, so we have a mental representation of our own traits, attitudes, and the like. And this mental representation is the self.
Cognitive psychology and cognitive science distinguishes between two forms of mental representation: perception-based representations of the physical appearance of objects and events, and meaning-based representations of their meaning. As a first pass, we can think of the self-image as a perception-based mental representation of the self, and the self-concept as a meaning-based representation.
If the self-concept is a concept, then we can think of the
self-concept as organized like other concepts: as a proper
set, as a fuzzy set represented by a prototype, as a
collection of exemplars, and the like. So you should know
something about how these views of conceptual structure could
be applied to the self-concept. For example, the symbolic
interactionsts promoted the concept of the "looking-glass
self" -- that each of us has as many "selves" as we have
significant others in the environment. These context-specific
self-concepts could be united by a superordinate
"prototypical" self. Or they could simply be represented as a
collection of exemplars, with no unifying prototype.
Multiple personality disorder, in genuine cases, may be a kind of radical instantiation of the exemplar view.
If the self-image is an image, then we can think of the
self-image as resembling the images that we have of other
people. Mita's study of photo preferences, which showed that
we prefer straight photos of our friends, but mirror-reversed
photos of ourselves, strongly suggests that we do indeed carry
something like pictures in our heads depicting what we, and
they, look like. Research on the body image in men and women,
including eating-disordered women, also suggests that we have
not only an image of our actual body, but also an image of our
ideal body. In eating-disordered women, the discrepancy
between actual and ideal body-image is particularly
great. You should have some
understanding of the various methods by which the self-image can be assessed, whether by
questionnaire, the paper-and-pencil Body Image Assessment,
or computer morphine software and other
If, as these lectures assume, the self
is just another person, then we can also think of the mental
representation of the self as a memory structure, similar to
the memory structures discussed in the earlier lectures on
person memory, and encompassing both episodic
(autobiographical) and semantic (trait) knowledge about the
self. As was the case with (other) person memory, priming
studies support an independent encoding of episodic and
semantic self-knowledge. Retrieval of trait-related
autobiographical memories primes the retrieval of other autobiographical memories, but not retrieval of
trait descriptions; and retrieval of trait descriptions
primes retrieval of other trait descriptions, but not the
retrieval of trait-related autobiographical memories.
In addition, the study of DB suggests that amnesic patients may not be able to project themselves into the future, in addition to their inability to remember their past -- but I didn't get lecture on this, so you're not responsible for knowing it!.
But there's more to autobiographical
memory than a "grocery list" of unconnected episodic
By representing one's character, reflecting the major epochs in one's own life, and by explaining why some events happened as they did, autobiographical memory serves not just as a record of the self, but also a theory of oneself -- how one came to be the person one is.
How different are mental representations
of oneself, compared to mental
representations of another person. We probably
have more self-knowledge than we possess about
others (though there are some people who seem
remarkably self-unaware). But we also possess
direct introspective access to our mental states --
our thoughts, feelings, and desires as various
events occurred; for other people, we can know these
things only by inference.
F&T's Chapter 5 covers some of this same ground, but in my view doesn't pay enough attention to the problem of the representation of self-knowledge. You should recognize such concepts as the working self-concept, the relational self, and self-schemas.
The material on t"the neural basis of self-views"
illustrates the problem of making social-neuroscientific
inferences: proper interpretation of brain imaging
results depends on proper analysis of the task that
subjects perform. So, for
example, it sometimes appears that there are distinct
regions of the temporal lobe activated when making
judgments of the self; but the same areas are
activated when making judgments about others whom the
subject knows well.
Of particular importance is the work on cultural differences
in self-perception (i.e., independence and interdependence) and self-enhancement.
The material on self-regulation is important, especially self-discrepancy theory and self-guides, self-efficacy, and self-focus and the cybernetic theory of self-regulation. The neural bases of self-regulation appear to be located in the pre-frontal cortex -- but, again, it's not clear that any of these areas perform functions that are specific to self-regulation. The motivational consequences of self-cognition are also important, including self-affirmation theory and self-evaluation maintenance theory, and social projection (note, too, the cultural differences in self-enhancement).
