Toward an Ecological Theory of Social Perception

Psychological Review

1983, Vol.

90, No.

3,215-238

Copyright

1983

by the

American Psychological

Association,

Inc.

Toward

an

Ecological Theory

of

Social

Perception

Leslie

Zebrowitz

McArthur

Brandeis

University

Reuben

M.

Baron

University

of

Connecticut

The

ecological

approach

to

perception

(J.

Gibson,

1979; Shaw, Turvey,

&

Mace,

1982)

is

applied

to the

social

domain.

The

general

advantages

of

this

approach

are

enumerated,

its

applicability

to

social

perception

is

documented,

and its

spe-

cific

implications

for

research

on

emotion

perception,

impression

formation,

and

causal

attribution

are

discussed.

The

implications

of the

ecological

approach

for

our

understanding

of

errors

in

social

perception

are

also

considered. Finally,

the

major

tenets

of the

ecological

approach

are

contrasted

with

current

cognitive

approaches,

and a

plea

is

made

for

greater

attention

to the

role

of

perception

in

social knowing.

Current

research

on

social perception

op-

erates within

a

very narrow

schema—the

schema.

The

questions addressed concern

the

knowledge

structures (e.g., schemata, proto-

types,

scripts) that impose meaning

on the

blooming,

buzzing confusion around

us. The

questions ignored concern

the

structured

stimulation

that exists

in our

social environ-

ment.

As

such,

we

have learned much about

the

processing

of

information

and

little about

what that stimulus information

is. We

know,

for

example, that there

are

primacy

effects

in

impression formation:

The

person

who is

first

perceived

as

industrious

and

then

as

stubborn

will

often

be

judged more positively

than

one who is first

perceived

as

stubborn.

But,

what information (other than

a

verbal

label)

communicates industry

or

stubbor-

ness?

This

we do not

know. Similarly,

we

know

that there

are

discounting

effects

in

causal attribution: Success

on a

difficult

task

will

be

attributed

to

ability

if

little

effort

is

present,

but not if

there

is a

great deal

of

effort.

But, what information communicates

effort?

This

we do not

know.

To

fully

understand impression formation

or

causal attribution

or

other aspects

of

social

The

authors

would like

to

thank Teresa

Arnabile,

Su-

san

Fiske, David Hamilton, Bert Hodges,

Joann

Mon-

tepare,

Dick

Nisbett, David

Schneider,

and Jim

Todd

for

their very helpful

comments

on an

earlier draft

of

this

paper.

Requests

for

reprints

should

be

sent

to

Leslie

Z.

McArthur,

Department

of

Psychology, Brandeis Univer-

sity,

Waltham,

Massachusetts

02254

or to

Reuben

M.

Baron,

Department

of

Psychology, U20, University

of

Connecticut,

Storrs,

Connecticut

06268.

perception,

we

must ultimately

identify

the

nature

of the

stimulus information that

re-

veals

industry, hostility,

and the

other attri-

butes that

we

perceive

in

people. Whereas

current

cognitive approaches

fall

short

in

this

endeavor,

the

ecological approach

to

percep-

tion

provides

a

fruitful

model

for

theory

and

research.

The

purpose

of

this article

is to ad-

vocate this approach. More

specifically,

we

will

(a)

summarize

the

basic tenets

and ad-

vantages

of the

ecological position,

(b)

dem-

onstrate

the

applicability

of

this approach

to

the

realm

of

social perception,

and (c)

con-

sider

implications

of

this approach

for our

understanding

of

errors

in

social perception.

The

Ecological Position

and Its

Advantages

for

the

Study

of

Social Perception

What

we

call

the

"ecological position"

is

not a

unified

theory

of

perception; rather,

it

draws

on

several recent theories (e.g.,

J.

Gib-

son, 1966, 1979; Shaw, Turvey,

&

Mace,

1982).

This approach

has

four

distinguishing

features.

First,

it

assumes that perception

serves

an

adaptive

function

and

that

the ex-

ternal

world must therefore provide infor-

mation

to

guide biologically

and

socially

functional

behaviors. Second,

it

assumes that

this information

is

typically revealed

in ob-

jective physical

events—dynamic,

changing,

multimodal stimulus information

as

opposed

to

static

or

unimodal

displays.

Third,

it as-

sumes

that

the

information available

in

events

specifies,

among other things, environmental

affordances,

which

are the

opportunities

for

acting

or

being acted upon that

are

provided

215

216

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

by

environmental entities. Fourth,

it

assumes

that

the

perception

of

these

affordances

de-

pends

upon

the

perceivers'

attunement,

that

is,

the

particular stimulus invariants

to

which

the

perceiver attends.

The

Adaptive Nature

of

Perception

The

ecological approach

to

perception

be-

gins

by

assuming that perception serves

an

adaptive

function.

That

is, by

informing

ac-

tion,

perception

is

assumed

to

promote

in-

dividual

goal attainment

as

well

as

species

survival.

1

J.

Gibson

(1979),

for

example,

ar-

gues:

"The medium, substances, surfaces,

objects,

places,

and

other animals have

af-

fordances

for a

given animal. They

offer

ben-

efit

or

injury,

life

or

death. This

is why

they

need

to be

perceived"

(p.

143).

The

idea

that

perception

is a

matter

of

discovering

and ad-

justing

to

utilitarian properties

of the

envi-

ronment

has

important

ramifications

for

the-

ory

and

research

in

social

perception.

Most

generally,

it

focuses

attention

on

the

what

of

perceptual

processing

(i.e.,

the

useful,

struc-

tured

information

in the

environment),

whereas

traditional approaches emphasize

the how of

processing

(i.e.,

structuring mech-

anisms

in the

head;

see

Shaw

&

Bransford,

1977).

More

specifically,

the

proposition that

"perception

is for

doing"

(J.

Gibson, 1979)

focuses

our

attention

on a

particular subset

of

information within

the

perceiver's envi-

ronment,

namely

that

information which

(a)

is

revealed

in

events,

(b)

affords

adaptive

ac-

tion,

and (c) is

accessible

to the

perceptual

systems

of

the

perceiver.

Information

in

Events

The

ecological

position. Within

the

eco-

logical

approach

to

perception,

the

units

of

information

in

structured stimulation

are

events.

The

idea

of an

event

is

both simple

and

complex. Viewed descriptively,

all we

mean

by an

event

is a

dynamic

as

opposed

to a

static stimulus

display—a

motion picture

versus

a

posed photo

or a

person

walking

versus

a

person standing still.

The

dynamic

changes

over space

and

time that characterize

events

come

in

many

varieties.

For

example,

they

may be

fast

or

slow (e.g., smiling

vs.

aging),

they

may be

rigid

or

elastic (e.g.,

ro-

tating

vs.

stretching), they

may be

reversible

or

nonreversible (e.g., rolling

vs.

growing).

And,

as the

foregoing examples illustrate,

events

occur

in the

social

as

well

as the

phys-

ical

environment.

Events

provide perceivers with structured

information

that

supplements

the

informa-

tion that

is

available

in

static

stimuli.

For ex-

ample,

one can

perceive certain properties

in

a

stationary sphere, such

as its

size,

its

color,

and its

texture.

But

there

are

other attributes

that

can be

perceived only

in

events. These

include

the

heaviness

of the

sphere

that

can

be

perceived

by

observing

its

rate

of

accel-

eration down

an

incline,

by

observing some-

one

lift

it, or by

lifting

it

oneself.

This

last

example

highlights

an

important point,

namely,

that events

may be

created

by the

perceiver;

dynamic stimulus information

may

result

not

only

from

the

displacement

of en-

tities

in the

environment,

but

also

from

the

exploratory

behaviors

of an

active perceiver.

Certain properties

of an

entity will change

during

an

event, whereas others remain

the

same.

Those

elements

that

remain

the

same

are

referred

to as

structural invariants.

For

example,

the

shape

and

color

of an

object

remains

the

same

as it

rotates;

similarly, most

of

your features remain

the

same

as you

smile

or

even

as you

age.

The

styles

of

change that

events

may

manifest

are

referred

to as

trans-

formational

invariants.

For

example,

we

rec-

ognize

a

particular style

of

biomechanical

movement

as

"walking"

over

a

wide range

of

structures

(e.g., humans, cows,

and

dogs).

Similarly,

we

identify

particular morpholog-

ical

changes

as

"aging"

over

a

wide range

of

species

and

forms

(cf. Todd, Mark, Shaw,

&

Pittenger, 1980).

Advantages

for

social

perception

research.

Whereas

current research

in

social percep-

tion

has

concentrated

on the

cognitive pro-

cessing

of

social

information—how

sche-

mata, memory,

and

various cognitive heu-

ristics

influence

the

meaning

we

extract

from

the

social

environment—research

within

the

1

The

proposition

that

perception

is

adaptive repre-

sents

a

metatheoretical

or

primitive

assumption that

is

not

directly

falsifiable.

Making

this

assumption

does,

however,

lead

to

empirical

hypotheses

that

are

capable

of

disconfirmation.

See,

for

example,

the

sections

on

emotion,

error,

and

attributions

in

this article.

SOCIAL PERCEPTION

217

ecological approach will reveal

the

meaning

that

is

communicated

by

social events. More

specifically,

such research will reveal what

it

is

in a

person's movements, gestures, voice,

and

facial

appearance that communicates

to

us

that

person's

momentary intentions, emo-

tional

state,

or

more stable qualities,

and

what

it is in the

interactions between

two or

more people that communicates

to us the

nature

of

their relationship even when

we

cannot

hear

their

words (cf.

Archer

&

Akert,

1977).

We

clearly make great

use of

such

in-

formation

in our

daily

life,

for

example, when

deciding whom

to ask for

directions, whom

to ask for a

date,

or

whom

to

steer clear

of

in

the

subway. However, current research

does

not

provide

us

with

a

description

of the

stimulus information

to

which

we are re-

sponding.

A

notable exception

is

research

on

nonverbal

communication,

an

excellent

ex-

ample

of

focusing

on

information

in the

stim-

ulus. (See Schneider,

Hastorf,

&

Ellsworth,

1979,

for a

comprehensive review

of the

per-

tinent literature.) However, this research

has

tended

to

concentrate upon

the

communi-

cation

of

affect.

There

is

other psychological

information

in the

extensional properties

of

people,

and it

needs

to be

identified.

The

ecological approach

not

only focuses

on

the

neglected

and

important question

of

what

information

is

provided

by

social stim-

uli,

but it

also emphasizes

the

importance

of

addressing

this

question

by

examining

dy-

namic stimulus information, something

which

is

rarely done

in

social perception

re-

search. Perceivers more

often

make social

judgments

from

written summaries

of

social

events

than

from

direct exposure

to

those

events

as

they unfold. Thus

we

study social

cognition,

not

social perception.

The

ecolog-

ical approach makes explicit that

if we

want

to

study social perception,

we

must give

our

subjects

a

chance

to

perceive people

and

their

behavior.

Integral

to

the

emphasis

on

dynamic stim-

ulus

information within

the

ecological

ap-

proach

is an

emphasis

on the

active perceiver.

Not

only

is the

stimulus information pro-

vided

by

inert objects shown

to

become

dy-

namic when perceivers

are

permitted active

perceptual exploration (e.g.,

J.

Gibson,

1966,

p.

195),

but

also,

it is

assumed

that

the

prop-

erties

of the

external environment

will

be

more accurately detected when perceivers

are

allowed

such active exploration.

The

ecolog-

ical approach emphasizes

the

intrinsic con-

nection between

action

and

perception,

something

that

has

been given

insufficient

consideration

in

traditional theories

of

social

perception.

The

Perception

of

Affordances

The

ecological

position.

Perception

in the

ecological approach

is not

concerned with

just

any

information. Rather,

it

concerns

the

pick-up

of

useful

information.

The

useful-

ness

of

information depends upon

its

rele-

vance

to the

perceivers'

actions

and

goals,

and

for

this reason,

the

ecological approach

stresses

the

perception

of the

affordances

of

the

environment. These

are

defined

by J.

Gibson (1979)

as

"what

it

offers

the

animal,

what

it

provides

or

furnishes,

either

for

good

or

ill"

(p.

127).

A

more

poetic

and

vivid

in-

dication

of

what Gibson means

by

affordance

is

provided

by his

quotation

from

Koffka

(1935,

p. 7):

"Each

thing says what

it is

...

a

fruit

says

'eat

me';

water

says

'drink

me';

thunder says

'fear

me';

and

woman says

'love

me'

"

(J.

Gibson, 1979,

p.

138).

Gibson

maintains that

the

action possibilities pro-

vided

by an

object

in the

environment

as

well

as the

consequences

of

interacting

with

that

object

may be

revealed

in its

extensional,

physical characteristics.

For

example,

the

edibility

of

fruit

may be

specified

extension-

ally

by

color, smell, size,

and

texture.

Of

course,

any one of

these properties

may not

be

sufficient

to

reveal

a

fruit's edibility.

Rather,

the

detection

of

this

affordance

may

require

event information such

as

that pro-

vided

by

grasping

the

fruit

and

squeezing

it

or

even breaking

it

open. This

fact

highlights

the

mutuality

of the

affordances

of the en-

vironment

and the

behavior

of the

animal.

