"SHE 5'5. 2233 V3 Shim 5.333313353313333 33 331.53? D PER..»C)NS

“i.” 0
(36453513.:

up 1316 3953'9‘6 a? 3““ fin

“13.3mm 53311:: uaazn=2asm=r

ea
93
3:»...
c
:3

anssna

’
-‘A 4 I A
L... L. 41:13 ’3 ml‘ a! 3-2: h

 

 

,LIBRA R Y
Michigan sm {'6
Pnivcrsity

Pic

.- “

i
‘1‘ £8334~1V21¢23 2.3.".

irh
- ‘2.

A THE SIS

Submitted to the College of Science and Arts
Hiehigan State University of Afrleulture and
Applied Science in pa rtial f 1lfillment of

11
the reo11irements for tie degree of

WAJTT? 5? ARTS

Department of Psychology

TL: 3137-231??? ILLUSICN
IE ELlfT P;PSCT3
12v

FernandO'Jolon

11 51:91.3 3:11—11
Submitted to the College of Science and Arts
Hichigen State University of Agriculture and

AppliedQ Science in partial Iulfillmer t of
the requirements for the digree of

‘r‘,',« firn‘rylj "it‘
,' ‘lLC' J. .,_1,'.. L \51

Department of Psychology

Fernando Colon
The present study was designed to investigate the occurrence
or non-occurrence of the Size-Peijht Illusion in congenitally

blind persons and also to determine whether the sithed person de-

('3
Ho

veIOps his reaction to the Size-T ght.Illusion by the use of

sirht as well as the kinesthetic and tactual senses.

The psychophysical method of constant stimuli was used to get
a measure of the frequency of occurrence of the Size-Height Illu-
sion. The apparatus consisted of two series of five blocks each.
The first series of blocks were designed to keep the size con-
stant as the weijht varied, while the second series of blocks was
'esijned to :eep the weifht constant as the size varied.

There were three groups of ten observers each: a sighted
group, a blind-folded sighted group and a congenitally blind
group. One hundred comparisons with each series of blocks were
made by the observers in each of the three sroups utilizing a
pincer lift (using just thumb and forefinger to lift the block.)
The congenitally blind group in addition, using the second series
of blocks, made 100 judgments with a palmar lift and 100 judg-
ments with a grasp lift.

The eXperimental results delinitely demonstrated that the
Size-Height Illusion does occur in the confenitally blind person
and that the sighted perscn does deveIOp his reaction to the Size—
Eeight Illusion by the use of siiht as well as the kinesthetic and
tactual senses.

It is concluded that neither the tactual sense, the visual

sense nor the kinesthetic sense alone is enough to elicit the

Fernando Colon
Size-height Illusion. It seems that at least two of the senses

must be involved in order for the illusion to occur.

T_
Batmmfirf' L O

 

w as 1;:1c r. . L we 113337111
I deeply appreciate the continuous help and constructive
sugfiestions given by 3r. S. Howard Bartley and Dr. Thomas M.
Nelson during the formulation and execution of this research.
I wish to thank Dr. Albert I. Rabin and Dr. Donald T.
Johnson for reading the manuscript and offering helpful cri-
ticism.

. Kartin Crump for his invaluable

[11‘

I am also indebted to
assistance in constructing the apparatus used in this research.

P1 (-1
0 U 0

Ho
Ho

LIST CET‘PA>I3§3....
I IJCEIJDVJTILT’
II METROD......

Subjects.

‘ 7???
1.! I T—LJ_J'~~/1a'

1v 7* I

1358 .tus

\
I-
.a..

ha
'1 3
F3
0')

Procedure

.rI‘SOOOOO

,—

1‘r
.J I“

1 .‘r

,_ T”
s.) J.K/ .'.'. O O

(/1

m

V SUT7ARY.....

SIPLIQ“f

7—

1

:EEZY. o o o o 0

54

L';

‘
J

(I

Ho

Ho

Ho

5-)
CD

EISUFE l Occurrence of t‘ Size-Eeight Illusion

}-10

Group I - Si~hted - pincer 1 ft

I
”:3
F0
r5
(3
(D
"S

Group II - Sifihted-Dlindfolded

1

1d — rince

(.2

I
, _
4

onjenitally rli

ya!

Group III -

,4&

L

[’

ll'tfla‘l

1-.
x

‘
LJ

1

‘1‘, T“
m" 1;;

r1 '1
_'-.

.18

The size-weight-illusion (SKI) refers to the fact that when
subjects are asked to make a judgment about the weight of two equal-
ly weighted objects of different size they tend to call the larger

.1

sized object lirhter and the smaller sized ooject heavier.

Actually the size-weight-illusion is not really an illus1on

but it is rather a case in which a person reacts to the density

of an object. If a large object weighing 50 grams is compared to
a small object weiching 50 grams ne laiter will have a fireater
oensity and appea ar heavier. Thus in the comparison of two objects
density becomes the primary factor when the weights are the same
and the sizes are varied and weight becomes the primary factor
then sizes are the same and wei hts are varied.