There is one qualitative difference between self-knowledge
and knowledge of other people: we have conscious awareness of
our own behaviors, feelings, and attitudes -- but can know
these features of other people only by inference from their
behavior. But, not to pick a point, the literature on
self-referencing again illustrates the importance of
task-analysis: exactly the same areas of prefrontal cortex are
activated when subjects make judgments
about other people.
F&T subtitled their textbook with reference to brains and culture, but they don't actually have much about either topic, which is why I asked you to read two papers that got this relatively new field going. The K&K paper focuses more on neuropsychology, while the O&L paper is more explicitly neuroscientific in orientation.
Social neuroscience is a variation on cognitive
neuroscience. Its primary task is to
identify the neural substrates of social cognition and
social behavior -- a task made much easier these days by the
availability of technologies like fMRI. But some cognitive and social neuroscientists
believe that neuroscience can do more than that -- that it
can resolve questions at the
psychological level of analysis. I call this the
rhetoric of constraint, and I gave several examples of
what I called "the rhetoric of constraint", just to show
that this is not a straw man. And I also argued
against it, on both conceptual and historical
grounds. A valid psychological theory is a
prerequisite for interpreting brain-imaging results;
and, anyway, there's no good example of constraint in the whole history of neuroscience.
You don't have to agree with me about the rhetoric of
constraint, but you do need to know what it
I also emphasized the Doctrine of Modularity, which is the governing idea in modern cognitive neuroscience, on which social neuroscience is modeled. You should understand the basic idea of modularity, and how it was anticipated by 19th-century phrenology. And you should know what happened to Phineas Gage, and that his case is a little more complicated than is usually presented in textbooks.
You should have some feeling for the various proposals for
social-cognitive modules, such as Jackendoff's and Gardner's. And you should have some sense, from
the Lieberman paper, about the sorts of social-cognitive
modules which have been identified so far. Don't
memorize them all! But do have a
sense that controlled and automatic, externally
and internally oriented, social-cognitive functions do seem to be
located in distinct areas of the brain.
Setting aside the debate about the rhetoric of constraint, it's clear that inferences about the neural bases of various social-cognitive modules depend on valid analyses of the tasks that subjects perform while in fMRI machines (or whatever). I went through one example in considerable detail: the controversy over the "fusiform face area", which Kanwisher and others have located in the a region joining the occipital and temporal lobes. The problem is that this same area also appears to be activated when expert subjects are engaged in subordinate-level categorization (i.e., naming) of non-face stimuli such as cars, birds, greebles, and snowflakes. The most recent evidence takes care of a methodological confound, which was that the "expertise" studies used relatively low-resolution scanners, while the studies ostensibly revealing the "true" fusiform face area haven't measured expertise. A recent study took care of this problem, and found that, lo and behold, areas of expertise overlap with face-selective areas, strengthening the case that the FFA isn't specialized for faces. You can disagree about the conclusion, but you should understand the issues at stake.
Note: There's a second area where this problem has reared its ugly head, which is the search for an area of the brain dedicated to self-referent thinking. I didn't get to lecture on this, so you're not responsible for it. Nor did I get to the third example, "mirror neurons", so don't worry about them, either (though they do represent a hot new emerging area in social neuroscience, and are discussed a little in F&T's Chapter 15).
The bottom line is that, although we are now in a position to begin to identify the neural substrates of various aspects of social cognition (and other social interactions), the neuroscience is only as good as the psychology which it draws on.
These not-too-technical articles, and all of the articles
assigned in addition to the text, are intended to be general
reviews. If you can understand the abstracts, of these
and the other papers, you'll be fine.
The article by Klein and Kihlstrom (1998) represents an early foray by social psychologists into neuroscience. Notice, however, that, in contrast to the next article, by Lieberman, K&K spent relatively little time discussing modularity or functional specialization. Instead, they relied on more traditional neuropsychological approaches, which used data from brain-injured subjects to shed light on normal mental processes, without being concerned with functional specialization per se. So, amnesic patients such as H.M. were studied for the insights they provided into the organization of self-knowledge in memory. And autistic children were studied for insights into the theory of mind. And prosopagnosics were studied for insights into the processes involved in face recognition. None of these insights depended on there being a particular location in the brain representing the neural substrate of self-knowledge, face-recognition, or theory-of-mind reasoning.