And,

although

Gibson

emphasizes

the

objec-

tive

reality

of

affordances,

he

also emphasizes

this

synergy: "Affordances

are

properties

of

things

taken with

reference

to an

observer

but

not

properties

of the

experiences

of the ob-

server.

They

are not

subjective values."

(J.

Gibson, 1979,

p.

137)

Thus,

a

fruit

affords

eating

by

some

observers

but not

others:

Monkeys

eat

bananas

and

coconuts, whereas

ants typically

do

not.

218

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

Advantages

for

social perception research.

Although

the

concept

of

affordance

is

closely

linked

to

Kurt Lewin's (1936) concept

of va-

lence,

it has not

been represented

in

social

perception research

any

more than

in

per-

ception research

in the

nonsocial

domain.

2

Rather,

research

in

social perception

has fo-

cused

on the

detection

of

structural invari-

ants—dispositional

properties such

as

abili-

ties

and

traits. Conceptualizing

our

social

perceptions

in

terms

of

affordances

may be

both

more valid

and

more useful than

the

trait approach.

There

is now

considerable evidence that

people

may not

have

traits

as

traditionally

conceived. Rather, behavior

may be

specific

to

situations

as

well

as to the

person with

whom

one is

interacting

(e.g., Mischel,

1968).

The

concept

of

affordance

deals nicely with

this possibility, because

an

affordance

is in-

herently specific

to a

particular

perceiver—

what

John

affords

me may or may not be the

same

as

what John

affords

you.

In

addition

to

better capturing

the

realities

of

human

behavior,

the

concept

of

affordance

may

often

surpass

the

trait approach

in

capturing

the

perceivers' phenomenology. Although

we may

sometimes

use

trait terms

to

think about peo-

ple

(and although

our

research subjects have

almost invariably been forced

to do

so),

it is

likely

that

we

often

want

to

know what people

qan

do to or for us and

what

we can do to

them, rather than some abstract information

regarding

the

number

of

situations

in

which

they

will

behave thus

and so.

Not

only

do

affordances

often

capture

the

properties

of

other

people

and our

percep-

tions

of

them better than traits

do, but the

affordance

concept

has the

further

advantage

of

permitting

us to

assess

the

accuracy

of

perceptions more readily than

the

trait con-

cept does. Although

it is

difficult

if not im-

possible

to find a

criterion

appropriate

for

validating

perceived traits, behavioral evi-

dence

can

serve

to

validate perceived

afford-

ances.

For

example,

if

someone

is

perceived

to

afford

protection,

one can

determine

the

accuracy

of

this perception

by

ascertaining

whether

that person

will

indeed provide pro-

tection

to the

perceiver. However,

if

someone

is

perceived

as

protective, then

any

instance

of

protecting

or

nonprotecting

can

neither

confirm

nor

disconfirm

this trait ascription.

Perceptual

Attunements

The

ecological

position.

The

fact

that

af-

fordances

are

perceiver referenced highlights

a

basic tenet

of the

ecological approach,

namely, that perception requires certain

compatibilities between

the

perceiver

and the

perceived.

At the

most fundamental level,

there must

be a

match between animals'

re-

ceptor capabilities

and the

stimulus

infor-

mation

to

which they

are

perceptually sen-

sitive. Indeed,

it can be

hypothesized

that

perceptual

systems have evolved

to be

sen-

sitive

to the

types

of

structured information

available

in a

given ecological niche. Thus,

animals

who

live

in

darkened areas, such

as

caves,

have poorly developed visual systems

but

very well-developed auditory systems

that

permit navigation

and

hunting.

The

concept

of

perceptual attunement

in the

ecological

approach

is not

limited

to

attunements

that

occur

through biological preprogramming

in

response

to

evolutionary pressures. Attune-

ments

may

also derive

from

what

J.

Gibson

(1966)

has

called

the

"education

of

atten-

tion." More

specifically,

the

stimulus

infor-

mation

to

which perceivers

are

attuned

may

vary

as a

function

of

their perceptual learn-

ing,

goals, expectations,

and

actions.

The

influence

of

perceptual learning

on

attunements

is

revealed when

one

compares

the

forest

that

is

seen through

the

eyes

of a

naturalist

to the

forest seen through

the

eyes

of

a

city dweller. Whereas

the

naturalist will

differentiate—that

is,

see—many

levels

of

structure,

the

city person

may

never

see

any-

thing more than

a

uniform

and

boring mass

of

green.

On a

less anecdotal level,

the

master

chess player sees more moves

on an

appro-

priately arranged

chessboard

than

the

novice

sees.

Evidence that

the

expert actually sees

the

board

differently

is

provided

by

Neisser's

(1976)

report

that

experts show more

board-

dependent

eye fixations

than novices

do.

Cross-cultural

differences

in

susceptibility

to

perceptual illusions

also

provide evidence

for

2

According

to J.

Gibson (1979,

p.

138),

Kurt Lewin's

concept

of

valence

is an

English translation

of the

term

Aufforderungscharakter,

which

has

also been translated

as

invitation character,

a

translation that probably makes

the

concept

clearer

as

well

as

more

readily identifiable

as

analogous

to the

concept

of

affordance.

SOCIAL PERCEPTION

219

the

role

of

perceptual learning (e.g., Segall,

Campbell,

&

Herskovitz, 1966).

The

influence

of

goals upon perceptual

attunements

is not a

domain

of

exclusive

in-

terest

to the

ecological approach. Indeed, this

topic

has

been extensively explored

in a

body

of

research known

as the

"new

look",

which

sought

to

demonstrate that

the

perception

of

stimuli

may be

inhibited

or

enhanced

as a

function

of the

needs

or

goals

of the

perceiver

(e.g.,

Postman, Bruner,

&

McGinnies,

1948).

Although

this research

was

earlier subject

to

considerable criticism (e.g., Goldiamond,

1958),

it has

recently been revitalized

in Er-

delyi's

(1974)

reformulation,

and its

primary

emphasis

is

quite compatible with

the

eco-

logical

view

that perceivers

are

attuned

to the

stimulus information

that

is

most relevant

to

adaptive

actions.

Like

motivational

influences

upon percep-

tion,

the

postulated

influence

of

expectations

is

not

limited

to the

ecological approach.

However,

one can

differentiate

between

two

expectancy

effects,

only

one of

which reflects

perceptual

attunement.

In the

effect

that

op-

erates

at a

perceptual level, perceivers with

one

expectation

may

partition

a

complex

stimulus

differently

from

those with another

expectation.

In

Asch's

(1952)

terms, these

expectancy

effects

reflect

variations

in the

object

of

judgment—the

effective

stimulus

information

is

different

for

different

sets.

Another

kind

of

expectancy

effect

operates

at a

more inferential level: Perceivers with

one

expectation interpret

the

same stimulus

information

differently

from

those with

an-

other

expectation.

In

Asch's (1952) terms,

such

expectancy

effects

reflect

variations

in

the

judgment

of the

object.

The

role

of

action

in

perceptual attune-

ments

is

obvious

but

often

overlooked

in

tra-

ditional

approaches that treat

the

perceiver

as a

passive receptacle

for

sensory stimula-

tion.

Actions constrain what

is

perceived.

Where

one is

walking, what

one is

touching,

where

one

looks, listens,

and

sniff's

will

all

influence

the

particular subset

of

information

that

is

available

to

guide subsequent actions.

Advantages

for

social perception research.

The

assumption that perception

is

selective

and

that what

is

perceived

will

vary

from

perceiver

to

perceiver

is not

unique

to the

ecological

approach. What

is

unique

is the

basis upon which selectivity

and

individual

differences

are

postulated

and, therefore,

the

factors

presumed

to

influence them.

According

to the

ecological position,

we

are

sensitive

to

adaptively relevant informa-

tion,

as

opposed

to all

possible information.

Although

this

may

seem

a

self-evident truth,

it

is

something that

has not

been central

to

research

and

theory

in

social perception,

and

in

underscoring this point,

the

ecological

ap-

proach highlights

new

areas

for

research.

In

particular,

the

assumption

that

what

we

per-

ceive

in the

social environment

is

likely

to

be

first and

foremost that which

is

most

es-

sential

to

adaptive action suggests

a

hierarchy

in

the

ease with which various social prop-

erties

are

perceived.

For

example, emotions

such

as

anger

and

fear

should

be

most

readily

perceived because they

are

most essential

to

adaptive action

on the

part

of the

perceiver.

3

Similarly,

the

stable

attributes

of

domineer-

ingness,

aggressiveness,

and

strength should

be

perceived more readily than kindness,

in-

telligence,

or

reliability

because

the

former

are

more essential

to

adaptive action.

The

ecological approach

not

only draws

our

attention

to the

fact

that

some

social

properties should

be

more readily detected

than others

by all

perceivers

but

also

it

sheds

new

light

on the

issue

of

individual

differ-

ences.

Different

information

is

essential

to

the

behavioral goals

of

different

people. What

low-status

people need

to

perceive

in

order

to

interact

effectively

with their environment

may

often

be

different

from

what high-status

people need

to

perceive; what people

in one

culture

need

to

perceive

may be

different

from

what those

in

another culture need

to

perceive; what people

in one

occupation need

to

perceive

may

differ

from what people

in

another occupation need

to

perceive;

and so

forth.

4

Thus,

the

ecological approach suggests

3

Some evidence consistent with this argument

is

pro-

vided

by

Averill's (1983)

finding

that

there

are

many

more

words

in the

English

language

referring

to

negative

emotions than

to

positive

ones.

4

The

truth

of

this

assertion

has

been

vividly illustrated

to the first

author recently

as a

consequence

of a new

hobby—oil

painting.

The

necessity

to

create

on

canvas

what

is

seen

in the

external world

has fine

tuned

her

perceptual capacities

in

ways never imagined

possible—

shadows,

for

example,

that

formerly went

unnoticed,

have

now

become

palpable

entities

with color,

texture,

and

form.

220

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

that

we may

discover much

of

interest

in the

domain

of

individual differences

in

social

perception

if we

begin

our

investigations with

a

careful

analysis

of

what

it is

that various

individuals

most need

to

perceive

in

order

to

interact

effectively

with their social envi-

ronment.

The

concept

of

perceptual attunement

is

pertinent

to

research

on the

development

of

social perceptions

as

well

as

individual dif-

ferences.

According

to J.

Gibson

(1966),

per-

ceptual development involves

the

"education

of

attention."

More specifically, Eleanor Gib-

son's

(1969)

perceptual

differentiation

theory

holds that

as

perception

develops,

the

organ-

ism

extracts

from

the

total stimulus

flux

cer-

tain

distinctive features

and

invariant rela-

tionships

not

previously detected. Because

developing

an

attunement

to

these

new

prop-

erties

of

stimulation will result

from

an

active

search

directed

by the

task

at

hand,

one

would expect

perceptual

attunements

to de-

velop

in

synchrony with

the

development

of

behavioral

capabilities.

As

such,

we may

learn

much about

the

development

of

social

perceptions

(i.e.,

the

social

invariants

to

which

perceivers

at a

given developmental

level

are

attuned)

if we

begin

with

an

analysis

of

the

behavioral goals that

the

social per-

ceptions

can

serve.

The

concept

of

perceptual

attunement

has a

further

implication

for

studying

the

development

of

social percep-

tions. Intrinsic

to the

concept

of

perceptual

attunement

is the

assumption that perceivers

can

extract

certain

information

from

the

stimulus

environment

before

the

develop-

ment

of

conceptual

structures

and

language.

Because

testing this assumption requires

re-

search

investigating

the

perceptions

of

young

children,

developmental research becomes

an

essential component

of

research

on

social

perception within

the

ecological approach

rather than

a

separate enterprise.

Applicability

of the

Ecological Approach

to

Social Perception

We

have shown

that

an

ecological

theory

of

perception

offers

several advantages over

current approaches

to

social

perception.

The

question

to be

considered

in

this section

is

"Can such

a

theory

be

applied

to the

realm

of

social perception?"

Adaptive

Function

of

Social

Perception

One of the

basic tenets

of the

ecological

approach

is

that perception serves

an

adap-

tive

function

for the

organism inasmuch

as

it

informs

(and

is

informed

by)

action.

The

fact

that

it is

biologically important

to

detect

properties

of

the

physical environment such

as

chasms

or

obstacles

or fires

bolsters

the

argument

that

we are

attuned

to the

stimulus

information

that reveals these properties

and

need

not

rely

upon

the

slower

and

less sure

processes

of

inference (cf. Nisbett

&

Ross,

1980).

Can we

provide

the

same kind

of un-

derpinnings

for an

ecological theory

of

social

perception?

The

adaptive value

of

attune-

ment

to

stimulus information

specifying

so-

cial events

has

been acknowledged

by J.

Gib-

son

(1979),

who

argued, "For

any

animal

needs

to

distinguish

not

only

the

substances

and

objects.

...

It

cannot

afford

to

confuse

prey with predator, own-species with another

species,

or

male with female"

(p. 7).