Sharpe ntiere was the first to describe the SKI. We obtained
it by using two balls with the same physical weight but of a dif—
ferent size. Flourney% elicited the illusion even though no tac-
tual contact was made with the objects when the O was told to put
his finger throufih a ring to which a stiff thread was suSpended
and attached to the objects to be lifted. He had his 0's repeat
the experiment with their eyes closed and there was no SUI. Van
Bierivliet* had his 0's lift a bottle by the neck and by the bot—
tom. Here the size was the same in the two situations. He cor nduct-
ed several experiments in this manner with the eyes Open and the
eyes closed. He got the SWI in all cases.

Koseleff (5) went further and made 0 compare the heaviness of

.0

a small object looked at directly with the heaviness oi the same

\‘a
1“\

Findings translated and reviewed by Koseleff (5)

object perceived through a convex lense. There was a SWI; the
thing different was the retinal images. Fe achieved another
reduction of the visual size by making 0 look throuwh a small
hole in a piece of cardboard and again the SflI occurred. Kose-
leff (5) was also able to elicit the SWI when he used two bloc:s
which were different in heijht but exactly alike in length and
width. When one was superimposed upon the other they seeminfly
formed one body. The smaller one was about nine times as heavy
as the larger one. When both blocks were firasped and lifted with
the heavier on top they felt lighter than the heavier one alone.
When the heavier one was underneath the illusory effect was smaller
and more uncertain than when it was on top. (4) Koseleff (6) ob-
tained data which seemed to signify that the STI was freater when
the larger object was measured by the smaller one than when the smal-
ler one was measured by the larger one.

Usnadze (15) had his O's compare two balls of the same size

99 mm) weigh n

I
H.

(diam a 286 grams and 880 drams respectively.

Then this was done with the eyes closed the lighter one appeared

to most persons to be larger. In this case the weight affected

the experienced size which was the reverse of the STI wherein the
size affected the experienced weight. Usnadze also placed eight
prisms of different heights but with the same weight on the resting
arm of the 0's. No active lifting was performed and again the SKI
was found. It also was found when the weights are placed on the
resting hand of the 0. However, in the latter case, when O's at-

tention was not on the size of the objects the SUI diminished

9‘1
{3,

Ind tende to disappear altOfether when the eyes were closed.

Huang (3) focused on the factors of density and wei ht. S's
were required to compare under conditions otherwise identical the
weights of objects of unecual size in one set of experiments and
their densities in another set. To Pet C's to make judfments hcs-
ed on density they were asked whether one of the objects (apart from
their weiéhts) appeared more "empty or "floating” than the other.
The results for the wei ht liftins eXperiments show the usual SYI

'I O

and the jPeater the difference in the size of the objects compared

the greater the illusion. The results for the densitv eXperiments

1‘

showed the same trerd with the 8?: effect beinj greater. However,

in the atter case the differential limens and the standard deVia-

‘4

tions of the ESE'S :ere slightly freater that the correSponding

values for the wei3ht experiments.
“hippie (14) in his "Ianual of ”ertal and Physical Tests,"
states that the SWI hardly exists up to the ass of three, is well
develOped by the afe of six, continues to increase up to the ate of
nine, and then slowly declines. Iyssen and Eourdon (9) conducted
an extensive experiment with 600 normal S's between the ares of

20 and 60 years. To study the SWI they used as weights paralleli-
pipeds in beechwood. Cne block was 21 cm. long, five cm. wide and
only two cm. hijh. The other block was the same xc pt that it was
seven cm. high. Both blocks weijsed 500 grams. They found that
the STI occurred in 99.17? of “he 630 cases when the comparison be-
tween these two blocks vas made. They found that the illusion ex-
isted not only when different sizes with the same weifiht are com-

pared but that it also remained when the wei ht of the larger size

4
was sensibly hiyher than the little sized one. Because of this fact,
szsen and Bourdon (10) M in 1< that the majnit ude of ihe illusion may
be rated by the phxs ically heaviest weieht for which the illusion
is still present. In comparinj data from 50 males and 50 females
they found that the magnitude of the illusion was the same for
both sexes. They also compared 20 universitv educated O's with 20
unskilled laborers and there was no sifnificant difference in the
.agnitude of the STI due to educational level.

.‘u '~

Applyinfi he same procedure to be veneral paretics, 42 senile

patients and 97 adult oliQOphrenics, all very generally deficient,
tLe SiI proved to be clinically worthless f’or the detec thIl of men-
tal deficiency or deterioration in adults. (9) They pursued his

idea further using Teseleff's (4) iorm of the illusion believing
it would be more agt to produce a less tenacious illusion, partic-
ularly in patho lo ical cases with pronounced deficiency or deter-
ioration but without success. Tyssen and sourdon (S) cite Clar-
oarede's observation of the ex
the peda o ical tr ackward S's of the special classes in weneva.
Doll (1) de es ed the illusion in lcOL of the abnormal S's whose
mental age reached eiyht years. On the other hand, Demoor, as

sported by fyssen and Bourdon (9), found the SYI absent or inVert—
ed in feeolew in ded cnildr en. however, most authors are :encrally
agreed that the ma;nitude of the illusion is of the same rate in
normal persons as in CWefi ient or dete :iora ted ones .