The review by Lieberman (2007), however, is completely focused on modularity, and identifies some 21 different areas ostensibly associated with different aspects of social cognition, from "theory of mind" to "fairness and trust processes". I would not expect you to be able to locate all of these 21 functions. However, Lieberman did suggest some more abstract localizations, such with controlled functions (the "C system") clustering in the lateral prefrontal and parietal cortex, and automatic processes (the "X system"). Similarly, internally focused processes seemed to be localized in the medial (interior) portions of the brain, while externally focused processes tended to be localized on the lateral (external) portions of the brain (nice mnemonic, that). Note, in passing, Lieberman's invocation of the rhetoric of constraint: he thinks that this is an organization of social-cognitive processes that would not be apparent without neuroscientific evidence. In this respect, he is, of course, quite wrong: for example, the distinction between automatic and controlled processes was well established by 1977, long before Posner and Raichle published Images of the Mind, introducing PET imaging to psychologists. Still, Lieberman provides a concise review of contemporary social neuroscience, which is why I assigned it.
My "Phineas Gage"
paper traces the evolution of social neuroscience, and offers some
caution about what I call the "rhetoric of constraint"
-- that is, the idea that knowledge of brain structure
constrains our theories of the mind. I'm not embarrassed to say that I incline toward a modern version of
dualism, because I think that psychology is
inherently dualistic, and
that you can study the mind without studying
its physical basis in the
brain. But you don't have to agree
with me. A more important point, as
illustrated by the problem of
self-reference and the "fusiform face
area", is that you have to get the
psychology right before the neuroscience can tell you how the
brain does it.
F&T don't have much on social-cognitive development,
either. And they don't have much on the social learning
processes by which social knowledge (declarative or
procedural) is acquired. These
lectures attempt to fill in the gap, looking at the development of social cognition from ontogenetic, phylogenetic, and
The ontogenetic approach to social-cognitive development focuses on the "cognitive starting-point", and is all wrapped up in two issues:
Most research on social-cognitive development, however,
focuses on the "cognitive endpoint", whether it be formal
operations (Piaget) or the development of a theory of mind
(Baron-Cohen and others). You should understand Piaget's
"three mountain" task as a measure of egocentrism, and its
relation to the theory of mind. And you should understand the
"false belief" task and its relation to the theory of
But that's not all. The false-belief
task has to do with intersubjectivity, but there are more
than just two levels of intersubjectivity.
As a rule, adults can handle about 4 levels of
intentionality. But while children pass the standard false-belief
task (entailing two levels of intentionality), they don't
pass a test for "second-order" theory of
mind (involving 3 levels of intentionality) until they are
9 or 10 years old. So there's
social-cognitive development beyond age 5. And
there's social-cognitive development beyond false
beliefs, too. For example, the same children who
do well on the standard false-beliefs task may not do so
well on tests of other aspects of the theory of mind,
such as the difference between real and apparent
And there's social-cognitive development before age 5, too. For
example, infants will typically pass a
nonverbal version of the theory of mind test, meaning that they have at least a
rudimentary ability to make inferences about
what other people believe.
Baron-Cohen's theory of
mindreading has provided an important framework for
development. He distinguishes
among four social-cognitive modules -- an intentionality
detector, an eye-direction detector, a shared-attention
mechanism, and a theory-of-mind mechanism -- each with its
own developmental trajectory, and each with its own
associated brain module.
