The effect of haptic information on vision : Using geometrical patterns with and without visual illusion(Summary of Awarded Presentations at the 24th Annual Meeting)

The Japanese Psychonomic Society

NII-Electronic Library Service

The JapanesePsychonomic Society

eneJmpanesefournatof Rs'.vchmtomicScience

2006,VoL 25,No.1,107-10S

Summaryof

Awarded

PresentationIP23

The

effect

of

haptic

information

on

vision

-Using

geometrical

patterns

with and without

visual

illusion-Keiko

OMoRI*i

**,

_Yuji

_WADA***,

_Yukio

_{ITsuKusHIMA*,}

_and

_Kaoru

_NoGucm*

Nihon

Uhiversit),*,

JaPan

Society

for

thePromotion

of

Science**,

and IVdtionalFood Research

institute***

We investigated the effect of

haptic

information on visual

illusions

and the appearance ef

simple geometrical patterns without

figural

components which induce visual

illusions,

ment

I

was

designed

to

examine the effect of haptic information on the visual illusionsof the

Hering and

Wundt

figures, The haptic stimuli with three differentcurvatures were made of

wooden board, The participants were asked to

judge

the apparent curvature of each testfigure,

with or without

the

haptic stimuli,

In

Experiment

II,

the

participants

judged

the apparent

curvature of each testfigurebut cornponents

inducing

visual illusions,with or without thehaptic

stimuli. Itwas

found

thatthe visual illusionswere biasedtowards the

direction

of the information

given by actively touching the

haptic

stimuli and thaL the visual perception of simple patterns

without any

inducing

components was not affected by actively touching.

Key words: multisensory perception,haptic information,visual

illusion

Vision

provides perceptual properties such as $hapq size,and texturewhen we recognize an object.

Simultaneously, we are able toknow the _properties

'

of an object by touch. A question to

be

answered is

how

we organize

information'

from

the

different

sen-sory modalities, vision and touch, when we perceive

an object, Ithas

been

demonstrated

that vision is

generpllydominant over theother modalities

jn

inte-gration

of multisensory information

(Rock

&

Harris,

1967),while a

few

case of haptic capture was

re-ported

(Ernst

&

Banks,

2002;

Gallace

&

Spence,

2005). However, there exists strong evidence that

haptic

information

plays an important role

in

the people with restored sight

(Torii

& Mochizuki,

1992).

In the _present study, we

investigated

the effect of

haptie

information

on visual

illusions,

and simple

geometrical patterns without figuralcomponents

in-ducing visual

illusions.

* _{Department of}

_Psychology,

_Graduate

School

of Literature and Social Sciences, Nihon

University,

3-25-40

Sakurajosui,

ku, Tokyo 156-8550

**

_Research

_{Fellow of}

_Japan

_Society

for the

Promotion

of Science,8Ichiban-cho,

ku,

Tokyo

102-8472

***

_National

_AgricuLture

_{and Food}

Research

Organization,

2-1-12 KannodaL Tsukuba,

Ibaraki305-1684

Experiment

I

Method

Participants

Seven

students who had normal

touch and vision _participated

in

the experiment.

Apparatus

and Materials The visual standard stimuli were visual illusions,the Hering and

Wundt

figures.InFigure

1A,

each of _thetwo vertical linesin

thesevisual standard stimuli was _physicallyconvex,

parallel,and concave lines.The visual comparison

stimuli consisted ofnine pairsof vertical lineswhich

changed incurvature

from

convex to concave lines

(Figure

IA). Haptic stimuli were three different

shapes

(convex,

parallel,and concave) rnade of

wooden

board

(Figure

IB).

Procedure

The

participants selected one of the

visual comparison stirnuli which seemed

to

match

the vertical linesin the visual standard stimulus,

The

standard stimulus was presented on the display

ina random order with or without haptic stirnulus

(Figure

1).

Results

and Discussion

The

participants'

judgments

of the degrees of

ap-parent curvature were expressed

in'terms

of PSEs.