Certainly

it is as

adaptive

to

differentiate

male

from

female

and

prey

from

predator

in the

social

environment

as it is to

detect properties

of

the

physical environment. Similarly,

the

help-

lessness

of

babies,

the

looks

of

fear

or

anger,

and the

ravages

of

disease

require

fast

and

sure

recognition

if our

species

is to

survive.

It

should

be

noted,

however,

that

the

adaptive

value

of

accurate social perception

is not

lim-

ited

to its

utility

for

species survival. Accurate

social perception also serves

an

adaptive

function

at the

level

of

individual goal

at-

tainment—for

example, getting

from

others

that

which

we

desire

and

escaping that which

we

dislike.

Stimulus

Information

in

Social Events

Although

one can

make

a

convincing

ar-

gument

that attunement

to

stimulus

infor-

mation

specifying

social events

has

adaptive

value,

applying

the

ecological approach

to

social perception requires more than dem-

onstrating

a

Darwinian

aesthetic

appeal.

One

must also argue

that

social perception rests

upon

the

same

types

of

information

as

per-

ceptions

of the

physical environment.

The

first

question

is

"Are social

properties

re-

vealed

in

events?"

J.

Gibson's arguments

on

behalf

of

event perception

are

actually more

telling

in the

realm

of

social perception

than

SOCIAL

PERCEPTION

221

in

the

realm

of

object perception. Perceiving

the

properties

of

people, even more than

those

of

inanimate objects,

will

often

require

a

dynamic stimulus display both because

hu-

mans

are

animate

and

because many

of

their

properties

can be

detected only

in

their

ac-

tions.

One can

thus argue simply

on

logical

grounds

that social properties must

often

be

revealed

in

dynamic stimulus displays.

A

sec-

ond

question that must

be

considered

in ap-

plying

the

ecological

approach

to the

social

realm

is "Do the

stimulus properties detected

in

the

realm

of

social perception include

structural

invariants, transformational

in-

variants

and

affordances,

just

as in the

phys-

ical

realm?"

Structural

invariants. Just

as the

physical

appearance

of a

windmill reveals

to

perceiv-

ers

structural invariants (e.g.,

a

vertical

shaft

and

sails),

so

does

the

physical appearance

of

people

reveal certain

structural

invariants

(e.g.,

a

backbone

and

limbs). However, people

are

also perceived

to

have invariant proper-

ties whose physical extension

is not so

readily

apparent. Consider,

for

example, Heider's

(1958)

concept

of

dispositional

properties,

which

he

defines

as

"the invariances that

make

possible

a

more

or

less stable, predict-

able,

and

controllable world." According

to

Heider,

these relatively unchanging struc-

tures

include

"such

object properties

as

color

and

size,

and

such person properties

as

char-

acter

and

ability"

(p.

80).

Can we

directly

perceive

human structural invariants such

as

character

and

ability? Although Heider sug-

gests that

we can

when

he

says, "Certainly

a

person's apparent self-confidence

often

influ-

ences

our

judgments

of his

abilities"

(p.

94),

his

writings stressed

the

role

of

inference

in

social

perception:

Thus,

it is

very

unlikely

that

we

will

be

able

to

coordinate

univocally

certain characteristics

of the

stimulus mani-

folds

to

impressions

of

personality traits, social

acts,

or

sentiments

in a

simple

way.

In

order

to

understand

the

connection between

the

stimulus

pattern

and the

impres-

sion,

we

have

to

resort

to

thought models which

are

more

complicated,

(p. 24)

Following

Heider's lead, most research

in

social perception

has

assumed that social

in-

variants must

be

inferred,

as

opposed

to

being

specified

in the

stimulus. However,

ev-

idence

that structural invariants such

as

abil-

ity

or

personality

are

indeed specified

in the

stimulus

has

been provided

in

research

ex-

amining

stimulus information presumed

to

be

important

on the

basis

of

ethological

the-

ories.

For

example,

Guthrie

(1970)

has

pro-

posed several morphological characteristics

that

may

signal dominance

by

virtue

of

their

roots

in our

phylogenetic past. And, consis-

tent with

his

postulates, recent research

by

Keating,

Mazur,

and

Segall

(1981)

has re-

vealed

that persons with

the

broader

of two

faces

or the

more

receding

of two

hairlines

are

perceived

as the

more dominant

by

peo-

ple

from

a

variety

of

cultures. Similarly,

drawing

on

Lorenz's (1970) postulates

re-

garding

infantile stimulus features,

Mc-

Arthur

and her

colleagues

(McArthur

&

Apatow,

Note

1;

Berry

&

McArthur, Note

2)

have

found

that adult males with relatively

"babyish"

facial

features, such

as

large eyes,

short noses

and

ears,

or low

vertical place-

ment

of all

features,

are

perceived

as

less

strong

and

domineering than those with more

"mature"

features—smaller

eyes, longer noses

and

ears,

or

higher placement

of all

features.

To

discover

the

stimuli that reveal human

structural invariants

may

require

not

only

examining

stimulus information that

is

well

grounded

ethologically

but

also scaling

up to

a

higher level

of

stimulus complexity.

For

example,

it may be

possible

to

detect person-

ality

dispositions such

as

shyness

or

nurtur-

ance when individuals

are

observed

in

dyads

or

in

groups

but not

when they

are

observed

in

isolation. Similarly,

it may be

possible

to

detect certain traits when people

are

observed

in

action

but not

when they

are

observed

in

more static displays. Consider,

for

example,

KofTka's

(1935) presumption

of a

mapping

of

inner qualities into overt actions: "The

slow

dragging movements

of the

depressed,

the

jerky, discontinuous movements

of the

irritable, correspond, indeed,

to the

leaden

state

of

depression

or the

disrupted state

of

irritability"

(p.

658). More substantial evi-

dence

that

a

person's

gait

can

reveal

struc-

tural invariants

is

provided

by the

work

of

Cutting

and his

colleagues.

By filming

people

in

darkness with point lights

on

their major

joints, Kozlowski

and

Cutting

(1977)

dem-

onstrated

that

a

walker's

sex can be

recog-

nized

from

the

moving point lights without

familiarity

cues. Subsequent research

re-

vealed

that

the

perception

of a

biomechanical

222

LESLIE

Z.

MCARTHUR

AND

REUBEN

M.

BARON

invariant

in a

person's

gait—the

center

of

moment—is

sufficient

for the

identification

of

a

walker's

sex

(Cutting, 1978; Cutting,

Profitt,

&

Kozlowsky,

1978).

The

foregoing

examples

suggest that

as we

begin

to

consider

the

stimulus information

in

events,

we may

find

that there

is

indeed stimulus information

sufficient

for the

perception

of

many human

structural

invariants.

The

question remains

as to

what structural

invariants

are

specified

by the

stimulus

in-

formation

in

social events. Although

the an-

swer

to

this

question

must

be

established

em-

pirically,

some boundary conditions

can be

derived

from

the

theory

of

ecological percep-

tion,

which suggests

that

there will

be

stim-

ulus

information

sufficient

for the

perception

of

structural invariants whose perception

has

important adaptive value

either

to the

per-

ceiver

or the

perceived. Drawing

on J.

Gib-

son's

(1979)

proposition that animals cannot

afford

to

confuse prey

with

predator,

own

species with another,

or

male with

female,

one

would expect social stimulus informa-

tion

to

specify benevolence versus malevo-

lence,

in-group

versus out-group,

and

gender

and

sexual receptivity. Similarly,

the

adaptive

value

of

detecting structural invariants such

as

physical strength

and

illness, mental

as-

tuteness

and

insanity,

and

social dominance

and

dependency argues

for

their specification

in

the

stimulus information that people pro-

ject.

In

contrast, there

are

other physical,

mental,

and

social qualities whose detection

has

less general adaptive importance (e.g.,

fine

motor coordination, mathematical rea-

soning ability, sense

of

humor),

and

these

structural invariants should

be

less clearly

specified

in

social stimulus information,

re-

quiring

more inferential processes

for

their

identification.

Of

course,

it

should

be

reit-

erated that, according

to the

ecological

ap-

proach,

the

detection

of any

structural

in-

variant

is

more likely when event informa-

tion

is

provided

and

when

its

detection

is

relevant

for the

perceivers' actions.

Transformational

invariants.

In

addition

to

manifesting various

structural

invariants,

a

windmill reveals

in its

motions

the

trans-

formational

invariant rotating. Similarly,

transformational

invariants

such

as

walking

or

running

are

given

in the

physical move-

ment

of

people.

For

example, Johansson

(1973)

has filmed

motion patterns

by

means

of

a

moving-dot technique

in

which

10

small

luminous points were attached

to the

main

limb joints

of an

actor

who was

then

filmed

in

near darkness while walking

or

running.

Although

the film

depicted nothing

but a

group

of 10

bright dots, each moving

in its

own

path, perceivers were able

to

detect

a

walking

or a

running

man

when viewing

the

film

for

as

little

as 200

msec. What

these

data

tell

us is

that perceivers

are

attuned

to

very

abstract information concerning

the

invari-

ances

in

their

environment.

The

abstract

character

of

this information coupled with

the

speed with which

it is

translated into

a

percept

indicates

that

kinematic

stimulus

in-

formation

is

sufficient

for the

perception

of

human

locomotion.

5

Recent research

by

Runeson

and

Frykholm

(1981)

indicates

that

such information

is

also

sufficient

for the

perception

of

human

effort,

inasmuch

as the

relative heaviness

of

lifted weights

was

per-

ceivable

from

moving point-light displays.

It

thus appears that trying

is

revealed

in the

speed,

direction,

and/or

smoothness

of mo-

tions.

The

transformational invariants

that

char-

acterize walking, running,

and

trying

in the

foregoing

research involved changes

in the

layout

(i.e.,

the

position)

of an

individual.

Changes

in the

layout

of two or

more people

may

also

be

directly specified

by the

moving

stimulus

display that they project. Thus,

again

using

the

moving-dot technique,

Jo-

hansson

(1973)

found that perceivers

are

able

to

detect dancing when

viewing

a film

that

depicts nothing

but a

group

of 20

bright

dots

that

are

affixed

to the

joints

of two

dancing

people. Similarly,

the

classic, animated

film

produced

by

Heider

and

Simmel

(1944)

re-

veals

that perceivers detect chasing

and fight-

ing

in

kinematic stimulus information.

It is

conceivable that there

are

transformational

invariants

specific

to

other social events such

as

lovemaking,

dominating,

submitting,

5

It is

recognized

that

the

point-light

research

involves

an

impoverished stimulus display

on all

dimensions

of

a

stimulus array except parameters

of

movement.

This

is

by

intention, because this research

is

designed

to

dem-

onstrate

that

kinematic stimulus information

is, in and

of

itself,

sufficient

for the

perception

of

various mean-

ingful

human behaviors.

SOCIAL PERCEPTION

223

mothering,

and so

forth.

If so,

then

sufficient

information

for the

detection

of

these events

may

be

provided

by the

motion vectors that

people project. Some evidence that complex

social

relationships

are

specified

in the mo-

tions

of two

interacting people

is

provided

by

Archer

and

Akert

(1977),

who

exposed some

perceivers

to

short

videotaped

clips

of a

nat-

ural

social interaction

and

other perceivers

merely

to a

written

transcript

of the

inter-

action.

The

results

of

this research were very

striking: Only those

who

were

able

to see the

interaction were able

to

accurately detect

a

variety

of

social

relationships,

including kin-

ship, friendship,

and

status

differentials.

Archer

and

Akert's

data

suggest that there

are

transformational invariants

specific

to

social events such

as

parenting, winning,

def-

erence,

and

friendship.

The

task remains

to

locate

and to

systematically describe

the na-

ture

of the

stimulus information that reveals

these invariants.

In

some cases,

sufficient

stimulus

information

may be

provided

by

changes

in

people's gross motor activity,

as

is

true

for

walking, running,

and

dancing.

In

other cases,

sufficient

stimulus information

may

be

provided

by finer

motor

activities.

Emotional expressions,

for

example, involve

characteristic

changes

in the

facial stimulus

array

as

well

as in

other nonverbal behaviors,

and the

sufficiency

of

such information

for

the

detection

of

emotions

has

been demon-

strated

in a

large number

of

studies (e.g.,

Rosenthal, Hall, DiMatteo, Rogers,

&

Archer,

1979).

Not

only

do

characteristic

facial

expressions, gestures,

and

vocal qualities pro-

vide information

sufficient

for the

detection

of

emotion, but, moreover, some recent

re-

search

by

Bassili

(1978)

using

a

modified

point-light

technique suggests that

the

emo-

tion

may be

revealed

in the

nonverbal changes

per se.

Because separate features,

facial

lines,

and so

forth

are not

visible with this proce-

dure,

it

allows

one to

determine whether

fa-

cial movement, apart

from

the

appearance

of

particular

features,

communicates

emotion.

The

results reveal that emotions

can

indeed

be

specified

by

facial

motor

activity occurring

over

time.