A survey of the work done on the strength or fr cue ency of the

illusion in relation to the various seise modalities revealed that

q

1’10

"‘2
L

:4. Seashore (11) found

[—10
f—L'o
g.)

“eneral agreement in the f 1

fa
\__‘

there was

5
tLat the 811 was weaker when the blocks mere viewed in indirect
vision and still weaker when judged by visual merory. lhen the
size was estimated by the combined effect of all the Spatial senses
the illusion was weaker than when demanding on the tactual and the
kinestbetic senses.

1

Nyssen and Bourdon's (10) results a ree with Seashore's. Tney

had a group of 50 3'8 that were blindfolded while they lifted the
blocks and a second group of SO S's that were not blindfolded but

who lifted the blocks with straps from which the wei ht was hun'

a:
(J \J.

Without visual perception the apparent weight was 265 grams, without
the tactual perception it was lol grams and with both tactual and
visual cues the apparent weiyht was 250 drama. Neyers (8) stated
that the illusion is strongest then the subject is permitted both

to see the objects and to graSp them. Huang (2) did several exper-
iments under five conditions, each of which involved a further re-

duction of size cues than did the preceding one. his results agree

with those stated ayove.

( ,0

Thus the degree of the KI obtained was generally found to be

oeoendent upon the amount and kind of sensory information made

H4

available. It was generally agreed tiat the tactua and kinesthetic
senses were more potent than the visual sense and that all the
senses totether resulted in a weaker illusion than the tactual and
kinesthetic alone. Huang (2) states that it is of great interest
that with all the sensory cues eliminated, knowledre and memory

alone of the difference in size of the objects seemed to lead to on-

ly a small degree of the illusion. He feels that this shows that the

6

!

lenendent upon actual sensory data rather than on ideational,

.-

C’}

VIi“: I i S

(

intellectual attitudes.

Several explanations of the SKI have been proposed. Muller
and Schuman as cited by Roodworth (15), theorized in 1989 that as
O prepares to lift th comparison weight he will do so with a mus-
cular force just previouslv found adequate for liftiha the standard.

L \4‘

He carries over a motor adjustment or hinstellun‘" from the

r
4" .
k.)

‘
,

lift to tie second. If the comparison weirht comes up quickly and
easily, it seems lith and is judfed lighter than tne standard; if
it resists and comes up slowly it seems heavy and is jud
than the standard. Therefore, a weight which looks heavier than it
is will be lifted with greater force than necessary and accordingly
feel light.

Loomis' (7) experiment tends to support this View. He used
a method of recording the movements which are executed in lifting
two boxes of egral wei;ht but of unequal size. We also had a way
of f'juring out how much energy was expended and found that an 0
uses more energy to lift the large box than he does in lifting the

small box. Thus, if the two unequally sized boxes that weifih the

same are lifted, the larger one will appear lighter because more

[.Jo

.1.
be

energy was utilized in liftinj

Charpentier, as described by Nyssen and Eourdon (10), ascribed
the phenomenon to a difference in tactile sensations. He explained

the illusion by positing the notion that since the largest object
would come in contact with a larger skin surface it would have its

»re

r.
’\__.

1"".

wei ht Spread over a 3ter number of sensorial recptors and there-

fore seem lighter.

7

Koseleff (3), after an extensive review of the research done
1“

on the SKI, prep sea as a working hypothesis ior an explanation of

epe ds on a number of

OJ

the illusion, that the experienced heaviness

nd the volume of the

on

factors; amongst others, the physical weifiht
object.

Huang (2) thilmls the there is a strikin: parallelism between
the problems, facts an t‘oeoretical issues of the SYI are the per-
ceptual constancies. He feels that an explanation o: the STI may

well lie in Wj not tion with the constancies . Thus, we have two

:03
'3

attempts at a connlete exr anation of the STI and two rather in-
complete, tentative explanations.

A sh ort perus Qal of ,he above reviewed experiments shows that
Sharpentier's explanation is inadequate when Flourney, as cited by
Koseleff (E), and hrssen and Bourdon (l‘) elicited the SKI without
tactual contact with the objects. The MU ller and Schuman theory
also falls short because as Seashore (ll) pointed out, even if the
0's are avar e of the excerin1er1tal situation, the SWI Jersists.
Huang (2) also demonstrated the insufficiency of this theory as he

showed that the STI is depe endent upon actus l sensory data and not

H.

deational, intellectual attitudes. Thus we are left wiuthowt a
complete explanation of the STI.

The purpose of this stud y is to focus on the re le of vision in
the SUI by studying its oc currerce or non-occurrence in congenital-

ious studies have only dealt with the role

<1

ly blind persons. Pre

i wit d It would seem that by

\J

(D

of vision by blindfoldin

*3
{D
O
'3
U)
0

using conyenitally blind persons one would have more success in

arriving at a clean-cut consideration of the 'isual factor in the

It is hrorn that sirhted ;eorle make the SCI judsment on tLe
basis of their 'isual, “inesthetic and tactual senses. Te would
expect the convenitally blind person (it) to behave differently
than the sirhted person (1) in a oil s1.tuation since t1e former
has never directly used his visua 1 me chanisr. The only vay SB
persons have contact with an object is throu h their kinesthetic

and tactual senses.

