1-B-42 視覚的文脈は触覚探索に転移するか?(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

TheJLipangsefournalofPsychonomicScience

2ee3,Vel.22,No.1,35-36

Prizewinner's

Summary1-B-42

Visual

spatial

contexts

Tomohiro

NABETA,

transfer

to

Fuminori

ONo,

and

IIiroshima

Uitiversity*

haptic

search

Jun-ichiro

KAwAHARA

Under incidentailearningconditions spatial ]ayouts can

be

acquired

implicitly

and

facilitate

visual search

(contextual

cueing effecO.

We

examined whether thevisual contextual cueing effect

is

specific tothemodality or transfersto thehapticmodality. The _{participantsperformed} visual search trials

based

on a typica]contcxtual cueing _paradigm, followed by haptic search trials

in

which halfof thetrialshad layoutsused in

the

previous visual search trials.

The

visual contcxtua] cueing effect was obtained in

the

learning

_phase.

More

importantly,

_the effect was transferred

from

visual to

haptic

search. This suggests a cornmonality of $patial memory toallocate focused attention

in

both

visual and

haptic

modalities.

Key words: transfer

from

vision tohaptics,contextual cueing, implicitrnemory

Spatial context facilitates

the

visual processes,

such as a visual search.

To

examine

how

spatial

context islcarned,

Chun

and

Jiang

(1998>

devised a centextual cueing _paradigm. In a typical experi-ment, participants conducted visuar search fora

tilt-ed "T"

among rotated '`L"s

and

indicated

the

direc-tion of Lhe target. Half of the search trialshad

unique stimulus

layouts

whereas theother ha]fhad

layouts that were repeated

throughout

_the

experi-menL Accordingly the _participants

identificd

the

orientatien of the target

faster

with the repeated

layouts

than with the unique layouts,although the

participantswere not able to recognise thercpcatcd

layouts.

Their

performance suggested that they

learned

the repeated spatial layouts irnplicit]yand used this

implicit

knowledge to_guide their attention

toward thelocatienof thetarget.

Although recent studies

have

shown that spatial

contexts yield robust contextual cueing effects, the nature of the spatial memory that mediates the ef-fectsisnot clear. Spccifically,isthe memory of

spa-tial

context specific tothe visua]

domain

or

is

it

viable across modalities ?

We tested this possibilityin

the

present study,

The experirnent consisted of 2 _phases. First,_the

participantsconducted visual search trialsbased on

a typicalcontextual cueing paradigm; then the

par-* _{Department of}

_Psychology,

_Hiroshirna

Univer-sity, 1-1-1 Kagamiyama, Higashi-Hiroshima

739-8524

ticipantsconducted

haptic

search trials.

Method

Stimulus materials and apparatus The visual

stimuli were disp]ayed on an LCD monitor. The stimuli consistcd of one "T"

and eleven "L"s. _The "T"

was the targetand was tilted

_90"

_tothc right or to

the left;the "L"s

were distractorsand were rotated

ee,

900.

1800,

or

2700

from

the upright orientation.

In the transferphase, 12 items inahaptic search

array were raised on a _plasticsheet for overhead

projection

(Figure

1).

Design

and _procedure

In

the ]carning

(visual

search> _phase.the 2 main variables were thelayout

(old

vs. new) and epoch

(1-5).

The old set of stimu]i

consisted of

8

randomly _generated

layeuts

through-out the learningphase,once per block. The repeated

layouts maintained thelocationof the itemsand the

identity of

the

distractors,

but

the

identity

of the

Figure 1. The search array used in the

The Japanese Psychonomic Society

NII-Electronic Library Service

TheJapanesePsychonomic Society

36 The

Japanese

Journal

of

Psychonomic

Science

Vol.

22,

No,

1

target was changed randomly from trialtotrial.The

new set consisted of

8

different

layouts

that

were generated anew

for

each trialtoserve as a control

baseline.Each

_participant

was

_given

20

blocks

of

16

trialseach

{8

old, 8 new}, fora totalof 320 trials.

Twenty blockswere _grouped insets of 4intoepechs,

resulting

in

5

epochs.

In the transfer

(haptic

search) phase The main

variable was the layout

(common

vs, specific), The

common set consisted of

8

layouts

which were

the

same as the old layouts used inthe

_previous

visual

search trials.The specific set consisted of 8 layQuts

that were generated anew,

Eachsession consisted of2blocks of 16trialseach

(8

common,

8

specific), resulting

in

atotalof

32

trials.

The 12 naive _participantssearched ferthe target

visually

(learning

_phase) and

haptically

(transfer

phase), and indicated its orientation

by

pressing

either the right or left

foot

switch throughout the

'expenment.

After

the

haptic

search trials,the participants im-mediately performed a hapticrecegnition testwhich consisted of 16trials,Eight were thecommon

layout

and the other

8

were fi11ersthat were _generated

anew.

Results

Searchtask

The mean reaction times of the

visual search task

for

all of the correct responses

within an epoch were calculated.

These

data

were

used for a repeated-measures ANOVA with the

layout

_(old

vs. new) and epoch

(1-5)

as within-subject

factors.

The analysis revealed significant main effects of

layout,

F(1,

11)=34.63.

P<,Ol,

and epoch,

F(4,

44}='='

3,02,

p<.05.

The

interaction

between

layout

and

epoch was not significant,

F(4,

44>=2.34, n.s.

How-ever when the analysis was restricted toepochs 1

and

5

only, the

interaction

was significant, F(1,11)=

6.00,p<.05.

In

the

transfer

phase,

the

mean reaction times for the correct trialswere calculated for both the

com-mon and specific conditions.

Analysis

of the

data

with a t-tcst revcaled that the

haptic

search was

fasterinthe common condition than inthe specific

condition,t(11)=2.35,P<.05.

Recognition test The mean accuracy

in

the

rec-ognition task was 47%, The _participantscorrectly classified the common layouts as old in23% of the trials

(hit

rate), and thiswas same astheirfalse-alarm

rate of

23%,

t(11)=O,19,n.s.

Diseussion

In

the

present study, we obtained

3

major results.

First,theresult of the learningphase showed thata

visual context was obtained through visual searches: the reaction times

in

the old condition were

sig-nificantly shorter than

in

thenew condition,

Second,

we

found

that the

haptic

search _performance was

fasterwhen the layoutswere identicalto_previously

experienced visual search layouts than when layouts

differed

from

thoseof thevisual search,

Finally,

the mean accuracy of therecognition testwas at chance

level.

Based

on

these

resu]ts, we suggest

that

im-plicitspatial memory acquired through visual search

isalso available

for

_guiding

the

focus

Qf spatial atten-tion inhapticsearch.

Our

results

demonstrating

thetransfer of a learned spatial context

for

a search across visual and

haptic

modalities suggest thatthevisual memory isshared toorient multiple spatially distributedobjects.

Reference

Chun, M. M.

&

Jiang,

Y. 1998

Contextual

cueing:

Implicitlearning and memory of visual context

guides spatial attention. CognitiveRsycholqgy, 36,