Changes

slower

than

those

transformations

that reveal emotional state

are

associated

with changes

in

age. Wrinkling,

drooping

skin,

and

changes

in

craniofacial shape

are

invariant transformations

of the

human

face

and

head with increasing age. Considerable

research supports

the

hypothesis that percep-

tions

of a

person's

age are

strongly tied

to

these

transformational invariants.

For ex-

ample, Shaw

and his

associates (Shaw

&

Pit-

tenger,

1977; Todd

et

al.,

1980) have inves-

tigated

perceivers'

sensitivity

to

differences

in

the

shape

of

facial

profiles

that

are

associated

with growth. Subjects' judgments

of the age

of

various

facial

profiles

supported

the hy-

pothesis

that

increasing levels

of a

cardiodal

strain

transformation (which simulates

ac-

tual growth) performed

on a

standard

profile

would

produce increases

in the

perceived

age

of

the

profile.

Furthermore, subjects' sensi-

tivity

to the

profile

differences

was

very acute:

(a)

They could

detect

and

associate with

age

shape

differences

produced

by the

strain

transformation

that were only

a few

times

greater

than

the

absolute limit determined

for

visual

acuity

in

resolving spatially adjacent

lines,

and (b)

these discriminations were

made

very

quickly. These

findings all

suggest

that changes

in

craniofacial morphology

are

sufficient

for the

identification

of

changes

in

a

person's

age.

Affordances.

It has

been argued that

the

structural

and

transformational invariants

that characterize social events

may be

spec-

ified

in the

physical

extensions

of

those

events

just

as

they

are for

nonsocial events. What

about

human

affordances?

It is

reasonable

to

argue

that people have physical properties

that

modulate

light

and

sound waves

in a

manner

that reveals their

affordances

just

as

the

structure

of a

windmill

may

reveal

its

climbability?

Although acknowledging that

human

affordances

are in

certain respects

unique,

J.

Gibson (1979) would answer

in the

affirmative.

The

perceiving

of

these mutual affordances

...

is

just

as

much

based

on

stimulus

information

as is the

simpler

perception

of the

support that

is

offered

by the

ground

under

one's

feel

. . .

other persons

can

only

give

off

information

about themselves

insofar

as

they

are

tan-

gible,

audible, odorous,

testable,

or

visible,

(p.

135)

Although

there

has

been little research

concerning

the

perception

of

affordances,

some

evidence

that

they

are in

fact revealed

in

social stimuli

is

provided

by

recent studies

examining

perceptions

of

people

who

vary

in

the

babyishness

of

their appearances.

Alley

224

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

(1981)

found

that

perceivers report

a

greater

desire

to

protect

from

attack

and to

cuddle

stimulus persons whose head shape

or

bodily

proportions

are

babyish than those

who

have

a

more mature appearance.

Alley

also found

a

greater desire

to

protect,

but not to

cuddle,

persons with

an

elderly craniofacial appear-

ance than

persons

who

were young

adults.

have been obtained

by

McArthur

and

Apatow (Note

1),

who

found

that

faces

varying

in the

babyishness

of

their

features

are

perceived

as

affording

different

social

interactions.

For

example, persons

with

more

babyish faces were

perceived

as

less

likely

to

turn

a

cold shoulder

to the

sub-

jects'

attempts

at

friendly

conversation,

less

able

to

move several boxes

of the

subjects'

heaviest books,

and

more likely

to be the

kind

of

roommate

who

would comply with

all of

the

subjects' wishes about furniture arrange-

ment, quiet hours,

and so

forth.

Attunements

in

Social

Perception

Although

we

have discussed

the

perception

of

social

affordances

in a

manner that does

not

distinguish among perceivers,

it is im-

portant

to

recall that

in the

ecological

ap-

proach

to

perception

"affordances

are

prop-

erties

of

things taken with

reference

to an ob-

server"

(J.

Gibson,

1979,

p'.

137). Thus,

the

affordances

detected

by

some perceivers

may

not be

detected

by

others.

For

example, per-

ceivers

of any

strength

may

perceive

the

structural invariant high physical power

in a

stimulus

person with

mature

facial

features.

But,

whether

that

stimulus person

is

per-

ceived

as

affording

defeat

in a

wrestling

match will depend upon

the

perceivers'

own

physical

powers. Similarly, perceivers

of any

status

may

perceive

the

structural invariant

low

social power

in a

stimulus person with

babyish

facial

features. However, whether

the

babyfaced

person

is

perceived

as

affording

compliance with

the

perceiver's

own

wishes

will

undoubtedly depend upon

the

perceiver's

own

social

powers.

6

It may

also depend upon

the

perceiver's social goals:

The

person

who

has

no

desire

to

elicit compliance

may not

perceive that

particular

affordance.

Although

the

correspondence between perceived

af-

fordances

and

one's

own

capabilities

and

goals

has not

been extensively studied,

one

intriguing

finding is the

report

that

prison

inmates convicted

of

assaultive crimes

are

highly

attuned

to

stimulus information

in a

person's gait that reveals

the

person's

"mug-

gability"—that

is, the

affordance

of

assault

(Grayson

&

Stein,

1981).

Although

we

have been illustrating

the in-

fluence

of a

perceiver's

own

action

potential

upon

the

detection

of

affordances,

it

should

be

noted

that

the

detection

of

structural

or

transformational

invariants

can

also

vary

across perceivers. Most notably, perceivers

may

be

blind

to

those structural

or

transfor-

mational

invariants

that

have

no

behavioral

utility

(i.e., that

do not

have

any

fordance).

We are all

familiar with

the ex-

ample

of

Eskimos

who

differentiate

several

varieties

of

snow.

In the

traditional Eskimo

life-style,

each snow structure undoubtedly

affords

different

activities.

For the

urban

dweller,

snow

of any

sort

affords

shoveling

and

slipping,

and the

different

structures

are

therefore

not

perceived. Blindness

to

percep-

tual information

specifying

structural

in-

variants

that have

low

behavioral utility

may

also

be

found

in the

social realm.

For ex-

ample, Gilson, Brown,

and

Daves (1982)

re-

cently

demonstrated that perceptions

are

sig-

nificantly

sexual preference. Using

a

binocular rivalry paradigm, these authors

found

that

gay men

tended

to

report

images

of

men

whereas straight

men

reported images

of

women.

The

perceiver's expectations,

as

well

as

behavioral

potential,

may

influence percep-

tual

attunements.

For

example, binocular

ri-

valry

research

has

demonstrated

that

people

more readily perceive photographic slides

whose

content

is

expected

by

virtue

of

being

drawn

from

their

own

culture than slides

drawn

from

an

unfamiliar culture (Bagby,

1957).

Similarly, people more readily per-

ceive

violent photographic slides

when

such

scenes

have become expected

by

virtue

of

police

training than when

no

such training

has

occurred (Toch

&

Schulte,

1961),

and

they

more readily perceived

by the

physician than

by

the lay

person,

not

only because this

in-

formation

is

more relevant

to the

physician's

behavioral

goals

but

also because extensive

experience

has

more

finely

tuned

the

physi-

cian's perceptual apparatus

to the

detection

of

these invariants.

A

physician

who

meets

on the

street

a man

with

red

discoloration

of the

cornea

and

notched

teeth

is

meeting

someone

who

openly displays

two of

Hutchinson's

signs

and

is

likely

to be

syphilitic.

Others

present, however,

medically

blind,

will

sec no

evil.

(Goffman,

1963,

p.

51)

Another

example

of

individual

differences

in

attunement

to

social invariants

is

provided

by

the

experiences

of the first

author

while

she

was

traveling through Europe. Several

people

whose acquaintance

she

made accu-

rately perceived

her

Jewish identity

from

her

appearance and/or demeanor.

As it

turned

out, these people were themselves Jewish,

and

they were

old

enough

to

have lived through

the

Holocaust.

One can

speculate that their

own

survival

required

a

keener attunement

to the

stimulus information that reveals

"Jewishness"

(see Savitz

&

Tomasson,

1959)

than

is

possessed

by the

average person whom

Jews

encounter

in the

United States. Given

the

likely

impact

of

perceptual experience

and

perceiver goals upon

the

detection

of in-

variants

in the

realm

of

social perception,

more research devoted

to

examining these

variables

should

be

enlightening.

Causal

Attribution

Most

of the

theory

and

research

on

peo-

ple's

causal

explanations

for

social

events

has

focused

upon

the

cognitive operations that

might

be

engaged

in by an

individual

in

order

to

infer

the

causes

of

reported

behaviors.

This

approach

to

causal attributions

is

certainly

a

useful

and

ecologically appropriate one.

Although

it has

been

argued

that

our

impres-

sions

of

others

are

more

often

than

not

based

upon

first-hand

perceptual information, this

is

probably

not the

case

for our

causal attri-

butions.

As

often

as we

ponder

the

causes

of

observed

behavior,

we

also ponder

the

causes

of

those behaviors that

we

learn

of

second

hand.

Although

the

focus

on

inferential pro-

cesses involved

in

causal attributions

is

well

placed,

we can

still lament

the

neglect

of

per-

ceptual processes.

Stimulus

information.

Michotte's

(1963)

work

has

demonstrated that there

is

stimulus

information

sufficient

for the

perception

of

physical

causality.

Yet

theories

and

research

SOCIAL

PERCEPTION

229

dealing with

the

attribution

of

causality

in

the

social domain have focused almost

ex-

clusively

upon inference. Research

that

in-

vestigates

the

stimulus information

specify-

ing

social causality

is

needed

to

redress this

imbalance.

The

ecological approach suggests

that

such

research

must allow perceivers

to

watch, listen

to,

and/or

interact

with

the

peo-

ple

for

whom they

will

be

making causal

at-

tributions.

In

this manner,

one can

ascertain

what information

in the

extensional prop-

erties

of a

person

or

persons

is

sufficient

for

making

a

given attribution. And,

one can

thus begin

to

describe

the

stimulus invariants

that

give rise

to the

perception

of

social cau-

sality just

as

Michotte

described

the

stimulus

invariants

that give

rise to the

perception

of

physical

causality.

Some

of

the

research

on

causal

attributions

is

consistent with

the

foregoing goals.

For

example,

Dix (in

press)

has

recently dem-

onstrated

that

the

concrete depiction

of

low-

consensus information

may

allow young chil-

dren

to

perceive personal causality before

they

have developed

the

cognitive sophisti-

cation

to

infer

such causality

via the

appli-

cation

of

logical schemata. Other research

(Bassili,

1976)

has

described

the

stimulus

in-

variants that give rise

to the

perception

of

chasing,

an

example

of

social causality. Sim-

ilarly,

Kassin, Lowe,

and

Gibbons

(1980)

have

identified perceptual information

that

yields

the

perception

of two

other

causes—

pushing

and

carrying.

Whereas

the

foregoing studies have

em-

ployed

either pen-and-ink sketches

or

ani-

mated geometric forms, other studies

em-

ploying

videotapes

of

real, interacting people

have

also taught

us

something

of

interest

re-

garding

the

stimulus information that

is

suf-

ficient

for

making

a

given attribution. When

we

watch

or

listen

to two

people

having

a

conversation,

the

person

who is for

some rea-

son

perceptually salient

is

seen

as

exerting

more causal influence than

the

person

who

is

less salient. This

effect

has

been demon-

strated

for a

wide range

of

salient attributes,

including

movement, bright lighting, louder

talking, boldly

patterned

clothing, novelty,

and

unit formation. (See

McArthur,

1981,

for

a

review

of

this

literature.)

Although

the im-

pact

of

salient stimulus

properties

upon per-

ceptions

of

causality

has

been amply dem-

onstrated,

the

question remains

as to why

this

occurs.

In

keeping with

the

inferential

ap-

proach

to

person perception,

a

number

of

cognitive

mediators have been proposed

and

tested (e.g.,

see

Fiske, Kenny,

&

Taylor,

1982).

On the

other hand, McArthur

(1980)

has

argued

that

the

ecological

approach

to

perception

may

suggest

a

more satisfactory

explanation

for the

tendency

to

perceive

sa-

lient

people

as

causal.

In the

context

of a

dynamic

social interaction, each person's

be-

havior

is

typically both cause

and

effect:

Per-

son A

reacts

to

Person

B and

that reaction

causes

a

reaction

in

Person

B. The

power

of

certain stimuli

to

draw attention

may

cause

the

perceiver

to

pick

up the

salient

person's

influence

on the

nonsalient person, rather

than

vice versa. Thus,

a

conversation between

a

soft

spoken person

and a

louder person

or

between

a

dimly

lit

person

and a

more

brightly

lit

person

may be

registered

in

units

reflecting

the

causal

influence

of the

louder

or

brighter actor

on the

quieter

or

dimmer

actor rather than

in

units reflecting

the re-

verse

causal

influence.

Although

no

research

has

successfully

assessed

the

perceptual

or-

ganization

of a

social interaction

involving

salient

and

nonsalient stimulus persons, there

is

evidence

to

indicate that such organization

is

responsive

to

salience manipulations

(Newtson,

Rindner, Miller,

&

LaCross,

1978)

and

that

it is

perceived causality

(Massad,

Hubbard,

&

Newtson, 1979). Such

evidence clearly warrants more research

de-

voted

to

investigating

the

impact

of

percep-

tual,

as

opposed

to

cognitive,

influences

upon

the

attribution

of

social causality.