1-3
L3”

e purpose of this study is a: attempt to answer the following

,0
g:
(L)
U)
(—1-
Ho
13

S

P
K}

a. Can the ooiee t Trovide tie C? uitn some basis for the
oeneral behavior we c all t] e SJI?

b. Can th; 33 by his 11ines thetic and tactual senses develop
an orc lyr reaction to the density of an object?

0. Does the simhted nerson deveIOp his reaction in accord-
ance with the density of the object by use of si ht as well
as the ainestlzet’l c and tactual Senses. Cr will blindfold-
ing a s Lhted person make a difierence?

N.)

tarsus
SUBJECTS: The S's for this experiment were composed of three groups:
Group I, ten si hted persons (S); Group II, ten blindfolded sighted
persons (SB); and Group III, ten congenitally blind persons (CB).
roups I and II were students from Psychology 201 and

500 at Michisan State University. They ranged in a'e from 18 to 67

years. In all, there were 14 males and six females.
Subjects in group III were obtained from the Kichigan School

for the Blind. They ranged in age from 13 to 17 years. There were
six males and four females in this group. In selecting subjects

for this :POUp it was necessary that they had been totally blind
since birth, were 12 years old or older and had an average I.Q. of
100 or better. This was felt to be necessary since their results
were to be compared to college students and it was desirable to e-
liminate any comnlicatinf growth factors by having them above the awe

of puberty .

Cf

It turned out 1st their I.Q.'s ranted from 105 to 129 except
for one subject whose I.Q. was 91. It might be argued that the blind
group was not comparable to the other two croups. However, the cru-

cial factor was whether or not the groups used world be aware of the

iniprle (14) found that

STI. The investigator reasoned that since

the SEI was well deveIOped by the age ,f six and that Tyssen and

Eourdon (9) found the SII to occur in 39.17i of 600 cases between

the ages of 20 and 60 years, it was felt that the SLI would pro-

/‘ "rj

bably occur, if it were possible, with the subjects in the on group

who ranjod in aje from 15 to 17 years. The results indicated that

lO

APPARATUS: The apparatus consisted of two series of wooden cubes,

five in each series. The cubes in Seriis I were used to get the O

H

accustomed to makinfi \eij ht discrimi nat ons. Therefore, these cubes

n Series II

,.-

H-

were constant in size and varie in weight. The cubes
were used to Get a measure of the frequency of occurrence of the

“

SVI and they were thus constant in weijht but different in size.

:1-
14-

tandardiza

(f)

on of the Stimuli:

Since the ultimate aim of the e} rcr“wcrt vs to elicit the
SZI if possible, the goal in constructing the blocks to be used was

LI

to space the blocks so that they were readily discrinin able with re-
ference to size or wei: _ht. A wooden cube LaDPCDEd to be available

t1O L ches S(Lare. This cube was rel-

O)

that weifhed 55 grass ard we

-’_4

atively easy to manipulate and therefore it was decided to use it
and arothor one like it as the standard in each series of blocks.

As all he blocks in Series I were to he the same size they were
all nfla e txso inches square. A multiple of Taber's fraction (800 x

1/40) for lifting wei hts was used to give asproximate values in

weifht that would be readily discernible. in addition, this multi-

a

CT'

ain a COES'BHE fraction in weisht

:15
93
Ho
14

ole of Labor's fract ior would

between the five blocks. The followin wove the calculated sizes and

‘“D

vei hts for Series I:

Stimulus A - L10 “2153.
Stimulus E - 45 {ms.
Stishllim C - 57.6 fms. (StanCafid)
Stim1 lus D - €9.12 fins.
it'133u1‘13 E - E22094: "INS.
All two inc hrs square
Using tnese wei hts {s a guide an empirical invest Lation vas

's- f. - 1- " I \ a ’. w *7 r~ r'~ K's r 1 ' r~ P1 7' J- :
irc;ctco to CleOvCT L s,nio of values that could actually be used

for the purjose of this expirine
differently Lei nteu blocks to

of beakers with total

‘ T1)vfl‘(‘, C“
-.I\,. ." ‘JI‘

 

\‘ICDUIhb-CNIDI-J

' ‘1 ~\
‘VV.

(3 (C1

A criterion of 100% we

to be correct in their judgments

trials with either

.1 _- W ‘
cube shaheo h ooks.

one of the

60 resoectivsly were used is thi

folded throughout this grocedure since permitting them to see the

differently weighted beakers

'—
.J .3

Using
was the first discernible

-,

To determine the next

used as the basis for comparison since a scale of values was cesired.

It tas found that 57.5 gm. met

3 erployed which required

hand before a

discernible weight

the criterion when

11

the i hts to be used,

 

70.0 ”

W(j F H
{be

75.0 "
77.5 ”
50.0 "

the sxfiiflsets

I

of weiiht five out of five

value would be accepted for use

would cue them as to weight.

d it was found that 47.5 gms.