insight that FB tasks can be administered nonverbally
permits us to ask questions about ToM in nonhuman primates
-- who also can't talk. Of course, the whole notion
of ToM arose from primate research. But (with the possible, partial exception of Sarah!),
chimpanzees and other primates don't
seem to pass nonverbal ToM tests that are passed by even
very young children. In Baron-Cohen's terms monkeys
appear only to have an "intentionality detector"; the great apes
appear to have an eye-direction detector as well, but
lack the shared-attention mechanism,
and no theory-of mind mechanism. On the other
hand, these tests have mostly been
performed on captive monkeys. Frans de Wall,
who works mostly with monkeys born, raised, and
living in more naturalistic environments, argues
forcibly that chimpanzees do engage in
Research on autistic children indicates that they generally lack the theory-of-mind mechanism, and lack the shared-attention mechanism as well -- though it might be said that they might perform better on nonverbal versions of the FB task. . As intuitively appealing as this conclusion is, there are problems with the "ToM" hypothesis of autism. For example, it turns out that many blind children are also late in developing a theory of mind, so this problem is not specific to autism. And it suggests that the failure to develop a theory of mind may be an effect of autism, rather than its cause.
The ontogenetic and
phylogenetic views of social-cognitive development
are connected by ToM, but they're also connected by
mirror neurons, which were "discovered" in monkeys who are observing the
actions of others. Subsequent brain-imaging
research has tentatively identified "mirror neuron
systems" (as opposed to individual mirror neurons)
in humans as well -- two MNSs, in fact: a MNS for
action in the premotor and somatosensory cortex,
and a MNS for emotion in the anterior
cingulate. It's been
suggested that mirror neurons comprise the
neural substrate of social cognition, and that
the failures of mindreading in autistic
individuals are caused by a "broken" MNS.
But these conclusions are
troubled by various problems. First,
it's not clear that, even
in monkeys, mirror neurons subserve action understanding as opposed to action selection
(for example, patients with Broca's
aphasia, resulting from damage in the
MNS, can still understand speech).
Second, initial findings suggestive
of a malfunctioning MNS in autistic
individuals have frequently -- too
frequently - -failed to
Turning to the cultural view, some anthropological and
cultural-psychology research suggests that some indigenous
people don't explain behavior in terms of
mental states -- that is, they don't seem
to have the same theory of mind that we are familiar with.
This suggests that ToM may be a
cultural product, acquired through socialization and acculturation, and not an
innate, modular cognitive faculty. But it's not
entirely a cultural product. Following
Chomsky's idea of a "Universal Grammar" (UG)
underlying language acquisition and use, Lillard
has suggested that there is a
"Universal False Belief" -- that is, no matter
what culture we've been raised in, we notice
when behavior does
not accord with the way the world really
is. The explanation of this discrepancy
may differ across cultures, but the capacity
to notice this discrepancy is, she
Gopnik and Wellman (1992) are pretty much the authors of the "theory theory" view of cognitive development, and their treatment of the child's theory of mind is very useful. Pay particular attention to their characterization (pp. 149-153 of developmental changes in children's theory of mind between ages 2-1/2 and 5, and to their evidence that children really do try to explain and predict the world of the mind (pp. 153-158). Don't be too concerned about the debate between "theory" theorists and "simulation" theorists -- that's for a more advanced course in cognitive and social development.
Tager-Flusberg discusses some problems with the very popular notion that autistic children (and adults) lack a theory of mind, or a capacity for mindreading. Her basic point is that autistic individuals have other problems too, so mindblindness is only part of the problem.
Lillard takes us away from the ontogenetic
and phylogenetic views of development, and into
the cultural view, by arguing, mostly with
anecdotal evidence from cultural
anthropologists, that there are some societies out there
which don't have the same "theory of mind" that
"folks" in the Western world
do -- which suggests
that the theory of mind is not
And, of course, there's Haddon's
novel, The Curious Case of the
Dog in the Night-Time, which is
relevant because it ostensibly portrays the mind of a boy with autism (or maybe
Asperger's Syndrome), and who, to some extent, has failed to develop a theory of mind.
The social-cognitive approach to personality begins with "social learning" theory, which itself began as a kind of "third way" between psychodynamic and trait theories of personality. Social learning theory, in turn, is an offshoot of behavioristic and neobehavioristic learning theory in general. As opposed to psychodynamic theories, such as Freud, which emphasized defenses against conflict generated by primitive drives such as sex and aggression, and trait theories, such as "The Big Five", which hold that basic personality traits give coherence, stability, consistency, and predictability to behavior, learning theories of personality assumed that personality was acquired and maintained through learning, and changed when the person learned to behave differently.