An

Ecological Perspective

on the

Nature

of

Error

in

Social Perception

Most

current

thinking

in the

domains

of

social perception

and

cognition either takes

as

axiomatic

the

proposition

that

our

knowl-

edge

of the

social environment

is

highly

error

prone (e.g., Nisbett

&

Ross, 1980; Ross,

1977)

or

simply

ignores

the

accuracy prob-

lem

altogether

in

favor

of a

process analysis

that focuses

on the

cognitive operations

in-

tervening

between stimulus

and

response

(e.g., Anderson, 1974; Newtson, 1976).

Moreover, when

the

problem

is

discussed,

in-

ferential

processes

are

typically assigned

the

230

LESLIE

Z.

MCARTHUR

AND

REUBEN

M.

BARON

role

of

straightening

out the

noisy data

of

perception

as in

Taylor

and

Fiske's (1978)

adaptation

of the

Brunswikian model

of

per-

ception. Indeed, within mainstream social

cognition, putting

the

term perceived before

a

construct,

as in

perceived control

or

per-

ceived

crowding,

automatically

confers

a

sub-

jectivity

and

lack

of

trustworthiness

to

that

source

of

data.

In

sharp contrast

to

such views,

the

eco-

logical

perspective

emphasizes

the

essential

accuracy

of

perception-based

knowledge (e.g.,

Ittelson,

1973). And, because

J.

Gibson (1979)

strongly

argues that perception

of the

social

environment

is

likely

to

follow

the

same basic

principles

as

perception

of the

nonsocial

en-

vironment,

an

ecological model challenges

the

current view that social perceptions

are

more

often

flawed

than not. Before

attempt-

ing

to

reconcile these contrasting perspec-

tives,

some comment

is

needed regarding

Gibson's equation

of

social

and

nonsocial

perception, since

one

must certainly

ac-

knowledge

differences

between

the

two. Most

notably,

people, unlike

oak

trees,

do try to

deceive

perceivers. However, there

is

substan-

tial

evidence that perceivers

can

detect

de-

ception

in the

stimulus information that

the

deceiver projects (e.g., Runeson

&

Frykholm,

1982;

Zuckerman, DePaulo,

&

Rosenthal,

1981).

Moreover, when perceivers

fail

to de-

tect deception, their error

may

reflect

lack

of

sufficient

motivation

or

even collusion with

the

deceiver

rather

than

inadequacies

in

per-

ceptual

ability (cf.

Goffman,

1959). These

considerations bolster

the

ecological view

that

social perception,

like

nonsocial percep-

tion,

is

essentially accurate.

The

question

re-

mains

as to how we can

reconcile this per-

spective

with

the

more current

view.

The

Meaning

of

Error

and

Bias

At

the

outset,

it

should

be

made clear that

there appears

to be a

qualitative

difference

in

the

meaning

of

error

when

one

approaches

social

perception

from

an

ecological perspec-

tive

as

opposed

to an

inferential-constructive

one. Within

the

ecological perspective,

the

criterion

for

accuracy

is the

efficacy

of

one's

behavioral adjustments

to a

distal

object:

Error

occurs when one's knowledge

of the

world

does

not

permit adaptive action

(i.e.,

does

not

allow

one to

accomplish behavioral

goals). Within

an

inferential framework,

the

criterion

for

accuracy

is the

logicality

of

one's

reasoning processes,

and

error occurs when

one

manifests problems

in

reasoning. Within

the

latter framework, error

and

bias

are

syn-

onymous. However, within

an

ecological

framework,

bias

is

different

from

error: Bias

is

simply

a

matter

of

selective attention

and

action,

and

whether

a

given bias leads

to

error

in

adaptive behavior

is an

empirical,

not a

logical, problem.

Sins

of

Omission

Although

consistently arguing

for the

gen-

eral veridicality

of

perception,

J.

Gibson

(1966)

leaves room

for

certain kinds

of

error

based

both

on

deficiencies

in the

information

available

at a

given

time

as

well

as on

the

inherent

selectivity

in any

perceiver's pickup

of

information. Thus, Gibson states:

In

an

eventful

environment with sights

and

sounds

and

smells

and

touches

all

around,

the

individual

cannot

reg-

ister

everything

at

once,

and his

perception must there-

fore

be

selective

. . .

What

the

object really

affords

may

be

missed

and

what

the

observer perceives

it as

affording

may

be

mistaken,

(p.

309)

Missing what

an

object really

affords

may be

called

an

error

of

omission. However,

from

an

ecological viewpoint such

as

that espoused

by

Shaw, Turvey,

and

Mace (1982),

to

label

all

such omissions

as

error

is

wrong because

such labeling ignores

the

adaptive specificity

of

the

perceptual system. More

specifically,

they argue

that

the

essential accuracy

of

per-

ception rests upon

the

ability

of

animals

to

be

sensitive

to

adaptively relevant informa-

tion

as

opposed

to all

possible information.

Thus, these theorists would

not

want

to

call

a

bat's

perception

in

error when

it

misses

purely

visual information,

nor

would they

say

that humans' perceptions

are in

error

when

they miss ultraviolet information

or

high-frequency

auditory information.

When omissions

occur

at the

species

level,

most would probably agree that

it is

dubious

to

even

refer

to

them

as

errors. However,

the

issue

within species

is

more controversial.

For

example,

if an

expert

in a

given domain sees

more than

a

novice,

do we

want

to

call

the

novice's perceptions

in

error?

If

the

Eskimos

see

more

varieties

of

snow than

we do, are

SOCIAL

PERCEPTION

231

our

perceptions

in

error?

If we use as our

criterion

for

accurate perception

the

detec-

tion

of

affordances

that

are

essential

to

adap-

tive

action, then

we

probably would

not

want

to

call even these omissions

"errors."

Of

course, there

are

omissions

in

which perceiv-

ers

fail

to

detect affordances essential

to

their

own

actions. Some

of

these would have

to be

labeled

'error'

within

an

ecological perspec-

tive,

for

example,

failing

to

perceive

an

attack

dog

as

vicious. However, even some

mal-

adaptive

omissions would

not be

labeled

er-

ror

within

the

ecological approach

if

they

re-

sulted

from

impoverished stimulus infor-

mation.

The

failure

to

detect

an

affordance

may

often

result

from

an

artificial

reduction

in the

information

available

to the

perceiver. Some-

times

the

reduction

in

information derives

from

static, temporally truncated,

or

other-

wise

nonrepresentative stimulus displays that

have

been constructed

in the

interests

of ex-

perimental purity

as

opposed

to

ecological

validity.

At

other

times,

it

derives from

re-

stricting

the

perceiver's ability

to

actively

ex-

plore

the

available stimulus information.

Certainly

the

veridicality

of

oner's

percep-

tions

is

typically enhanced

as

more complete

stimulus

information

is

provided.

In the so-

cial

domain,

for

example, accurate identifi-

cation

of

personal identity, gender,

and

type

of

affect

from

Johannson-type

point-light

displays

is

possible

only when

dynamic,

as

opposed

to

static, displays

are

used.

7

Other

research indicates that

the

perception

of

higher

order social properties,

such

as ma-

levolent

and

benevolent intent, requires

not

only

dynamic information,

but

also rela-

tional

information such

as

that provided

by

the

joint spatial

and

temporal trajectories

of

the

entities involved (Bassili,

1976).

Thus,

information

from

an

interpersonal behavior

sequence

is

likely

to

yield more complete

and

accurate perceptions than information about

the

behavior

of a

single actor

at a

single point

in

time.

Just

as

impoverished information

can re-

sult

from

nondynamic stimuli,

it can

also

result

from

an

inactive perceiver. Consider

for

example

the

fact

that touch comes

to the

aid

of

vision when

a

stick

in the

water that

looks

bent

feels

straight. Similarly,

a

closer

look

corrects

a

more distant one,

and a mi-

rage

disappears. Within

the

social domain,

being

able

to

interact with another person

as

well

as to

passively watch

and

listen

to him

or her may

also dispel

a

number

of

errors.

Because

an

intermodal

and

temporally

ex-

tended

sampling

of

environmental stimulus

information

is the

natural

condition

for ef-

fective

functioning

of the

perceptual systems,

omissions that occur when such sampling

is

precluded

are not

viewed

as

error

in the

eco-

logical

approach

to

perception.

Sins

ofComission

J.

Gibson,

as

quoted before, suggests that

what

an

object

is

perceived

as

affording

may

sometimes

be

mistaken,

an

event that

can be

called

an

error

of

comission. However,

from

a

radical ecological viewpoint,

it is

inappro-

priate

to

label

all

such mistakes

as

error

in

the

sense that many

of

them

may

have

no

bearing

on

adaptive

actions.

Consider

the fol-

lowing

example: Based

on the

fact

that

the

prey

of

sharks normally produce

a

charac-

teristic

electric

field, it is

possible

to

lure,

a

shark

to

strike

at an

artificial electric

field as

if

it

were prey.

If

edibility

and

certain

electric

field

properties

are

correlated

in the

shark's

natural niche,

is the

shark really

in

error?

Shouldn't

the

criterion

for

error

be the

nat-

ural,

as

opposed

to the

experimental, struc-

ture

of

reality, because

the

perceptual systems

are

attuned

to

natural contingencies?

Of

course,

there

are

comissions

in

naturalistic

settings

and one

would probably want

to

call

these

"errors."

However, there

is an

impor-

tant

difference

between

an

ecological per-

spective

on

such errors

and

more traditional

views.

From

the

ecological perspective,

the

nature

of

these errors should

not be

random.

Rather, such errors should

be

rooted

in

strat-

egies

of

information

pickup

and/or

attune-

ment

to

particular invariants that usually

serve

an

adaptive function

for the

perceiver.

Thus, what

we

label

"errors

of

comission"

are

likely

to be

overgeneralizations

of

highly

adaptive perceptual attunements.

The

phe-

7

Needless

to

say,

the

point-light

research

does

not

provide

complete

stimulus

information.

It is

cited

here

only

to

demonstrate

that

with

increasing

stimulus

in-

formation (e.g., dynamic

as

opposed

to

static

displays),

there will

be

increased

veridicality

of

perceptions.

232

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

nomena

of

illusory causation

and

illusory

correlation that have received considerable

attention within

the field of

social perception

may

be

instances

of

such overgeneralization.

Illusory

causation.

The

tendency

to at-

tribute causal

influence

to

perceptually

sa-

lient persons

has

been called

the

illusory

cau-

sation

effect

(McArthur,

1980).

It

seems plau-

sible

to

suggest

that

such

"errors"

may

result

from

the

overgeneralization

of a

highly

adap-

tive

perceptual attunement. More

specifi-

cally,

it may be

that perceptually salient stim-

uli

are

actually more likely than nonsalient

stimuli

to

exert causal

influence

in the en-

vironment,

and our

perceptual apparatus

may

have evolved

to

process information

in

a

manner that

is

maximally sensitive

to

this

reality.

This proposal gains credence when

one

considers

the

causal

efficacy

of

percep-

tually

salient stimuli

in the

natural environ-

ment:

Bright lights, such

as

lightening

or fires,

and

loud sounds, such

as

thunder,

a

roaring

animal,

or a

screaming baby,

are

more

apt

to

exert causal influence than their less

in-

tense

counterparts. Similarly, moving stim-

uli,

such

as a

charging bull,

and

unit-forming

stimuli,

such

as a

herd

of

buffalo,

are

more

apt to

exert causal influence than

a

stationary

or

unrelated collection

of

animals.

If our

per-

ceptual

systems were

not

more attuned

to the

causal

influences

of

salient stimuli

on

non-

salient

stimuli then vice versa, then

we

might

not

live

long enough

to

detect causal invari-

ants through more "objective" registration

of

the

information available.

Illusory

correlation.

Another error

of so-

cial

perception that

may be

rooted

in

adap-

tive

perceptual functioning

is

illusory

corre-

lation:

People's perceptions

of the

correlation

between

events tend

to be

unduly

influenced

by

certain event pairs. These

effects

have been

demonstrated both

in the

realm

of

nonsocial

perception (e.g., Chapman, 1967)

and in the

realm

of

person perception, where

the

illu-

sory

correlation concerns

an

actor-behavior

link.

More

specifically,

certain types

of be-

havior

are

perceived

to be

more representa-

tive

of

certain categories

of

people than they

really

are

(e.g., Chapman

&

Chapman,

1967;

Hamilton

&

Gifford,

1976; McArthur

&

Friedman, 1980). Within

the

framework

of

an

ecological approach

to

perception,

the in-

teresting question posed

by the

phenomenon

of

illusory correlation

is

"What

strategies

of

information

pickup and/or what kinds

of co-

occurrences will lead

the

perceiver

to

detect

an

invariant relationship (i.e.,

a

correlation)

that

is not

present?"

Since

perceivers cannot possibly process

all

available information,

it

seems adaptive

for

the

perceptual systems

to be

geared

to

pick

up

information that

is the

most ecolog-

ically

significant.