1

tifference that met the

the 47.5 gm. beaker was

-\

}.Jo

constructing many

beskors so that there was a series
fCllO'H-JS:
Total eirht
57.5 gms.
40.0 "
42.5 "
45.0 "
47.5 "
50.0 "
52.5 "
55.0 "
57.5 ”
60.0 "
53.0 R ”
V5.1 O V
35.0 ”
57.5 ”

n

Three male subjects E:cd 5o, 40 an
study. The subjects were blind-

criterion.

comnared with 47.5.

Lorkin: 1n the other oirection from 55 jms., 67.5 fms. and 77.5
.4 ‘- - ,4 ' 1 V .7 ~ A .
Qms. w re found to meet the ariterion.

rm. -. ,- - 4. Ln . 1 -. J- .. - ..
gne record of the three suogects UQE as follots:

\‘IOUCJHh-CJQ
\7‘3 QWQQ
0
010100101
\Iowjnbc».
\‘Iflcnx‘lxd
0
0101001 m
m
C
O
3
U)

0

Since the data shoved food regularity it was decided to aim
for these tei hts in makinf the blocks for Series I. After construc—

tion the actual weights as measured on an analytical scale were:

 

Stimulus ”eirht
a 77.19 Cms.
‘ 3 6'70‘3r7 "
I“ l:

[A] U K
l4:
Q
.
LI;
(3‘)

A

Thus all blocks were within .5 Sm. of the dc ired weights.

(.0

Th.

A

w-

s discrepancy was considered negliviole since it was found that
a 2.5 gm. difference in weight was not discernible by the subjects
on whom the empirical study u:s conducted.

‘ O

a similar investigaticn was conducted to determine tne Sizes

U)

of the blocks to be used in eries II. -S‘nce the standard of each

1

series of blocks had necessarily to be of equal size and weight a

nt in deter-

*4.

piece of wood two inches wide served as the starting 0

*d

mining the values to be used in Series II. Thus two 1LGh-S served

0

as the value above which and below which the other values would range.

Fifteen pieces of wood were cut at 1/15 incn intervals so that

they ranged in width from two inches to two and 14/16 inches. Ey

15
aat described above in determining the weights,
the 8's again being blindfolded, it was found that 5/15ths of an
incn was the difference needed in size in order for a person to de-
tect a discrepancy in the sizes of tv 0 blocks. This was true both

above and below the standard value of two inches. Afain the 100%

50% or 753

H,

criterion was used. If the conventional criterion 0
was used it would have narrowed tne ranre of sizes in the blocks and

SKI. This study

0
O
:3”
Cu
{:5
O
(D
(.0
O
m
(D
H
Ho
O
H.
C?
H.
P
' 1
Cf
,3"
CD

probably have reduced th-
was conducted on two males aged 25 and 52, and three females aged
15, 28 and 57.

The goal, with reference to the weights of the blocks in Series
II, was to set them to all weigh 55 gns. After construction the

blocks in Series II had the followinr dimensions and Vei hts:

 

Stimulus Size Weifht
A 1.375" or la/s" sq. 54.61 gms.
B 1.750” or 15/4" sq. 54.55 "
C 2.000" or 2" sq. 54.57 ”
D 2.312" or 2C/1e" sq. £4.62 "
s 2.es7" or 2 11/16 " sq. 55.79 "

All these blocks were within .5 gms. of 55 ans. except the larg-
est. These weight discrepancies from the desired 55 fms. can be con-
sidered neglisible since,.on the basis of the above mentioned in—
vestigation, a difference of 2.5 :ms. in we ei'ht cannot be detected
by human subjects when a standard weighing 55 ans. was used.

The cubes were constructed out of white pine wood and were made
of two to four pieces of white pine wood elued togeth or depending
upon the size requi red. To add the gross we if_hts lead was placed

in the center of the blocks. Then it was necessary to take auay

Veifht the bloclcs W re hollowed out correSpondingly. The blocks were

14

finished with one coat of wood filler, one to four coats of plastic

finish and one to four coats of hard wax dependino upon how much
acditional weijht Lad to be ailded to th e individual bloc 1%

A box 5" high, 7 5/4" wide and 24 1/2" lonf was constricted to
serve as a shield so that the si hted group could not see the man-

ipulation of the clocks by the ex per rimente The box was made out

'15
O

L

of 1/4" plywood and has a blotter strip attached to its too and to
its inside bottom to deaden sounds and to protect the surfaces of
the bloclrs. This box was also used as a stand for Group I and

Group II on which the experimenter placed the pair of blocks to be

compared on each trial.