In some ways, this learning theory approach to personality is exemplified by B.F. Skinner's radical behaviorism: for Skinner, traits and drives were of no consequence. The important variables in behavior were the individual's history of reinforcement, the discriminative stimuli present in the environment, and the contingencies of reinforcement currently operating. Put another way: there was nothing inside the person that was relevant to personality; all the action was in the external environment. While Skinner just sort of tossed this idea off, Arthur and Carolyn Staats made the Skinnerian approach to learning the core of their analysis of personality, human motivation, and social interaction. The "learning" approach to personality was strengthened by the "behavior therapy" movement within clinical psychology, and when behavior therapy took a "cognitive turn" in the late 1960s, so did learning-theory approaches to personality.
Actually, social learning theory goes back further than that, to the work of Miller and Dollard, who argued that personality was a system of habits acquired through learning which took place in a social context -- hence their label of social learning theory. M&D's work was, in truth, little more than a translation of Freudian psychodynamic theory, de-sexualized, de-aggressivized (?), and de-biologized, into the terms of a learning theory proposed by Clark Hull, who argued that drives activated behavior, and that learning was reinforced by drive-reduction.
Social learning theory took a cognitive turn with the work of Rotter, who fused the drive-reduction theory of Hull with the cognitive learning theory of Berkeley's own E.C. Tolman. What made Rotter's social learning theory cognitive was his emphasis on the expectancy and value of reinforcement, as opposed to the objective contingencies of reinforcement, and whether the reinforcement was objectively positive or negative. Thus, the important individual differences lay not in habits, or in contingencies of reinforcement, but rather in the subjective definition of reinforcement and its contingencies. Thus, as opposed to the learning theories of Skinner and Hull, the learning situation was defined subjectively. Did I say that everything was subjective? That's what made it cognitive in 1954.
The next step in the evolution of cognitive social learning theory was made by Bandura, who emphasized the social aspect of social learning (whereas Rotter emphasized the learning aspect). For Bandura, some social learning occurred by virtue of the direct experience of the consequences of behavior, but most social learning was vicarious, or observational: we learn from the example of other people, or we learn by being taught by other people. Bandura also put a cognitive spin on imitation. Whereas M&D had thought of imitation as a habit acquired through learning (and reinforcement), Bandura argued that imitation was itself governed by cognitive processes of perception, attention, and memory. And whereas Skinner had emphasized the external control of behavior, Bandura stressed the importance of self-reinforcement and self-regulation.
Outside the mainstream of social learning theory was Kelly's "personal construct" theory of personality. Kelly argued that experience, thought, and action was filtered through a conceptual structure that was unique for each individual. The way we construe events will determine how we respond to them, and individual differences in behavior reflect individual differences in our personal construct systems. To the extent that people overlap in their personal construct systems, they will behave similarly in the same situation. To the extent that people understand each others' personal construct systems, they can have a productive social interaction.
Walter Mischel brought all of this together in
his cognitive-social learning reconceptualization of
personality. (Note for the historically minded: Mischel did his graduate
work at Ohio State University, where he
studied with both Rotter and Kelly; and for many years
he was on the faculty at Stanford,
where he was colleagues with Bandura.)
Like other social
learning theorists, Mischel rejects traits as the basis of
individual differences in personality. For him, the
important individual differences are: (1) cognitive-behavioral
construction competencies (basically, what you know how to
do), (2) encoding strategies (selective attention to some
things as opposed to others), (3) personal constructs (pretty
much as defined by Kelly), (4) expectancies (pretty much as
defined by Rotter and Bandura), and (5) self-regulatory
systems and plans (again, echoing Bandura).
