ll

3

145

 

THE RELATKQ‘K 9F CfiRTé-JN TARGE? FACE‘QRS
TO PHENC‘IMEML DESTANCE

Thesis for ”is Degree 05 M. A.
MICHIGAN STATE UNIVERSITY
Richard Warren Thompson

1959

 

PLACE ll RETURN BOX to roman this checkout from yam record.
To AVOID FINES mum on or bdoro duo duo.

DATE DUE DATE DUE DATE DUE

 

ygtvrj

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I. An Afflnncflvo Halon/Equal Opponunlty lnstltwon
Wanna-9.1

 

‘40411‘1

‘ H4ullflli'

 

.5?!”

 

THE RELATION OF CERTAIN TARGET FACTORS
TO PHENOMENAL DISTANCE

BY
Richard Warren Thompson

A THESIS
Submitted to the College of Science and Arts
Michigan State University of Agriculture and
Applied Science in partial fulfillment of
the requirements for the degree of

MASTER OF ARTS

Department of Psychology

1959

THE RELATION OF CERTAIN TARGET FACTORS
T0 PHENOMENAL DISTANCE

BI
Richard Warren Thompson

AN ABSTRACT
Submitted to the College of Science and Arts
Michigan State University of Agriculture and
Applied Science in partial fulfillment of
the requirements for the degree of

MASTER OF ARTS

Dopartment of Psychology

1959

Richard Warren Thompson

The present study was designed to investigate the
effect of picture size on the phenomenal distance of objects
in pictures and, the effect of which way a man is facing in
a picture on the phenomenal distance of that Ian.,

A psychophysical method was used that enabled an
observer to make comparisons between a small fixed standard
target and larger comparison targets. By turning a crank,
an observer could move the comparison target to a position
so that it was phenomenally equal to the fixed target. The
distance betweendhe observer and the comparison target
was a measure of the phenomenal difference in the two
pictures.

Five photographic prints of increasing size of each of
two scenes were used as comparison targets. A.smaller print
of each of these scenes was used as the standard target.

One scene depicted a man standing in a road facing the viewer,
the other depicted the same man.with his back to the viewer.
Each standard target was compared only to comparison.targets
that depicted the same scene.

Eighteen observers made five comparisons with each
comparison target, or a total of twenty-five comparisons
with each standard.

Specifically this experiment was designed to test the
hypotheses that as picture: size increases, the apparent
distance of objects in the picture would decrease, and
that a man in a photograph with his back to the viewer will

appear closer than a man facing the viewer.

Richard Warren Thompson

If the hypotheses are true then the observer would place
the larger targets farther from him than.the smaller targets,
and the photograph depicting a man.with his back to the
viewer farther than the photograph of a man facing the viewer.

The experimental results supports both hypotheses. As
target size increased, the targets were moved farther from the
observer. The targets with the nan.with his back toward the
viewer were moved farther away than the targets showing
a man facing the viewer. The distance obtained for both
scenes were greater than would be predicted by the law of
visual angle but did not demonstrate 'letric indifference'l
that would be expected by the law of constancy. That the
results were a compromise between.the two laws was rejected
on the grounds that there can.not be a compromise between

a finite distance and an infinite number of distances.

Approved by: 3““ am:

\
Date:JhHu"‘fi

 

ACKNOWLEDGMENT
I deeply appreciate the continuous help and
constructive suggestions given by Dr. 3. Howard

Bartley during the formulation and execution of
this research.

I also wish to thank Dr. Charles Henley and
Dr. Stanley Rather for reading the manuscript and
offering helpful criticism.

RJLT.

11

TABLE OF CONTENTS

LIST OF TABLES...........................................1v
LIST OF FIGURES.......................................... Y
I INTRODUCTION........................................ 1
II METHOD.............................................. 7

Subjects......................................... 7

Apparatus........................................ 7

Procedure........................................lO
III RESULTS.............................................13
IV DISCUSSION..........................................19
V SUMMARX.............................................23
BIBLIOGRAPHI.............................................2u
APPENDIX.................................................26

111

TABLE 1

TABLE 2

TABLE 3

LIST OF TABLES

Obtained Mean Distance in Inches and

-Distance,Predicted by Law of Visual

Angle for Each Comparison.Target...........16
Summary of Analysis.of Variance for.the
Transformed Data Based on the Median

of Each S for Each Target..................l6
Confidence Interval at the One Percent

Level for the Mean of Each Target Based

on the Transformed Data and the Square

Root of the Distance Predicted by the

Law of Visual Angle for Each Target.........18

iv

LIST OF FIGURES

Figure 1 Sample Target Photographs......................9
Figure 2 Picture of the Appartus........................9
Figure 3 Graph Showing the Relationship

Between Target Size and Phenomenal

Distance and the Racing of the Man

in the Target and Phenomenal Distance.