34

P35333733: Each 0 in groups I and II was brou .t into the experimen-

51.

tel room, seated at the apparatus and was read the following dir c-

(D

tions:

I will present to you on each trial a p ir of bl oc1:s.
They will be placed on the tOp of the box th t you see or
feel before you. You are to take a block in each hand at
the same time, usinj onlv yor_r thumb and your forefinjer
to lift the bloc1c. Your part in this experiment is to
ansver the cuesti n: which of the two blocks jiv n to you
wei hs the most? You are not to move the blocks up and
down. Five practice trials will be tiven before the actual
experiment be irs. There will be two sets of 10 O Juoorcnts
each. A break will be Civen be ween the sets. I1 you are
‘ndecided as to which of the two blocks is heavier, just
guess. Are the ere any questions?

{11 (0

80211138

OJ

The directions for Group III were somewhat alter ed

‘

ne total procecure of that group was different. However, the

d‘

(1)

directions were e seLtially the some.

The psvc hophysica 1 method of const: nt st12nuli was used in

C)

presenting the stimuli to the subject. In both series block

(.0

U)
(D
"5
1-410
C)
U]

erved as the sta 25a rd acainst which the other blocks in its

15
ne basis of weight. To determine which hand of
the S was to be gwiv n the standard a half dollar was thrown 100
times brior to tne experiment. If it landed Leads, the °tcit rd
was presented to the subject's rifht hand and if it landed tails

the standard was presented to the subwect's left ha-nd.

9

he order

P3

‘
.5

fie basis of a

i

of presents tion 0: the stinuli was ran omized en t

Froups I and II Led to make a total of lUO jud'ments in
both Series I and II using a pincer lift. A pincer lift meant
that he blocks were to be l’fted with the thunzb and index fin-
ger. For Croup III the procedure was slightly altered. They had

ents in Series I followed by ISO judsments in Series II

0 1
L.
{Cu

0 j1
using the pincer lift. This was followed by 10 judfments in Series
I using the pincer lift and then they had 100 JL”"W8DTS in Series
II using the palmer lift. Fi D23 lly, they again made 10 judgments
with the pincer lift followed by 100 judgments with the fireSp
lift. Five minute breaks were given at the completion of each

way of lifting the blocks.

Fifure 1 shows the plotted lines for the occurrence of the
STI for the various groups. The lines were plotted by the use of
ed Z scores, a method described by Koodworth and Schlossberg
(16). The ordinate represents tne Z scores and the abscissa repre-

sents the stimuli used in Series II which were usei to el'cit the

P

Jomnarirq all groups with reference to the pincer lift on the
basis of the occurrence of the STI, it is seen that Group I was
highest, followed by Croup III and Group II. It is also seen that

the congenitally blind, Group III, had successively higher occurrences
of the SWI with reference to the three conditions of the pincer,
palmar and graSp lifts. In all three of these conditions they had

a higher occurrence of the SKI than the blindfolded-sifihted, Group

II, and also came progressively closer, but not quite up to, the

1

,erformance of the sifhted,

}

U)
c“.
D
93
d‘
c‘1'
£0

The graph also demonstrate

hi
(D
r:
C l"
CD
*5
C1"
{3"
(D
C:
Ho
U)
C)
"5
(D
J
$3
"5
O
:4

ft
SJ
(D

in the size of the stimuli, the greater is the occurrence of

.‘W
.i I.

U)

Statistical tests of sionificance between the various groups
were computed. The med’an chi—square test for two independent
samples as described by oiegel was used as the test of sig-
nificance. Ihe one per cent level of significance was employed.

Table I lists the comparisons made. It can be seen that the
results for Group I and Group II, both groups using the pincer lilt,

are in general Significantly different from each other. The same

was true for the comparison between Group II and Group III Wiere

Croup

Occurrence of the SKI

I - Sighted - pincer lift

 

Group II - Sighted Blindfolded - pincer lift

Group III — Congenitally Blinded - pincer 11ft
palmar 11ft

+2.0

+11)

'“ras

p lift

 

 

 

g

\

 

 

 

 

 

 

/ 1.

1

 

 

l

 

 

|

 

 

 

 

 

I?
1.57“ 1. 75"

in

2.31”

STIMULI

Figure 1

2.68”

17

Summary of Statistic

1

square Te

I'.
analys

 

Stimulus

of signif.

 

Group I
versus
Group II

Both pincer

F1 +4

r- “
La, ,.
l k} .. 3

 

Group II
versus
Group III

Both pincer

a
U3

U

‘\.« I

 

Group I
versus
Group III

Eoth pince

C4!
:b

(J
m

4o; approx.
39¢ "

'7' g H
005;)

 

Group I dA .37 e5; approx.
versus
CE .31 54 "

Group III

4
‘\
t; ‘

 

 

F4 f4 F‘ r4 F’ +4 I4 F4 l4 F1 +4 i4 F’ r4 14 Fa

Pincer vs palmar :D (.27
CE .00 997
Group I CA 4.27 5p
versus
Group III S? 5.2 10;
Pincer vs grasp ,3 .20 55; approx.
C E . OO {:79 ;i".--

 

 

 

 

 

Table

l

 

both prongs used tie p nécr lllto however, the corner son oetueen
Croun I and Sroup III, where drain boo: grvuns us—S the pincer lift,
is not sipn"ficantly diiferent. r’"h’s yes also generally true of the
comparisons of Group I apa’nst ‘roup III with reference to both the
palmar and the grasp lifts.