Note for the interested (but I didn't lecture on it, so you're not responsible for it): More recently, Mischel (working chiefly with his student, Yuichi Shoda), has proposed that personality is based on a "Cognitive-Affective Processing System". Mischel distinguishes between consistency across situations (like what is expected from the Doctrine of traits) and consistency within situations (or, if you will, temporal stability in behavior within a class of situations). He argues that consistency within situations is a much more salient feature of personality. From this he concludes that behavior is a product of "If-Then" productions that take the form of "If I'm with my family, I'll be introverted; but if I'm with my friends, I'll be extraverted". S/he wouldn't be consistently extraverted (or introverted), as the Doctrine of Traits suggests; but s/he would be consistently introverted in family situations, and consistently extraverted in peer-group situations. These If-Then productions are in turn linked to Cognitive-Affective Units which encode representations of self, others, and situations; goals, expectations, beliefs, and feelings; and specific memories of people and past events. Personality, then is manifest by "behavioral signatures" in which activation of CAU leads to consistent behavior. But, in line with the cognitive interpretation of personality, everything depends on how the situation is construed.
Further Note: I also didn't get to lecture on my own "social intelligence" view of personality, which was included in some previous versions of this course, so you're not responsible for it, either. But, just for your information, social cognition offers a conception of social intelligence roughly analogous to academic intelligence -- a way of ranking people in terms of how "socially smart" they are. That's what Nancy Cantor and I call the ability view of social intelligence. We've proposed an alternative knowledge view, which abjures ranking people from socially smart to socially stupid, and simply builds on the idea that individual differences in social behavior - -which, after all, is where personality manifests itself -- are mediated by individual differences in the social knowledge, the social intelligence, that each person brings to the situation.
F&T don't have much to say about the relation of social cognition to personality, either, which is too bad, because there's lots of interesting stuff. But while the lectures have no parallel in F&T, but the final chapters in the text do present some very interesting and important material on the relation between social cognition and emotion. This material is so well presented that I decided not to lecture on it at all, and to try something quite different instead.
From Chapter 13, you should understand the vocabulary of emotion, like the two-dimensional structure of affect (which I also discussed in the lectures on social categorization, in terms of the dimensional resolution of problems with the classical fourfold typology of personality), and Ekman's basic emotions. You should also know something about the various theories of emotion.
Of these theories, of course, the most important are those with construe emotions as the product of social cognition, such as:
You should know why affective forecasting is so difficult.
And you should understand some of the problems associated with
the goal of affective neuroscience, which is to identify specific
brain areas associated with the various emotions.
Chapter 14 reverses the direction of causality, and
looks at affective influences on cognition. So you should know
something about the effects of mood on memory, judgment,
persuasion, and well-being. Note the role of the
prefrontal cortex in judgment, choice, and decision-making;
but also remember that assignment of any part of PFC to
explicitly "social" functions requires
that the same areas are not involved in "nonsocial"
functions -- and this is not clearly the case.
You should also know something about the "affect versus
cognition" debate: is affect a byproduct of cognition, or are
affect and cognition mediated by separate systems. As
F&T point out, a lot depends on how you define "cognition"
-- and how you define "emotion".
Bruner famously reminded us that "the purpose of perception is action", and in Chapter 15 F&T take up the relation of social cognition to social behavior. It turns out that cognition and behavior are not necessarily related, and you should understand some of the reasons for this. For example, the differences between assimilation and contrast in goal-directed behavior. Of particular interest are the factors that mediate the relation between attitudes and behavior. You should also know something about action-identification theory. And, of course, we can use our own behavior to shape the impressions that others form about us, which is what impression-management (including self-handicapping) is all about.
The cognition-behavior link is complicated
by the fact that goal-related self-regulation isn't only a
matter of conscious choices. Goals can be
automatically and unconsciously activated by environmental stimuli, and and these goals can
automatically and unconsciously lead to behavior
(which, itself, may be unconscious!). All of which
underscores the point that "cognition" doesn't necessarily
mean conscious cognition.