This might include

infor-

mation

from

stimuli that

are

intense,

un-

usual,

rare,

or

aversive, because events

in-

volving

stimuli with these characteristics

are

more

apt to

require adaptive responding than

events

involving more mundane stimuli.

In

view

of

this argument

it is

significant

that

research reveals greater visual

fixation of so-

cial

stimuli that

are

extreme (i.e., intense

and

atypical),

infrequent,

or

negative

(Fiske,

1980;

McArthur

&

Ginsburg,

1981).

Moreover,

these

are the

very stimuli that tend

to be in-

volved

in

illusory correlation

effects

(Ham-

ilton

&

Gifford,

1976;

Rothbart,

Fulero, Jen-

son,

Howard,

&

Birrell,

1978).

It

thus appears

that illusory correlation

effects

may

derive

from

the

selective registration

of

information

that

it is

particularly important

for

perceivers

to

detect, such

as

unfamiliar

people perform-

ing

unusual, negative,

or

intense acts.

Illusory

correlation

may

also derive

from

the

selective registration

of

information that

is

typical

or

representative

of the

natural

en-

vironment.

A

perceptual system attuned

to

typical

correlations would

be

more

efficient

than

one

that discovered each correlation

anew.

Thus,

for

example, large objects (in-

cluding

people)

may be

perceived

as

affording

more

physical resistance, more danger,

or

even

more noise than small ones, because this

is

typically

the

case.

8

This example illustrates

8

A

similar explanation

may be

offered

for the

per-

ceived correlation between physical

attractiveness

and

social

and

intellectual competencies

(e.g.,

Berscheid

&

Walster,

1974).

Specifically,

this

illusory

correlation

may

derive

from

a

veridical

correlation

between

the

unat-

tractiveness

that results

from

various

physical

abnor-

malities

and the

corresponding

disorders

of

social

and

intellectual functioning.

The

perceived

correlation

be-

tween

physical attractiveness

and

honesty (e.g., Berry

&

McArthur,

Note

2) may

also

have

its

roots

in a

verid-

ical

correlation.

It has

been found

that

when

people

are

telling

the

truth,

they look

more

attractive

than

when

they

are

lying

(Krauss,

1981).

SOCIAL PERCEPTION

233

the

possibility that illusory correlations

in the

realm

of

person perception

may

derive

from

attunement

to

co-occurrences whose adap-

tive significance

is

grounded

in the

physical

realm.

It

also illustrates

the

perception

of

cross-modal correlations,

a

phenomenon that

has

been

well

documented

by

Marks

(1978).

Indeed, even newborn infants

associate

par-

ticular

auditory stimuli with particular visual

ones (Born, Spelke,

&

Prather,

Note

4),

which

indicates that perceiving certain kinds

of in-

formation

as

correlated

is not

dependent

upon extensive learning about natural cor-

respondences,

but

rather

is a

fundamental

perceptual

preparedness.

Self-fulfilling

prophecies.

A

third

error

of

commission

that warrants discussion

in

con-

sidering

an

ecological approach

to

social per-

ception

is

that

of

self-fulfilling

prophecies.

Such

effects

fall

into

two

general categories:

(a)

eliciting

from

others

the

behaviors

we ex-

pect

to

perceive,

and (b)

perceiving

in

others

the

behaviors

and

traits

we

expect

to

perceive.

Although

these

findings

seem problematic

for

a

theory that emphasizes

the

essential accu-

racy

of

social perceptions,

a

closer analysis

reveals

that they

may be

reconciled

with

the

ecological approach.

First,

the

research documenting

a

tendency

for

perceivers

to

elicit

the

very

behavior they

expect

to

perceive provides

no

real problem

for

an

ecological theory

of

perception.

In

these

cases,

the

perceiver detects properties

in

the

target person that

are

actually mani-

fested

in

behavior. Indeed,

the

accuracy

of

these perceptions

is

often

affirmed

by

ratings

of

the

target person's behavior

by

blind judges

who

have

no

"prophecy"

(e.g., Snyder,

Tanke,

&

Berscheid, 1977). Thus,

the

"error"

in

these

self-fulfilling

prophecy

effects

is not

misperception

but

rather misguided

action—

that

is,

action that

has not

been

sufficiently

informed

by

perception.

A

second type

of

self-fulfilling

prophecy,

perhaps better labeled

a

self-confirming

prophecy,

is

more problematic

for the

eco-

logical position.

This

is the

tendency

for

per-

ceivers

to

detect

in

other people those prop-

erties that they expect

to find,

such that very

different

properties

may be

detected

in the

same behavioral information

by

perceivers

with

different

expectations (e.g.,

Kelley,

1950;

Langer

&

Abelson,

1974;MassadetaL,

1979;

Snyder

&

Frankel,

1976).

Like illusory cor-

relation

and

illusory

causation, such

effects

may

reflect

in

part

the

overgeneralization

of

an

adaptive

perceptual attunement:

To the

extent

that

our

expectations

are

more

often

right

than wrong,

it may be

functional

to be

particularly

sensitive

to

confirmatory stim-

ulus

information.

These

effects

may

also

be

a

reflection

of

ambiguities

in the

stimulus

information.

More

specifically,

they

may oc-

cur in

situations

in

which

the

behavioral

in-

formation

has no

underlying

structural

in-

variant—that

is,

there

are no

properties that

remain

the

same under changes

in the

sam-

pling

of

behavior across space

or

time.

If so,

then

two

observers, taking

different

samples

of

behavior, would

not

necessarily detect

the

same

underlying property.

It

follows

from

this

argument

that

self-confirming

prophecies

should

not

occur

in the

case

of

attributes

that

are

indeed invariant. Although researchers

have

begun

to

question whether people have

dispositional

invariants (e.g.,

Mischel,

1968),

some people

do

seem

to

have some

traits

(e.g.,

Bern

&

Allen,

1974).

In

these cases,

one

would

expect that

different

samples

of the

same person's behavior would tend

to

reveal

the

same invariant

disposition.

One

would

also expect that

the

detection

of

this invariant

disposition could

not be

overpowered

by er-

roneous

initial prophecies

of a

self-confirm-

ing

nature.

In

addition

to the

possibility that some

people

will

provide information

for

struc-

tural

invariants that

is

immune

to

self-ful-

filling

prophecy

effects,

it is

also likely that

there exist transformational invariants

and

affordances

that

are

immune

to

such

effects.

More

specifically,

invariant properties whose

detection serves important adaptive

actions,

such

as the

transformational invariant

fight-

ing

or the

affordance

danger,

may be so

strongly

manifested

in the

stimulus

infor-

mation that people project that they could

not be

overpowered

by

erroneous prophecies.

Some evidence

for

this assertion

is

provided

in a

recent study

by

Woll

and

Martinez

(1982),

who

found that biasing labels

for a

pictured facial expression influenced recog-

nition memory when

the

expression depicted

a

pleasant

or

neutral emotion,

but not

when

it

depicted

an

unpleasant emotion. Insofar

as the

accurate detection

of

unpleasant

emo-

234

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

tions

has

greater adaptive

significance

than

the

accurate

detection

of

neutral

or

pleasant

ones,

this

finding is

consistent with

the

eco-

logical

view.

So is the finding

that

the

influ-

ence

of

biasing

labels upon recognition

of any

of

the

emotions

was

limited

to the

condition

in

which there

was a

15-minute

delay

be-

tween

the

original presentation

of the

emo-

tional expressions

and the

recognition

task.

In

a

1-minute

delay condition, where per-

ceptual

information would

be

more salient,

the

labels

had no

impact

on

recognition.

Although

the

ecological approach must

acknowledge

errors

in

social perception such

as

illusory

causation, illusory correlation,

and

self-fulfilling

prophecy,

it

incorporates

such

errors into

a

theoretical framework that

argues

for the

general veridicality

of

percep-

tion.

In

particular,

it

suggests that

errors

of

comission

may

reflect

the

overgeneralization

of

highly

adaptive perceptual attunements.

Conclusions

The

ecological approach

to

social percep-

tion

provides

an

innovative conception

of

what

is

perceived,

how it is

perceived,

and

who

perceives

it.

What

is

perceived

is first

and

foremost what perceivers need

to

per-

ceive

for

adaptive interaction with their

en-

vironment.

Thus,

the

ecological approach

assumes

that perception

is by and

large

ve-

ridical.

It

also assumes that

we

often

perceive

affordances

as

opposed

to the

isolated struc-

tural

properties that have traditionally been

studied.

How

these affordances

are

perceived

is

through

the

dynamic, multimodal

infor-

mation

in

events. This

is

necessitated

by the

fact

that

affordances

are

typically complex

properties that have

no

one-to-one

connec-

tion

to the

static, stimulus elements that

are

provided

to

perceivers

in

traditional research

paradigms.

Who

perceives various

afford-

ances

are

those

for

whom these properties

are

behaviorally

relevant. Perceptual altune-

ments

thus

vary

not

only

as a

function

of the

cognitive

factors that

are

studied

in

tradi-

tional

approaches,

but

also with perceptual

experience, behavioral

capabilities,

and on-

going

behavior.

The

advantages

of

social perception

re-

search

within

the

ecological framework

are

numerous.

The

emphasis

on

information

in

the

stimulus provides

a

needed balance

to the

current emphasis

on

constructive

processes

in

the

perceiver.

The

emphasis

on

dynamic

stimulus

displays provides

a

needed balance

to

research that examines perceptions

of

peo-

ple

who are

neither seen

nor

heard.

The em-

phasis

on the

dynamic relationship between

perception

and

action directs attention

to the

rather neglected question

of

social

percep-

tions

within ongoing relationships, where

perceptions

are

informed

by

actions

and

where people really

do

have

the

opportunity

to

perceive

one

another's invariant attributes.

The

emphasis

on

perceived

affordances

pro-

vides

a

vital alternative

to the

trait analysis

of

social perception. And, more generally,

the

emphasis upon

the

adaptive function

of

per-

ception places formerly disconnected

re-

search issues

into

an

integrated

conceptual

framework

that generates

new

questions

and

promises

to

greatly enrich

our

understanding

of

social perception.

As

the

foregoing conclusions reveal,

the

ecological

approach

to

social perception dif-

fers

in a

number

of

respects

from

traditional

information-processing

models.

An

addi-

tional, metatheoretical

difference

between

the

two

approaches warrants some explicit

discussion. Whereas

the

ecological

approach

views

perception

as the act of

picking

up in-

variant

information

from

the

environment,

the

information-processing approach views

perception

as a

process

of

inference

that must

be

elucidated

by

specifying

the

cognitive pro-

cesses intervening between sensory inputs

and the

perceptual outcome.

This

difference

between

the

information-processing

and

eco-

logical

approaches

is

perhaps best exempli-

fied

by

their divergent

perspectives

regarding

the

meaning

and

utility

of the

schema con-

struct.

From

the

ecological perspective,

the

assumption

of an

internal,

reified

state

that

causally

mediates meaning

is

neither neces-

sary

nor

desirable. Instead

of

postulating

such

a

representational structure,

the

ecolog-

ical

approach simply assumes

that

the

past

history

of

one's interaction with

the

environ-

ment

consistently retunes

the

perceptual

ap-

paratus

on an

online basis.

Thus,

the

"ghost

in

the

machine" becomes superfluous. (See

Baron,

1980,

for a

more extended discussion

of

logical problems with using

the

schema

construct.)

Furthermore,

a

close look

at the

SOCIAL

PERCEPTION

235

properties

of

events suggests

that

they serve

many

of the

same epistemic

functions

as

schemata.

Both

schemata

and

events

are

abstract,

global,

and

generalized structures involving

higher

order relationships that hold over par-

ticulars. Indeed,

at the

conceptual level

the

major

difference

between these constructs

is

whether

one

assumes that such abstract

knowledge

or

information structures exist

solely

in

one's head

or

have counterparts

in

the

world

as

events.

For

example,

is the

causal

efficacy

of

personal

effort

an

event

to be de-

tected

or the

creation

of a

schema? Similarly,

are

higher social relationships, such

as de-

pendency, events

to be

perceived given

a

properly

attuned observer

or are

they

the

products

of a

higher level constructive pro-

cess?

The

ecological view

is

that most,

if not

all,

adaptively relevant properties

are

exten-

sionally

projected. However,

it is

acknowl-

edged

that

the

detection

of

some

of

these

properties requires extensive perceptual

learning and,

in the

absence

of

such learning,

may

require inference. Thus,

it is

possible—

even

likely—that

there

are

both internal

and

external sources

of

organization

and

struc-

ture. Still,

we

would argue that

the

internal

structures must

be

based upon

the

external

ones.

As

such,

to

fully

understand schemata

requires

an

understanding

of the

stimulus

in-

formation

that

is out

there serving

as

grist

for

the

schema mill.

There

are

several types

of

evidence relevant

to

differentiating

the

direct detection

of

stim-

ulus

information

from

constructive, infer-

ential processes.