Phenomenal Distance is Indicated by

the Distance in Inches Each Target was

loved. Distance Predicted by the Law

of Visual Angle is also Plotted...............14

.OOOOOO

INTRODUCTION

Painters and photographers have for many years manipu-
lated the factors that determine the apparent distance of
elements in paintings and photographs to achieve the effect
of distance they desired. Only resently has psychological
investigation of some of these factors been undertaken.

Amos (3) and Schlosberg (10) have listed several ways
in which the depth effect in pictures can be enhanced.

They suggested, to give only a partial list, viewing the
picture from a distance, monocularly, or through a small
hole.

Gaffron (9) presented an introspective analysis of the
difference in phenomenal distance of elements in.paintings
due to their lateral orientation. She presented an 'empirical
glance curve' as a partial explanation for the apparent
difference in the distance of objects in.a picture and its
mirror image. The object in the left half of a picture
was seen as closer than an object in the right half. Adair
and Bartley (l) and Bartley and Thompson (7) have investigated
this problem of the phenomenal difference between the left
and right in photographs by a psychophysical method. Adair
and Bartley presented their 38 with a picture and its mirror
image. They used several scenes of varying asymmetry, the
picture and its mirror image serving alternately as the
standard and comparison target. The standard picture was
always smaller than the comparison picture. The 38 moved
the comparison target out a track until it appeared the same

distance from him as the standard.

2
The results indicated that the left half of .‘ photograph was
seen as closer than.the right half. Bartley and Thompson
using the same psychophysical method systematically varied
the position of a man in a photograph from left to right
and obtained essentially the same results.

The effect of viewing distance on the apparent distance
of objects in pictures has been studied by several investiga-
tors using different methods. Smith (11) had his Ss view a
large photomural of a corridor monocularly, their view was
restricted to the image area of the mural. The 35 estimated
how many paces from their viewing point to a point in the
corridor depicted in the mural and from that point to the end
of the corridor. The results indicated that as the distance
from which a picture was viewed decreased, the apparent
distance of objects in the picture decreased,‘;..g., the 8s
estimated more paces for the photograph viewed farther away
than for the one viewed closer.

Smith (12) in another study presented his 83 with a
photograph of a field with several stakes in it. His Ss
viewed this photograph monocularly from each of two positions
and estimated the distance in yards to one of the far stakes.
The 83 estimated the stake was closest to them in the photo-
graph that was viewed closest to them. Smith and Gruber (1h)
used another method and obtained substantially the same re-
sults. In this study, a photomural of a corridor was viewed
monocularly from five different distances. This was the

comparison target.

 

The standard target was the actual corridor depicted in

the photomural. The Ss made ratio-judgments of the apparent
depth of the mural in comparison.withthe corridor. As the
photomural was moved closer to the eye of the 8, its apparent
depth decreased.

Bartley and Adair, (6) using the same phychcphysical
method described earlier by which a 3 could move a comparison
target until it was perceptually equal to a standard target,
get results in agreement with the workers mentioned above.
These investigators presented their 38 with.a small standard
target at each of three distances. The Be moved each of three
larger comparison targets out a track until they appeared
equal in distance to the standard. They found that as viewing
distance decreased so did the distance at which the comparison
target was placed, it appeared closer.

The effect of picture size is yet another variable that
has been investigated. Smith, Smith, and Hubbard (13) used
the following method to investigate this factor. The Be in
this eXperiment viewed four targets, a photograph, a line
drawing with full detail, a shaded line drawing, and a line
drawing, all of the corridor depicted in the photograph.

A slide was made of each of these targets and projected on

to a screen from each of five positions. These were the com-
parison targets. The standard was a slide of the photograph
projected from the middle distance of the comparison targets.
The Se made ratio-judgments of the distance to the end of the

corridor in the comparison targets in relation to the standard
target.

4
The results of the experiment indicated the apparent distance
of the end of the corridor increased as the projected image
size decreased. Ratio-judgments were greater for the slide
projected from near the viewer than for the far position.

The obtained ratio-judgments were significatly different

than those predicted by the law of visual angle. They found

no consistent differences due to the amount of detail in

the targets but this result was obscured because the slides

for the various targets were found, subsequent to the experiment,
to be of different sizes.