Here's where F&T talk a little about
mirror neurons, but they have so little to say about this issue, don't worry about
This Material Was
Not Covered in Fall 2015, and Is Not on the
Due to changes in the academic calendar, I had to drop my lectures on "Social Construction" from the course, so you're not responsible for this material. But, in the context of an end-of-course review, you might want to reflect on some of these issues anyway, so I've kept the following material in. These lectures, when I gave them, connected themes raised at the very beginning of the course to wider sociological and philosophical issues. Cognitive psychology and cognitive science is about knowledge and knowing; in social cognition, the objects of knowledge are social in nature. But beginning with the philosophical debate between realists and idealists, psychologists and other cognitive scientists have debated the extent to which the objects of cognition are independent of the mind (realism) -- the contrasting position being that the objects of cognition depend somehow on our own mental activities. In general, the "Enlightenment stance" assumes (1) that there is an real world out there, which has an objective existence that is independent of the mind; (2) that our minds form a more or less accurate subjective representation of that world; and (3) that the laws of nature are objective, meaning that they don't depend on the subjective beliefs, attitudes, feelings, and desires of those who seek to know reality.
That's all well and good, but cognitive psychology, and especially social cognition, contains more than a few hints that, in important respects, our minds create the reality that we're trying to know -- meaning that there may be important senses in which the real world isn't independent of our minds after all.
The first hints of this lack of independence come from the constructivist approach to perception (Helmholtz et al.), which assumes that the proximal stimulus is inherently ambiguous, and that perception must go "beyond the information given" by the proximal stimulus, drawing on knowledge, expectations, and beliefs to make inferences about the distal stimulus. Similarly, Bartlett argued that remembering must go "beyond the information stored" in the memory trace, and entails reconstructive activity that also relies on knowledge, expectations, and beliefs (don't worry about the slides about memoir, the point of which is that memoir-writers don't actually remember things the way they happened). In these cases, of course, there's still a real world out there to be perceived and remembered. And Kelly argued that what's important is the way we construe events. But it's our perception of the world, and our memory of the past, and our construal of events, that determine our behavior -- and those percepts and memories and construals are very much the product of our own minds. So, in that sense, the reality that counts -- the reality we actually respond to -- is not independent of the observer.
But in the cases outlined above, there's still an objective reality that exists independently of our subjective representation of it. It's just that our subjective representation may not be precisely accurate. Things get more interesting when we consider the self-fulfilling prophecy, in which a false belief -- an incorrect subjective representation -- leads to behavior which makes the false belief come true by altering objective reality. While perceptual confirmation effects bring the subjective mental representation of reality in line with expectations, behavioral confirmation effects bring objective reality in line with expectations. With the self-fulfilling prophecy and behavioral confirmation, we are not just knowing reality; we are creating reality -- a reality that is very much the product of our minds.
But it gets even better. While the standard modern view distinguishes between an objective and a subjective reality, "post-modern" thinkers argue that reality is a social construction -- not objective at all, highly contingent on historical events and social structures. Taken to an extreme, post-modernists argue that everything from gender to quarks is socially constructed -- that it's not just true that people once thought that the world was flat; it really was flat, because people thought it was flat. It's like a philosophical TV show: "Idealists Gone Wild".
The philosopher Ian Hacking detailed the stages through which social constructivism seems to pass, and distinguished among various types and subtypes of social constructivism: historical, ironic, unmasking, reformist, rebellious, and revolutionary).
Some post-modern claims concerning social constructivism can seem downright silly, but the philosopher John Searle finds that there is actually a deep and interesting philosophical claim: that there is "an objective reality that is what it is only because we think it is what it is". As proof, he offers the examples of money, marriage, and property rights. Money is money only because we say it is; but money is money, objectively -- it's not merely a matter of someone's belief. Same holds true for marriage and property rights, among many other things.
The trick to understanding how this can happen, in Searle's view, is to understand that the terms "objective" and "subjective" have two quite different meanings, ontological and epistemic. In terms of ontology, something has an objective mode of existence if it does not depend on the subject's experience; and a subjective mode of existence if it exists only as experienced by some subject. In terms of epistemology, knowledge is subjective if its truth-value depends on the attitudes and feelings of the subject; knowledge is objective if its truth value is independent of the attitudes and feelings of the subject.