Before

enumerating these,

it

is

important

to

state

that direct perception

basically

refers

to the

sufficiency

of the

per-

ceptual apparatus

to

extract certain environ-

mental meanings (e.g.,

affordances)

without

the

intervention

of

higher order cognitive

operations. Directness

in

this view does

not

mean directness

in the

phenomenological

sense

of the

immediacy

of the

perceptual

ex-

perience, although this

is

often

a

concomitant

of

direct

perception.

The

most basic evidence

for the

direct

de-

tection

of

functional information indepen-

dent

of

constructive cognitive operations

is

provided

by

research

that

demonstrates

the

ability

of

organisms with very limited cog-

nitive abilities (e.g., young infants

and

infra-

humans)

to

detect

relatively

abstract,

cross-

modal invariances

in

stimulus information.

(See

Born, Spelke,

&

Prather, Note

4, and

also

E.

Gibson

&

Spelke,

in

press,

for a

com-

prehensive

review

of the

infant

literature

on

this

topic.)

A

second source

of

relevant evi-

dence

is

naturalistic

and

experimental situ-

ations that demonstrate people's

ability

to

preserve

the

essential characteristics

of

rap-

idly

changing stimulus displays

in

their

online

responses.

For

example,

the

tennis player

at

the

net is

able

to

adroitly adjust

his or her

racket

to the

trajectory

of a

speeding

ball.

9

Similarly,

in the

social realm, people adroitly

adjust

their

own

nonverbal responses

to

rapid

changes

in

such responses

by

those with

whom

they interact (e.g.,

the

"eyebrow

flash",

Eibl-Eibesfeldt,

1975).

In

these examples,

perceiving

and

responding

are

instanta-

neously

coordinated,

and to

posit intervening

cognitive

processes seems gratuitous.

A

third

type

of

evidence

for the

sufficiency

of the

perceptual apparatus

to

extract environmen-

tal

meanings

is

demonstrations that changes

in

those meanings

are

closely,

if not

perfectly,

tied

to

changes

in the

available stimulus

in-

formation.

For

example,

in

research

on the

perception

of

aging, Shaw

and

Pittenger

(1977)

have argued

for a

psychophysical

re-

lationship between

the

objective, physical

changes that characterize

the

aging

face

and

the

perception

of

aging:

As

certain systematic

changes

occur

in the

cranium

and the

jawline

of

a

face

(the cardiodal strain transforma-

tion),

so

will

there

be

systematic changes

in

the

perceived

age of

that

face.

In

sum,

the

direct detection

of

meaning

in

objective stim-

ulus

information,

as

opposed

to the

subjec-

tive

construction

of

meaning,

is

evidenced

by

(a)

adaptive responses

to

stimulus informa-

tion

by

cognitively limited organisms,

(b)

adaptive, online responses

to

stimulus

infor-

mation that

is

rapidly changing,

and (c) re-

9

The

online coordination

of

seeing

and

doing speaks

to the

directness

of

perception

in at

least

two

ways:

(a)

The

joint

locomotor

movements

of the

perceiver

and the

target modulate

the

visual

flow

field

in

ways that help

specify

the

location

of the

target

(J.

Gibson,

1966;

Turvey,

1977),

and (b) the

requirement

for

both

speed

and ac-

curacy

of

responding suggests that

the fine-grained

con-

trol

of

action occurs

at

lower levels

of

brain functioning

that

do not

involve

the

higher order cognitive mechanism

of the

executive (Turvey, 1977).

236

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

sponses

that vary directly with

specifiable

changes

in the

objective stimulus

informa-

tion.

Whether

or not

readers accept

all of the

specific

suppositions

of the

ecological frame-

work,

we

hope that they

now do

accept

the

distinct

and

important role

of

perception

in

social

knowing—a

role that goes

far

beyond

a

primitive,

first-state,

registration

of

inputs.

And

we

hope

that

at

least some readers

will

be

challenged

to

pursue research

that

can

explicate

the

nature

of

social stimulus

infor-

mation

and

social perception. Perceptual

in-

formation

is

sometimes

sufficient

for

social

knowledge,

sometimes

it is

only

the

necessary

datum

for

social

inference

processes.

In

either

case,

we

cannot hope

to

fully

understand

so-

cial

cognition without understanding

the

per-

ceptual

information

on

which

it is

based:

"The idea exists only

by

virtue

of the

form."

Reference

Notes

1.

McArthur,

L.

Z,,

&

Apatow,

K.

Impressions

ofbaby-

Jaced

males.

Unpublished

manuscript,

Brandeis

Uni-

versity,

1983.

2.

Berry,

D., &

McArthur,

L. Z.

Impressions

of

college

males

rated

as

baby-faced versus mature-faced.

Manuscript

in

preparation, Brandeis University, 1983.

3.

Zajonc,

R. B., &

Markus,

H.

Affect

and

cognition:

The

hard

interface.

Unpublished manuscript, Uni-

versity

of

Michigan, 1981.

4.

Born,

W.

S.,

Spelkc,

E. S., &

Prather,

P.

Detection

of

auditory-visual relationships

by

newborn

infants.

Pa-

per

presented

at the

International

Conference

on In-

fant

Studies,

Austin, Texas, March 1982.

References

Alley,

T. R.

Caregiving

and the

perception

of

matura-

tional

status.

Unpublished

doctoral

dissertation, Uni-

versity

of

Connecticut,

1981.

Anderson,

N. H.

Cognitive algebra: Integration theory

applied

to

social attribution.

In L.

Berkowitz

(Ed.),

Advances

in

experimental social

psychology

(Vol.

7).

New

York: Academic Press, 1974.

Andrews,

R. J. The

origins

of

facial

expressions.

Sci-

entific

American, 1965,

213,

88-94.

Archer,

D.,

&

Akert,

R.

Words

and

everything

else:

Ver-

bal

and

nonverbal cues

in

social interpretation. Jour-

nal of

Personality

and

Social

Psychology,

1977,

35,

443-449.

Asch,

S. E.

Social psychology.

Englewood

Cliffs,

N.J.:

Prentice Hall, 1952.

Averill,

J.

Anger

and

aggression:

An

essay

on

emotion.

New

York:

Springer-Verlag,

1983.

Bagby,

J. W. A

cross-cultural study

of

perceptual pre-

dominance

in

binocular

rivalry.

Journal

of

Abnormal

and

Social

Psychology,

1957,

54,

331-334.

Baron,

R. M.

Social knowing

from

an

ecological

event

perspective:

A

consideration

of the

relative

domains

of

power

for

cognitive

and

perceptual modes

of

know-

ing.

In J.

Harvey

(Ed.),

Cognition,

social behavior,

and

the

environment.

Hillsdale,

N.J.:

Erlbaum,

1980.

Bassili,

J. N.

Temporal

and

spatial

contingencies

in the

perception

of

social

events.

Journal

uf

Personality

and

Social

Psychology,

1976,

33,

680-685.

Bassili,

J. N.

Facial motion

in the

perception

of

faces

and of

emotional expression. Journal

of

Experimental

Psychology:

Human

Perception

and

Performance,

1978,

4,

373-379.

Bassili,

J. N.

Emotion recognition.

The

role

of

facial

movement

and the

relative importance

of

upper

and

lower

areas

of the

face.

Journal

of

Personality

and

Social

Psychology,

1979,

37,

2049-2058.

Bern,

D.

J.,

&

Allen,

A. On

predicting

some

of

the

people

some

of the

time:

The

search

for

cross-situational

con-

sistencies

in

behavior. Psychological Review, 1974,

81,

506-520.

Berscheid,

E.,

&

Walster,

E.

Physical attractiveness.

In

L.

Berkowitz

(Ed.),

Advances

in

experimental social

psychology

(Vol.

7).

New

York:

Academic

Press,

1974.

Buck,

R.,

Baron,

R.,

&

Barrette,

D.

Temporal organi-

zation

of

spontaneous emotional expression:

A

seg-

mentation analysis. Journal

of

Personality

and

Social

Psychology,

1982,

42,

506-517.

Chapman,

L. J.

Illusory correlation

in

observational

re-

port. Journal

of

Verbal

Learning

and

Verbal

Behavior,

1967,

6,

151-J55.

Chapman,

L.

J.,

&

Chapman,

J. P.

Genesis

of

popular

but

erroneous psychodiagnostic observations. Journal

of

Abnormal

Psychology,

1967,

72,

193-204.

Cutting,

J. E.

Generation

of

synthetic

male

and

female

walkers

through manipulation

of a

biomechanical

in-

variant. Perception, 1978,

7,

393-405.

Cutting,

J.

E.,

Proffitt,

D.

R.,

&

Kozlowski,

L. T. A

bio-

mechanical invariant

for

gait perception.

Journal

of

Experimental

Psychology:

Human

Perception

and

Per-

formance, 1978,

4,

357-372.

Darwin,

C.

Expression

of

the

emotions

in man and an-

imals. London: Murray, 1872.

Dix,

T. H.

Inference complexity

and

informational

sa-

lience

in the

development

of a

consensus-based causal

attribution.

Child Development,

in

press.

Eibl-Eibesfeldt,

I.

Ethology:

The

biology

of

behavior.

New

York:

Holt,

Rinchart,

&

Winston, 1975.

Ekman,

P.

Universals

and

cultural

differences

in

facial

expressions

of

emotion. Nebraska Symposium

on

Motivation,

1971,

18,

207-283.

Engel,

E. The

role

of

content

in

binocular

resolution.

American Journal

of

Psychology,

1956,

69,

87-91.

Erdelyi,

M. H. A new

look

at the new

look: Perceptual

defense

and

vigilance. Psychological Review,

1974,

81,

1-25.

Fcldman,

R.

S.,

&

Donohue,

L.

F.

Nonverbal commu-

nication

of

affect

in

interracial dyads. Journal

of

Ed-

ucational

Psychology,

1978,

70,

979-987.

Fiske,

S. T.

Attention

and

weight

in

person

perception:

The

impact

of

negative

and

extreme behavior. Journal

of

Personality

and

Social

Psychology,

1980,

38,

889-

908.

Fiske,

S.

T.,

Kenny,

D.

A.,

&

Taylor,

S. E.

Structural

models

for the

mediation

of

salience

effects

on

attri-

bution. Journal

of

Experimental Social

Psychology,

1982,

18.

105-127.

SOCIAL

PERCEPTION

237

Gibson,

E. J.

Principles

of

perceptual

learning

and de-

velopment.

New

York:

Appleton-Century-Crofts,

1969.

Gibson,

E. J. The

concept

of

aftbrdances

in

development:

The

renascence

of

functionalism.

In W. A.

Collins

(Ed.).

Minnesota Symposia

on

Child

Psychology

(Vol.

15).

Hillsdale, N.J.:

Erlbaum,

1982.

Gibson,

E.

J.,

&

Spelke,

E. The

development

of

percep-

tion.

In J. H.

Flavell

& E.

Markman

(Eds.),

Cognitive

development.

In P. H.

Mussen

(Ed.),

Handbook

of

child

psychology

(Vol.

3). New

York: Wiley,

in

press.

Gibson,

J. J. The

senses considered

as

perceptual sys-

tems. Boston: Houghton

Mifflin,

1966.

Gibson,

J. J. The

ecological approach

to

visual

percep-

tion. Boston: Houghton

Mifflin,

1979.

Gilson,

M.,

Brown,

E.

C,

&

Daves,

W. F.

Sexual ori-

entation

as

measured

by

perceptual dominance

in

binocular rivalry. Personality

and

Social Psychology

Bulletin, 1982,

8,

494-500.

Gitter,

A. G.,

Black,

H.,

&

Mostofsky,

D.

Race

and sex

in

the

communication

of

emotion. Journal

of

Social

Psychology.

1972,

88,

273-276.

Goffman,

E. The

presentation

of

self

in

everyday

life.

Garden

City,

New

York:

Doublcday,

1959.

Goffman,

E.

Stigma: Notes

on the

management

of

spoiled

identity.

Englewood

Cliffs,

N.J.: Prentice-Hall, 1963.

Goldiamond,

I.

Indicators

of

perception:

I.

Subliminal

perception, subception, unconscious perception:

An

analysis

in

terms

of

psychophysical

indicator meth-

odology.

Psychological

Bulletin,

1958,

55,

373-412.

Grayson,

B., &

Stein,

M. I.

Attracting assualt: Victims'

nonverbal

cues. Journal

of

Communication,

1981,

31,

68-75.

Guthrie,

R. E.

Evolution

of

human threat display organs.

In

T.

Dobzhansky,

M. K.

Hecht,

& W. C.

Steere

(Eds.),

Evolutionary

Biology (Vol.

4). New

York:

Appleton-

Century

Crofts, 1970.

Hamilton,

D. L., &

Gifford,

R. K.

Illusory correlation

in

interpersonal perception:

A

cognitive basis

of

ste-

reotypic

judgments. Journal

of

Experimental

Social

Psychology,

1976,

12,

392-407.

Hastorf,

A.

H.,

&

Myro,

G. The

effect

of

meaning

on

binocular

rivalry.

American Journal

of

Psychology,

1959,

72,

393-400.

Heider,

F. The

psychology

of

interpersonal

relations.

New

York:

Wiley,

1958.