It ill be noticed that there “as a lien in the consistency of
of the statistical finiin s. ihi occurred in the conéarison of
if? in 'POU‘ I and Crown III in the pincer versus pa mar comperison
and as snown the result was significant at the I level. Since
LO “empaLisons wane vade one could expect at least one of the com-
jparisrns rode to be si nigi;ont, purely on the basis of ch nee.
'ihis possiiility is also enhanced by the fact tact 18P:€ N's uer
not reed. It is also seen \ith tnese groups that the two extrem
stimulus values, in and 33, were no vhcre rear sijnifieant an” it
follows tVst ,; O“ J- titld LeTE eten less chance of beia; sijrif-
icant. ‘kir these: ransors, iniis dis pWILGflSV'iIi'tha stmfithstical
finfin s may tell he cue to chance.

Another possibility is that ghere mipht have been a confound-
in“ effect at york Since Troun I used a pincer lift and Group III
used a ralmar lift. Chance, or a CCnfCUhTihf effect, nay also have

occurred then G

“raso crmnarison, sin
.L

. L. ,. '_ ."‘ . . \F‘ .’

for tgiS CumISFlSOu 1

n C)
v ‘4
,3‘1...‘
5 _ ‘3

4

‘7

v .

to iroup I; rers s

(

(J

' 1":
L.

ings

(W

'ectl}

The results of this exceriment gave answers to the questi”
set forth in the first “art of this naoer.
a. It was found that the object c:n provide tie con; enitally

er nd rer son with some basis for tr e general behavior we

b. The results also indicated that the confenitally blind per-
rinosthetic ta ctual senses, can develon an order-
ls reaction to the density of an object.
o. Finally, the sifnted yerbOJ, it seens, develons his reaction
to the SYI by the use of si ht as well as the kir esthe tic
and tactual senses, since blin‘foldirj a sifinted person ce-
creases th frequency oi occurrence 0
The general findinas show that in re ierence to the frequency
of occurrence of the STI the sighted, Group I, fave the hi hest

frequency of occurrence of the SKI; tie congenitally blinded, Group

III, were next; the blindfolded—sifihted, “roun II, were last.

(10) demonstrated that the STI could be obtained without the use
of the tactual sense. “he present study showed that the illusion
could be obtained nit}1out the use of the visual sense. However,
Jenacze (15) was not able to elicit the illusion when only the

of these

m
J
CL
LO
' 3
CO
H
A
N
( ')
1.1
[C

sider ation

O

tactual sense was involved. A
studies, alonf with th, others review. ed in the first 3art of this
paper (5,4,: 0), seems to show that tbe tactual sense, the visual

sense or the kinesthetic sense alone is not enough to elicit the

 

SKI. It seems that at least two of the senses must be involved in

21
order to get the illusion.
In the present study it Was found that the congenitally
blind group have a hi her frequency of occurrence of the SKI teen

’1

the blindfolded-sifihted nersoss. TLis result is underS‘endable 1n

CD
H;
0)
(3

view of th not that when one blindfolds a siihted person and asks

him to do t

CD

in which he would ordinarily use his sijht he will

0‘;

naturally be handicanped to a certain extent. The greater frequen-

cy of occt re ace of the SKI in tie con enita lly-bli nd .ed in this in-

14-

n effe

O

stance is not surprising oecause tLey, t, save been "blind-

H

folded all their lives Ehf have learned to utilize their tactual
and kinesthetic senses to a treater extent than the sighted person

who relies more on his sight.
In line with Huang's (2) experiment in which he got les SIT

by the successive reduction of the size cues, it was fou1d in this
study that by increa81no the amount of tactual and kinesthetic in-
formation made avai sole to the subject a higher frequency of occur-
rence of the S‘I was obta ined. This was demonstrated in the congen-
itally blind @roun as they gave successively hifiher frequency of
occurrence of the SEI under the three conditions of the pincer,
palmar and creep lifts. it may be that they obtained a higher fre-
quency of occurrence of the SYI not because of the increased amount
of sensory data but simrly because of the effects of learning. Un—
fortinately, the number of congenitally blind sub ects available
precluded the ossibility of directly testing this hypothesis. It
is probably safe to conclude that the increase in their performance

was in all probability due to a combination of both factors.

A question that mi ht well be xcrth investi atins is if more

22
tactual and kinesthetic information is made avialable to the blind-
folded-sijhted person, would he than five a frequency of the occur-
rence of the illusion more comparajle to the congenitally blind per-

son? This could readily be done by gettinj two more blindfolded-

H.

s éhteu groups and have one :POUp utilize the palmar lift and the

other froup utilize the grasg lift. Also to check on the possible

\5

influercing effect of lezrninf, one mirtt test two alditional

r
(W

W

froups of congenitally blind persons who would use the palmar and
the grasn lifts respeetively.

In summary, the major findir: of this study is that the S*I

person haniicans his ability to be atare of the STI.

QTrv" ‘n 1:1 r
N1= - t-‘ILLL-A.