The point is that ontological objectivity-subjectivity and
epistemic objectivity-subjectivity are not perfectly
correlated. Using Searle's examples:
Understanding how someone could categorize an object as a paperweight, or evaluate the moon as beautiful, are matters of individual psychology. They're what cognitive psychology is all about. But understanding how someone could state that earthquakes are bad for real-estate values requires an understanding of the social context in which the individual's thoughts occur. Put bluntly, earthquakes can't be construed as bad for real-estate values unless one is in a particular kind of society -- a society in which people share a concept of real estate (and property rights) to begin with -- which themselves are constructions of the human mind.
Searle notes that intrinsic facts are the domain of the physical sciences, such as math and physics, while observer-relative facts are the domain of the social sciences, such as sociology and economics. Psychology, for its part, seems to be both a physical science and a social science. In part, psychology tries to discover universal -- intrinsic -- principles about how the mind works (and how the brain does it). In that respect, it's a natural science. But it's also true that a great deal of the subject-matter of psychology are matters of meaning -- facts that are observer-relative, because they are products of mental activity. Some of these observer-relative facts are the product of individual mental activity, while some of these are the product of what Searle (and other philosophers) call collective intentionality.
And, just to bring this all back to earth, social cognition tries to understand how people acquire, represent, transform, and use knowledge about the social world. But some of these social facts are themselves the product of individual mental activity -- they're not independent of the mind of the person who knows them. And some of these social facts are true only in certain social contexts -- because they are the product of the collective mental activity. Still and all, these facts are objectively true, even though they are observer-relative. In this sense, social reality is, at least in part, a social construction. In Hacking's terms, the challenge of social cognition is to understand things that are true, but which might be true only for those who think they are -- things that are true, but which might not be true for everyone, everywhere.
The lectures may have no parallel in F&T, but Zerubavel covers related topics in his Chapters 7 and 8 -- which were assigned for the lectures on 'Personality and Social Cognition" in order to give you time to read them. The point of Chapter 7 is that time, which we usually think of as having an observer-independent existence (which is why physics has such constructs as space-time), is also a social construct, as when we divide the years into BC (BCE) and Ad (CE), and the Jewish and Muslim calendars are different from the Christian one. Chapter 8 simply claims social and institutional reality as the subject matter for cognitive sociology.
You needed time to read them because Searle's (2006) article on social ontology can be rough going for people who haven't had an introduction to philosophy, or to Prof. Searle, which at least some in class haven't. Anyway, Searle discusses his important distinction between the ontological and epistemic senses of the objective-subjective distinction, arguing that it is possible for something to be ontologically subjective yet epistemically objective -- and, in fact, that a lot of social and institutional reality is just like that. You should have some understanding of the primitive structure of social reality: collective intentionality (remembering that Searle is using "intentionality" in its German-philosophical sense of the "aboutness" of mental states); assignment of function, and constitutive rules. Hey, here's a question: if the assignment of function is epistemically subjective, and social neuroscience is about assigning functions to parts of the brain, does that make social neuroscience a social construction?
The last lecture ended
with the question with which the course
began: the relationship between social cognition and cognition
in general. In some respects, social cognition is simply
cognition of social objects -- implying that social cognition
is simply derived from cognitive psychology in general.
Setting aside the question of priority, there are at least two important differences between the social and the nonsocial case. In the first place, as discussed in the previous lectures, the objects of social cognition are in important respects not part of an observer-independent reality. Instead, the objects of social cognition are mental products, of either individual or collective intentionality. In the second place, the objects of social cognition are also subjects, sentient beings who know they are being cognized, and who are actively trying to shape the cognitions of those who perceive, remember, and think about them.
Social cognition can proceed at several different levels: at the level of cognitive psychology, viewing people as objects, just like any other; at the level of social psychology, viewing people as conscious, sentient beings who are engaged in impression management as well as impression formation; at the level of sociology, viewing groups, institutions, and whole societies as engaged in cognitive activity much as individuals are; and at the level of neuroscience, examining the biological substrates of social-cognitive processes.
This page last modified 12/02/2015.