2-B-47 Biological Motion知覚におけるERP計測(2002年度 日本基礎心理学会第21回大会優秀発表賞)

The Japanese Psychonomic Society

NII-Electronic Library Service

The JapanesePsychonomic Society

The

Jmpanese

foufmal

of

Psychenom･ic Science

2003,

Vol,

22,

No.

1,

55-56

Prizewinner's

Summary2-B-47

Specific

negative

component

biological

motion

dniveTsi

elicited

by

the

_perception

of

An

ERP

study

Masahiro

HiRAi*

and

Kazuo

HiRAKi*'

**

c3,

of

7bhyo*

and

RRESTO,

]ST

(laPan

Science

and

712chnology

Corporation),laPan**

Recent

studies

have

revealed

that

the

superior

temporal

sulcus

(STS)

_plays

an

important

role

in

'social

perception'.

In

this

study, we measured event-related

_potentials

(ERPs)

during

the

perception

of

biological

motion

that

can

be

_perceived

from

locomotion

which

is

monitored

by

only

points

of

light.

Twelve

subjects

_participated

and

they

were

shown

representations

of

biological

rnotion and scrambled motion.

In

the

scrambled motion, each

light

_point

had

the

same velocity

vector

as

in

the

biological

metien,

but

the

initial

starting

positions

of

the

lights

were

randomized.

The

perception

of

both

the

biological

and scrambled motion clicited ncgative

peaks

at around

200

(N200)

and

240

(N240)

ms.

Notably,

the

N240

component

was significantly

larger

over

the

bi]ateral

occipitotemperal region

for

the

biological

motien

than

for

the

scrambled metion condition.

These

findings

do

not

contradict

_previous

neuroimaging

results,

but

imply

that

the

N240

component

seems

to

be

specific

to

biological

motion

perception.

Furthermore,

this

component might

be

similar

to

the

detection

of

intcntion

component

that

was

found

in

_gaze

direction.

Key

words:

biolegical

motion, event-related

_potentials,

superior

temporal

sulcus, motion

tion

Our

brain

can reconstruct rich visual

images

from

sparse

input.

Biological

motion

is

such a well

known

phenomenon

that

we can

_get

a vivid

impression

of

human

form

and

activity

from

just

1

1

light

_points

of

the

motion.

Previous

human

neuroimaging

studies

have

revealed

that

the

superior

temporal

sulcus

(STS)

plays

an

important

role

in

_perception

of

bio-logical

motion.

However,

_the

dynamics

of

brain

acti-vation

during

the

perception

of

biological

rnetion

have

not

been

studied,

In

the

present

study, we observed

the

dynamics

of

brain

activity

during

the

perception

of

biological

motion

by

measuring

event

related

potentials

{ERPs}.

Methods

Twelve

naive

subjects,

9

males

and

3

females,

aged

between

23

and

29

years

(mean

±

SD;

26

±

2.3

_years)

participated

in

the

experiment,

The

subjects viewed

two

different

kinds

of

animation

(Fig,

IA).

The

animations

(5.7X

5.70,

2,O

_gait/s)

were

displayed

on

a

17-inch

CRT

monitor,

AII

of

the

_points

of

light

(9,8

*

_Department

_of

_{Multidisciplinary}

_Science,

Gradu-ate

School

of

Art

and

Science,

The

University

of

Tokyo,

3-4-1

Komaba,

Meguro-ku,

Tokyo

153-8902

arcmin> were

displayed

as

black

against a white

background,

The

experiment consisted of

5

blocks

with

inter-block

intervals

of one minute.

Each

block

had

50

biological

motion

stimuli

and

5e

scrambled

motion

stimuli

in

a

pseudo-random

order,

In

each

trial,

the

stimulus

was

presented

(biological

motion

or scrambled motion)

for

1,OOO

ms

and

followed

by

a

fixation

point

(a

cross

bar,

O.41Xe,410)

for

3,OOO

ms,

The

electroencephalogram

(EEG)

activity was re-corded

by

using a

_geodesic

sensor net

(Electrica]

Geodesics>

consisting of

64

silver-silver chloride

elec-tredes

evenly

distributed

across

the

scalp

of

a

par-ticipant.

The

electrical

_potential

was amplified and

filtered

with a

O.1

to

50-Hz

bandpass,

and

then

digit-ized

with

a

sampling rate of

250

Hz.

The

trials

in

which

the

signal

variation

exceeded

50"V

in

either

the

EEG

or

EOG

were excluded

from

the

averaging.

The

analysis window was extended

for

1,OOO

ms

fol-lowing

the

onset

of

each

stimulus,

and

a

pre-stimulus

period

(1OO

ms)

was

used

as

the

baseline,

The

vertex

served as a reference, and

the

ERPs

were re-referenced

to

the

average

potential

over

a

subject's

scalp.