Heider,

F.,

&

Simmel,

M. An

experimental study

of ap-

parent behavior. American

Journal

of

Psychology,

1944,

57,

243-259.

Henley,

N. M.

Body politics. Englewood

Cliffs,

N.J.:

Prentice-Hall, 1977.

Ittelson,

W. H.

Environment perception

and

contem-

porary

perceptual

theory.

In W. H.

Ittelson

(Ed.),

En-

vironment

and

cognition.

New

York: Seminar Press,

1973.

Johansson,

G.

Visual perception

of

biological motion

and a

model

for its

analysis. Perception

and

Psycho-

physics, 1973,

14,

201-211.

Kassin,

S.

M.,

Lowe,

C.

A.,

&

Gibbons,

F. X.

Children's

use

of the

discounting principle:

A

perceptual

ap-

proach. Journal

of

Personality

and

Social

Psychology,

1980,39,

719-728.

Keating,

C.

F,

Mazur,

A.,

&

Segall,

M. H. A

cross-cul-

tural

exploration

of

physiognomic traits

of

dominance

and

happiness. Ethology

and

Sociobiologv,

1981,

2,

41-48.

Kelley,

H. H. The

warm-cold variable

in first

impres-

sions

of

persons.

Journal

of

Personality,

1950,

75,431-

439.

Knapp,

M. L.

Essentials

of

non-verbal communication.

New

York: Holt,

Rinehart,

&

Winston, 1980.

Koffka,

K.

Principles

of

gestalt

psychology.

New

York:

Harcourt

Brace, 1935.

Kozlowski,

L. T, &

Cutting,

J. E.

Recognizing

the sex

of a

walker

from

a

dynamic point-light display. Per-

ception

&

Psychophysics,

1917,21,

575-580.

Kramer,

E.

Judgment

of

personal

characteristics

and

emotions

from

non-verbal properties

of

speech.

Psy-

chological

Bulletin, 1963,

60,

408-420.

Krauss,

R. M.

Impression formation, impression man-

agement,

and

non-verbal

behaviors.

In E. T.

Higgins,

C. P.

Herman,

& M. P.

Zanna

(Eds.),

Social

cognition:

The

Ontario symposium (Vol.

1).

Hillsdale, N.J.: Erl-

baum,

1981.

Laird,

J.

D.,

Wagener,

J.

J.,

Halal,

M.,

&

Szegda,

M.

Remembering

what

you

feel.

Journal

of

Personality

and

Social

Psychology,

1982,

42,

646-657.

Langer,

E.

J.,

&

Abelson,

R. P. A

patient

by any

other

name

. . .

clinician group difference

in

labeling bias.

Journal

of

Consulting

and

Clinical

Psychology,

1974,

42,

4-9.

Lanzetta,

J. X, &

Orr,

S. P.

Stimulus

properties

of

facial

expressions

and

their

influence

on the

classical con-

ditioning

of

fear.

Motivation

and

Emotion,

1981,

5,

225-234.

Lewin,

K.

Principles

of

topological

psychology.

McGraw-

Hill,

1936.

Lorenz,

K. Z. Die

angeborenen

Formen

Moglicher

Er-

fahrung.

Z

Tiernsychol.,

1943,

5,

235-409.

Cited

in

P.

H.

Mussen

(Ed.),

Carmichael's

manual

of

child

psychology

(Vol.

1).

New

York: John Wiley

and

Sons,

1970.

Marks,

L. E. The

unity

of

senses: Interrelations among

the

modalities. Academic Press:

New

York, 1978.

Massad,

C.

M.,

Hubbard,

M.,

&

Newtson,

D.

Selective

perception

of

events.

Journal

of

Experimental Social

Psychology,

1979,

15,

513-532.

McArthur,

L. Z.

Illusory causation

and

illusory corre-

lation:

Two

epistcmological accounts. Personality

and

Social

Psychology

Bulletin,

1980,6,

507-519.

McArthur,

L. Z.

What grabs you?

The

role

of

attention

in

impression formation

and

causal attribution.

In

E.

T.

Higgins,

C. P.

Herman,

& M. P.

Zanna

(Eds.),

Social cognition:

The

Ontario symposium (Vol.

1).

Hillsdale,

N.J.: Erlbaum, 1981.

McArthur,

L. Z.

Judging

a

book

by its

cover:

A

cognitive

analysis

of the

relationship between physical

appear-

ance

and

stereotyping.

In A.

Hastorf

& A.

Isen

(Eds.),

Cognitive social

psychology.

New

York: Elsevier, 1982.

McArthur,

L.

Z.,

&

Friedman,

S.

Illusory correlation

in

impression formation: Variations

in the

shared dis-

tinctiveness

effect

as a

function

of the

distinctive per-

son's

age, race,

and

sex. Journal

of

Personality

and

Social

Psychology,

1980,

39,

615-624.

McArthur,

L.

Z.,

&

Ginsberg,

E.

Causal attribution

to

salient stimuli:

An

investigation

of

visual

fixation

mediators. Personality

and

Social

Psychology

Bulletin,

1981,

7,

547-553.

McArthur,

L.

Z.,

Solomon,

M.

R.,

&

Jaffe,

R. H.

Weight

differences

in

emotional responsiveness

to

proprio-

238

LESLIE

Z.

McARTHUR

AND

REUBEN

M.

BARON

ceptive

and

pictorial

stimuli. Journal

of

1'ersonalitv

and

Social

Psychology,

1980,

39,

308-319.

Michotte,

A. The

perception

of

causality.

New

York:

Ba-

sic

Books,

1963.

Mischel,

W.

Personality

and

assessment.

New

York:

Wiley,

1968.

Neisser,

U.

Cognition

and

reality.

San

Francisco: Freed-

man,

1976.

Newtson,

D.

Foundations

of

attribution:

The

perception

of

ongoing behavior.

In

J.

Harvey,

W. J.

Ickes,

&

R. F.

Kidd

(Eds.),

New

directions

in

attribution

re-

search

(Vol.

1).

Hillsdale,

N.J.: Erlbaum, 1976.

Newtson,

D.,

Rindner,

R.,

Miller,

R.,

&

LaCross,

K.

Effects

of

availability

of

feature changes

on

behavior

segmentation.

Journal

of

Experimental Social Psy-

chology,

1978,

14,

379-388.

Nisbett,

R.,

&

Ross,

L.

Human

inference:

Strategies

and

shortcomings

of

social

judgment.

Englewood

Cliffs,

N.J.: Prentice-Hall,

1980.

Ohman,

A.,

&

Dimberg,

U.

Facial expressions

as

con-

ditioned stimuli

for

electrodermal

responses:

A

case

of

preparedness?

Journal

of

Personality

and

Social

Psychology.

1978,36,

1251-1258.

Orr,

S., &

Lanzetta,

J.

Facial expressions

of

emotion

as

conditioned stimuli

for

human autonomic

responses.

Journal

of

Personality

and

Social

Psychology,

1980,

2,

278-282.

Postman,

L.,

Bruncr,

J.

S.,

&

McGinnics,

E.

Personal

values

as

selective

factors

in

perception.

Journal

of

Abnormal

and

Social

Psychology,

1948,

43,

142-154.

Rosenthal,

R.,

Hall,

J.

A.,

DiMatteo,

M.

R.,

Rogers,

P.

L.,

&

Archer,

D.

Sensitivity

to

nonverbal commu-

nication. Baltimore, Md.:

Johns

Hopkins University

Press, 1979.

Ross,

L. The

intuitive

psychologist

and his

shortcomings:

Distortions

in the

attribution

process.

In L.

Berkowitz

(Ed.),

Advances

in

experimental social

psychology

(\o\.

10).

New

York: Academic Press, 1977.

Rothbart,

M.,

Fulero,

S.,

Jensen,

C,

Howard,

J.,

&

Bir-

rell,

P.

From individual

to

group impressions:

Avail-

ability

heuristics

in

stereotype formation. Journal

of

Experimental Social

Psychology,

1978,

14,

237-255^

Runeson,

S.,

&

Frykholm,

G.

Visual

perception

of

lifted

weight.

Journal

of

Experimental

Psychology:

Human

Perception

and

Performance,

1981,

7,

733-740.

Runeson,

S.,

&

Frykholm,

G.

Kinematic specification

of

dynamics

as an

informational basis

for

person

and

action perception: Expectation, gender recognition,

and

deceptive intention.

Uppsala

Psychological

Re-

ports (No. 324).

Uppsala,

Sweden:

Department

of

Psychology,

University

of

Uppsala, 1982.

Sackett,

G. P.

Monkeys reared

in

isolation with pictures

as

visual input: Evidence

for an

innate releasing mech-

anism. Science, 1966, 154, 1468-1473.

Savitz,

L.,

&

Tomasson,

R. The

identifiability

of

Jewish

identity through appearance, gesture, voice. American

Journal

of

Sociology,

1959,

64,

468-475.

Scherer,

K. R.

Personality inference

from

voice quality:

The

loud voice

of

extroversion. European Journal

of

Social

Psychology,

1978,

8,

467-487.

Schneider,

D.

J.,

Hastorf,

A. H., &

Ellsworth,

P. C.

Person

perception.

Reading,

Mass.:

Addison-Wesley, 1979.

Secord,

P.

Facial features

and

inference processes

in in-

terpersonal perception.

In R.

Tagiuri

& L.

Petrullo

(Eds.),

Person perception

and

interpersonal behavior.

Stanford,

Calif: Stanford University Press,

1958.

Secord,

P. F,

Dukes,

W.

F.,

&

Bevan,

W.

Personalities

in

faces:

I. An

experiment

in

social perceiving. Genetic

Psychology

Monographs, 1954,

49,

231-279.

Secord,

P.

F.,

&

Muthard,

J. E.

Personalities

in

faces:

IV.

A

descriptive analysis

of the

perception

of

women's

faces

and the

identification

of

physiognomic

deter-

minants.

Journal

of

Psychology,

1955,

39,

269-278.

Segall,

M.

H.,

Campbell,

D.

T,

&

Herskovitz,

M. J. The

influence

of

culture

on

visual

perception.

Indianapolis,

Ind.:

Bobbs-Merrill,

1966.

Shaw,

R.,

&

Bransford,

J.

Introduction: Psychological

approaches

to the

problem

of

knowledge.

In R.

Shaw

& J.

Bransford

(Eds.),

Perceiving, acting,

and

knowing:

Toward

an

ecological

psychology.

Hillsdale, N.J.: Erl-

baum, 1977.

Shaw,

R.,

&

Pittenger,

J.

Perceiving

the

face

of

change

in

changing faces:

Implications

for a

theory

of

object

perception.

In R.

Shaw

& J.

Bransford,

(Eds.),

Per-

ceiving,

acting,

and

knowing:

Toward

an

ecological

psychology.

Hillsdale, N.J.: Erlbaum, 1977.

Shaw,

R.,

Turvey,

M.,

&

Mace,

W.

Ecological psychology:

The

consequences

of a

commitment

to

realism.

In W.

Weimer

and D.

Palermo

(Eds.),

Cognition

and

sym-

bolic

processes (Vol.

2).

Hillsdale,

N.J.: Erlbaum,

1982.

Snyder,

M.,

&

Frankel,

A.

Observer bias:

A

stringent test

of

behavior

engulfing

the field.

Journal

of

Personality

and

Social

Psychology,

1976,

34,

857-864.

Snyder,

M.,

Tanke,

E.

D.,

&

Berscheid,

E.

Social per-

ception

and

interpersonal behavior:

On the

self-ful-

filling

nature

of

social

stereotypes.

Journal

of

Person-

ality

and

Social

Psychology,

1977,

35,

656-666.

Taylor,

S.

E.,

&

Fiske,

S. T.

Salience, attention,

and at-

tribution:

Top of the

head phenomena.

In L.

Berko-

witz

(Ed.),

Advances

in

experimental

social

psychology

(Vol.

11).

New

York: Academic

Press,

1978.

Toch,

H.

H.,

&

Schulte,

R.

Readiness

to

perceive vio-

lence

as a

result

of

police training. British Journal

of

Psychology,

1961,

52,

389-393.

Todd,

J. T,

Mark,

L.,

Shaw,

R.,

&

Pittenger,

J. The

per-

ception

of

growth. Scientific American, 1980, 242,

132-144.

Turvey,

M. T.

Preliminaries

to a

theory

of

action with

reference

to

vision.

In R.

Shaw

& 3.

Bransford

(Eds.),

Perceiving,

acting,

and

knowing. Hillsdale, N.J.: Erl-

baum, 1977.

Woll,

S. B., &

Martinez,

J. M. The

effects

of

biasing

labels

on

recognition

of

facial

expressions

of

emotion.

Social cognition, 1982,

/,

70-82.

Zuckerman,

M.,

DePaulo,

B.

M.,

&

Rosenthal,

R.

Verbal

and

nonverbal communication

of

deception.

In L.

Berkowitz

(Ed.),

Advances

in

experimental

social psy-

chology

(Vol. 14).

New

York: Academic Press,

1981.

Received

September

1,

1982

Revision

received

February

19,

1983

•