When an appar,ent curvature was

judged

as

being

parallel

{the

comparion stimulus

(D)

the PSE was

designated

as zero. The PSEs were _positivevalues

The Japanese Psychonomic Society

NII-Electronic Library Service

The JapanesePsychonomic Society

108

The

Japanese

Journal

of

Psychonomic

Science

VoL 25,No. 1

Visual Etand-rd stitrul1

:1,1",,･1tttt/ttt･･1･za,'IEl,"'//-li"ttt/e

t/tttt/"''t'tt/.tti':t.tttttt.tL','

/epley

ee

Teble Vlsual mi.t/

.ca..ttttantttttttt.t..

rieonetithuli

t/t

ig,,.n.Ili,II'lll"li

t･l"-i,l#.:･;･l-1//･il･11/:1,1,l-lttt

(.ml(.Iii)

,t,g,;-/#tal

ltlre,:l;1('ii)/1la.e OlspleweptioGtimulus Eble c" e)

Figure

1.

(A)

Front view

{B)

Side

view,

from 4to 1when the

judgment

was ranging from

the

comparison stimuli

O

to

@

respectively; and the

PSEs were negative values

from

-1

_to

-4

when the

judgrnent

was ranging

from

thecomparison stimuli

(Q)

to

(SD

respectively. To testtheeffect ef the haptic

information

on thevisual illusion,thePSEs

for

each apparent curvature in the visual

illusion

were

ana-lyzed using a two-way ANOVA with visual illusions

and

haptie

information

_(convex,

_parallel,concave,

and nonhaptic information) as main

factors,

The

analysis of the Hering

figure

data

showed a

sig-nificant main effect of haptic information

[F(3,

18)=

34,43,p<.OOI]. The analysis of the

Wundt

figure

data also showed a significant main effect of

haptic

information

_[F(3,

18)=74.03,p<.OOI]. A significant

main effect of

haptic

information

implied that the

haptic

information

had influenceson theappearance of thevisual illusion.

Experiment

II

Method

Participants

Six

students who had normal touch

and vision _participatedinthe experirnent.

Apparatus

and

Materials

Three

_pairsof vertical

lineswhich were physically convex, parallel, and

concave

lines

were used as the visual standard stim-uli. Inother words, they were simple _geometrical patterns without figuralcomponents

inducing

the

visual illusionswhich were used inExperiment

I,

The

visual comparison stimuli and the hapticstimuli

were the same as inExperiment

L

Proeedure

The same as inExperiment

L

Results and Discussion

The

judgments

of apparent curvature were

ex-pressed

in

the same manner as inExperiment

I.

To

testtheeffect of the haptic information on the geo-metrical _patterns,PSEs

for

each apparent curvature

inthe _geometrical patterns were analyzed using a

two-way

ANOVA

with the_{geometrical pattern}

(con-vex. _parallel,and concave) and

haptic

information

(convex,

paralleL concave, and non

haptic

inforrna-tion) as themain factors.The analysis showed _that

the main effect of haptic information was not sig-nificant, The result suggests thathapticinformation didnot alter the appearance of the simple geemetri-calpatterns,

General

Discussion

In

Experiment

L

visual illusionswere biased

to-ward thedirectionof theinformation _{given bytouch.}

InExperiment

II,

however,

the visual _perception of

simple _patterns which did not have any inducing

components was not affected

by

the

information

given by touch. These results might beexplained by

thebptimal

integration

hypothesis':the interaction

between

visual and hapticinformation may

function

as a maximum-likelihood estimation

integrator

that

minimizes variance in_the

final

estimate

(Ernst

&

Banks,

2002).

If

this

is

thecase, theappearance of the

visual

il]usions

was more ambiguous than thatof the

sirnple _patterns.

To

_perceive a stable world, our

visual system should be organized multisensory

in-formationcomplementarily,

References

Ernst,M.

O.

&

Banks, M. S.

(2002).

Humans integrate

visual and

haptic

information

ina statistically

timal

fashion.

AJdture,415,429-433.

Gallace,

A,& Spence,C,

_(2005).

Examining

the

rnodal consequences of viewing the MUIIer-Lyer

illusion,Emperimen tal

Brain

Researuh,

162,490T496,

Rock,I.& Harris,

C,

S,

_(1967).

Vision and touch.

Scien-tific

Amert'can,

216,

96-104.

Torii,

S.

&

Mochizuki,

T,

_<1992).

R)nnation

of

visual

'

Percoptzon

l:1bstQPerativelacalizationand

discrimi-nation

of

color and shape

in

thecases

of

the

tat(yblind.

Tokyo:

Baifukan.

(In

Japanese)