Figure

1(B}

shows

the

total

averaged

ERPs

across

the

12

subjects under each condition.

The Japanese Psychonomic Society

NII-Electronic Library Service

The JapanesePsychonomic Society

56

The

Japanese

Journal

of

Psychonomic

Science

VoL22,

No.

1

A

bielo

'

.

.

t-

,

-xB

xxs

PeakAmpkita}de (V4V)

.3

a caLmotion

.

--t-

-･

---

--''

-

--x.. -t rx tme'i

Cz

itT6

.L

_l

fi

h serambledmatien

,

,

,

-

----t

-t

-

--

, -- "t

.x.

....

..

Gm:.s.

N2op

240

-biolesietmoim

't.x

.'wwpmmUndttmbts

Peekintonqy pt]n / TSMooT6TS T6 N240

"totsOlll"Vv

4oomsec -bioteticalthcrtie! -emEtiNedmmhe

-t/1TSmooT6ffN240T6

'tpq.o]

Figure

1.

(A)

An

example

of

the

stimuli,

(B)

The

total

of

the

averaged

EPRs

across

the

12

subjects elicited

by

the

biological

and

bled

motion,

Reslllts

and

Discussion

During

perception

of

the

stimuli of

both

the

bio-logical

and

scrambled

motion

two

negative

compo-nents could

be

distinguishcd:

one was approximately

200ms

and

thc

other approximately

240ms.

We

named

the

components

N200

and

N240,

respectively.

The

amplitude and

latency

of

N200

and

N240

were subjected

to

a

two-way

analysis of variance

(ANOVA)

with

the

repeated measurement

factors

of

hemisphere

(left,

T5;

and

right,

T6)

and

condition

(biologica]

motion and scrambled motion),

For

the

arnplitude

of

N200,

thc

interaction

of clectrode

posi-tion

and condition was significant

(F[1,

11]=7.6,

P<

.05).

The

simple main effcct of

hcmisphere

was

significant

for

the

biological

motion condition

(F[l,

22]=8,5,

_P<.Ol)

and

it

refiected

the

fact

that

the

amplitude

for

the

biological

rnotion

condition

was

larger

in

the

right

hemisphere,

The

simple

main

effect of condition was also significant

for

the

right

hemisphere

_(F[1,

22]=7.2,P<.05),

indicating

that

the

arnplitude

of

the

biological

motion

condition

was

larger

than

that

of

the

scrambled motion condition

in

the

right

hemisphere.

We

also observed

that

the

simple rnain effect of condition was significant

for

the

amplitude

of

the

N240

(F[1,

11]=20.3,

P<.Ol).

This

result

indicated

that

the

amplitude

for

the

bio-logical

motion was

larger

than

the

scrambled motion

for

both

hemispheres.

In

contrast, we

did

not

find

any

significant

factors

in

the

latency

of

both

compo-nents.

Previous

neuroimaging studies

have

shown

that

the

V5/MT

area

is

responsible

for

motion

per-ception.

In

both

of

the

stimuli,

the

number

of

_points

and

the

velocity

vector

were

the

same,

and

this

indi-cates

that

the

V5/MT

area

rnight

be

activated

equal-ly.

However,

we

found

amplitude

differences

in

both

conditions.

which

imp!ied

that

there

is

a

specific

mechanism

that

responds

to

biological

motion

in

addition

to

that

in

the

V5fMT

area,

Recent

human

neuroimaging studies

have

reported

that

the

right

STS

was

activated

during

the

perception

of

biologi-cal

motion,

Wc

hypothesizcd

that

the

observcd

two

ncgativc

componcnts

can

be

cxplaincd as

follows]

the

N200

component

reflected motion

_perception

in-volving

the

V5/MT

region,

and

the

N240

component

was associated with

higher

proccssing

of

the

motion stimuli

involving

the

STS

region.

A

recent study

has

revealed

that

the

STS

is

activated not enly

by

bio-logical

motion

perception,

but

also

by

other

stimuli

of

biomechanical

motion

(e.g.

_gaze

direction

and sign

language}.

This

type

of

_perception

has

been

called

'social

perception'.

We

intend

to

c]arify

the

relation-ship

between

the

N240

component

and

social

inter-action

in

future

invcstigations.

References

Allison,

T,,

Puce,

A.

&

McCarthy,

G,

2000

Social

perceptton

from

visual cues: role of

STS

region.

Trends

Cogn.

Sci.,

4,

267-278.

Hirai,

M.

Fukushirna,

H.

&

HirakL

K,

2003

An

event-re]ated

_potentials

study of

biological

motion

perception

in

humans.

AJeurosci.

Lett.,

344,

41-44.

Watanabe,

S.,

Miki,

K.

&

KakigL

R.

2002

Gaze

tion

affects

face

_perception

in

humans,

IVeurosci,