An eyneriment was conducted to determine: 1) whether an ob-
ject can provide the concenitally blind person with some basis
for the general behavior we call the STI; 2) whether the congen-
itally blind person, by his kinesthetic and tactual senses, can

a 1 a

veIOp n orderly reaction to the density of an object; and 5)

w

F1
1’")

0

whether the sighted person develOps his reaction to the STI by the
use of sight as well as the kinesthetic and tactusl senses.

There were three groups of ten observers each; a sinhted group,

he

F1

folued-si~hted group sLd a confienitally blind group. T

Q;

a blin
apparatus consisted of two sets of five blocks each. One set was
desigred to oe all of equal size but with different weifhts. The
other set was to be all of equal we1ght but different in size. One
hundred comparisons with each series of blocks was made by the

method of constant stimuli. The sijhted and blindfolded-sighted

ment using the pincer lift. The congen-

Ho

groups performed the egper
itally blind group, in adcition, had to rake lOO judjments with a
palmar lift and 100 judgments with a firesp lift.
The r sults indicate that the following conclusiOLs can be
drawn:
a. The object does provide the congenitally blind person with
eneral behavior we call the STI.

some basis for the

(‘3
kg

(D
(71'
[—10
O
C“)
*3
,2
ct
(D
0
Ci‘
C
{D
H

b. The congenitally blind person, by his kinesth

senses, can deveIOp an orderly reaction to the “ensity of an

Object.
o. The sifhted person develops his reaction to the STI by the
use of -i ht as well as the kinesthetic and tactual serses.

1.)

1‘0

4

scussion of the variables influenCing the occurrence or

8‘

W
Q
P-

non—occurrence of the STI was Siven together With some SUTTeStiOUS

for future research.

1.

C51
0

6)

C13
0

L0

10.

}..J

chol., 1957, 15, 242-25

Doll, 3., The Demoor Size- eight Illusion. Traininh School

 

Ful1., 1915, IX.
Huang, 1., The Size- eight Illusion in Relation to Percept-

ual Constancies. J. of Gen. Psychol., 1945, 55, 45-65.

 

Huang, 1., The Size—Height Illusion and the Teight-Density

Illusion. J. of Gen. Psychol., 1945, 35, 65-84.

 

Koseleff, P., A Jodification of the Chareentier Effect.

Psychol. Porsch., 1939, 21, 142-145.

 

Foseleff, P., Studies in the Perception of Heaviness I, Some

relevant facts concerning the size-weight effect. Acta Psv-

m

Kcseleff, P., Studies in the Perception of Heaviness II, Con-

(15

cerning the ouantitative determination of the size-weight ef-

fect. Acts Psychol., 1958, 14, 109-130.

 

 

Psychol. Rev., 1907, 8, 534-3a8.
Heyers, C. S., A Textbook of Exierimental Psycholosy, Yew Yorgz

 

Lonfimans, Freen, 1928, 207.
Fyssen, 3, and Bourdon, J., Contribution to the Study Of the

Size-Weight Illusion by the Ie.hod of P. Koseleff. Acta

Psvchol., 1955, 11, 467-474.

w

szsen, . and Bourdon, J., A flew Contribution to the Exoer-

L

Q

imental Study of the size-Hei;ht Illusion. Acta Psychol.,

 

1995, 12, 157-175.

seashore, 3. 5., Keesurenents of illusions and hallucinations

)3

in normal life. Stud. Vale Psychol. ab., 1395, 3, l- 7.

 

 

‘ A . -v- n ‘v A -. f: 1.- F '. . ( N .
12. bl€_el, 3., .onnfrenetric gottlfitléfi for the eFSVJoral
1‘ —° 1"- -~ 1" - " ,- 1'?" r: ,-. . 'r 5' 2
nciences, reL lork: Howrah—1111 nook companV, inc., lQe ,

 

1: TTfiw-N r ’7 Q '\ “(-1)" '3 c) 1 q 1. “ f ,3 J . r7 " A, 7 "at V1.11) . f‘ Y\ t‘ \ 1. 4'
9“. . \JbLiQLAZV’ .J. , Jll¢lel -.|._Lrl\-fl l-.v “la—'8... (I; J. *USlK/L— CkllC‘ LLIS
. -- ”‘ .w L. ' " t‘ P " 3 "
“1,10 ues. ‘SQC C1. lesser., l9ol, li, set-U79.

 

JV
'5
I;
H
r)
f—U
\D
.3
Cr
Cu
1...!
‘ 'J
”3‘
f
A
4
(.0
H.
’J
\1
H
r-f‘
\3
CO
CT
(0
V
' '21
£3
*5
cf
1 4
H
0

14° "53173318, rm», w --

 

15. toodworth, R. 3., Experitent€l 3333r91013’ lew York: Henry
Eolt and Joméanj, 19?3: 440-441'

1 7*

16. Koodworth, R.3., and Scnolossberg, d., Exnerimental Psychol-

Osv, Few York: Henrv Folt and Company, 1954, 205-206.
I,

R3355 {33}: MY

MICHIGAN STATE UNIV

llllll If

31293 304

R I

11111111“
6 6514

O