Bartley and Adair in the study mentioned earlier not only
varied viewing distance but also target size. They used four .
photographic prints of the same picture, a 1.x 1, 3 x13, u:x h,
and a 6.x.6 inch enlargement. The l x 1 served as the standard
and the other three the comparison targets. They found that
as target size decreased apparent distance of objects in the
target increased. This is the same kind of results as ob-
ltained by Smith, Smith, and Hubbard but by use of a different
method.

Bartley and Adair also indicate that their results do
not follow either the law of visualiangles or the law of
constancy. The law of visual angle predicts a definite
distance for each target, the law of constancy predicts, that
the targets would appear equal regardless of their distance
to the eye. That is, any place on the track would be equally
good for matching the apparent distance of the objects in
the comparison target with objedts in the standard (“,5).

5
The results they obtained range from 76 to 88 percent of what
is eXpected by visual angle. To say that the results are
a compromise between the two laws is not logical. They say,
“How can a numerical compromise between.a fixed value, and
an infinite number of values all of which.are equally good
(constancy) occur?' It appears as if the law of constancy
is not the salient factor involved.

Another factor mentioned briefly by Gaffron concerns
the disposition of objects in pictures. In her analysis of
Rembrandt's Return 9; Egg Prodigal Sag she stated that the
sense of nearness of the figure in the left foreground,
kneeling with his back turned toward the viewer, was in
part due to the sense of personal identification with the
person.with his back turned toward the viewer. She says,

‘A person standing in the left foreground with
his back turned toward us arouses a decided feeling
of identification with ourselves, because his
position comes nearest to the one we assume as
spectators. For the same reason we feel that a
person looking out of the picture from the left
foreground is directly apposed to us.

To summarize, five factors have been investigated that
have been found to influence apparent distance in pictures.
The overall depth effect in a picture can be increased in
several ways, one of which is viewing the picture monocularly.
The left side of a picture is seen as closer than the right.
Decreasing the size of a picture or increasing the distance
from which it is viewed will increase the apparent distance
of an object in a picture. It was also suggested that a

front view of a person in a picture will appear farther

from the viewer than a back view.

These findings must be taken into consideration in
designing any experiment to.investigate factors involving
apparent distance in pictures.

The present study was designed to investigate the effect
of picture size on the apparent distance of objects in pic—
tures and Gaffron's statement that the direction a person
is facing in a picture will, in part, determine his apparent
distance. Thelspecific hypotheses to be tested are:

a. As picture size increases the apparent distance of
objects in the picture will decrease.

b. That in.a picture depicting a man.with his back
turned toward the viewer the man.will appear closer than

in.a picture showing him facing the viewer.

METHOD

SUBJECTS: The Ss for this experiment were 15 men and 3
women students fromman introductory course in psychology.
The 88 were volunteers and received class credit for par-
ticipating in experiments. All the women Ss were 19 years
of age, the men Ss ranged in age from 19 to 29 years. Five
additional 89 participated in the experiment but their data
are not included in the analysis because more than three
of their judgments for any one target were beyound the range
of the apparatus. None of the Ss had any knowledge regarding
the purpose of the experiment. To discourage questions
about the eXperiment, the Ss were told it was part of a
larger experiment to be explained later.
APPARATUS: Six photographic prints of eachcnf two scenes
were used as stimuli or targets in this experiment. In
Fig. l is a print of each of these scenes. The scene in
each case depicts a man standing in the middle of the road
and are symmetrical except at the extreme left and right
edges. The only difference in the two scenes is that in
one the man is facing the camera and in the other his back
is turned toward the camera. The photographs were taken
with a # X.5 Speed Graphic camera with.a 135 mm Graflex
Optar lens. The photographs were taken at a height of five
feet two inches above the ground. The man was 52 feet
from the camera.

.A 2 X 2 inch crOpping of both the original negatives

exactly centering the man in the scene was used to make a

8
l X 1. inch reduction, a 2 X 2 inch print, and a 3 X 3, 1+ X 4,
6‘X 6, and a 8.x 8 inch enlargement on white matt photographic
paper. The l X.l print for each scene was called the
Front and Back standard. The larger prints were the com-
parison targets. Each of these prints had a 1/8 inch
white border and were mounted on.a plywood standard painted
flat black so that a 1/8 inch black border surrounded the
print. This made the over-all size for each target greater
by one half inch.

In.Fig. 2 is a picture of the main part of the apparatus
with a blank standard and comparison target in the positions
as they would appear to the S. The main part of the
apparatus was a 280 inch track along which adjustments were
made. .A movable rider on the track held the comparison
targets. This rider could be moved, by turning a crank
on the left side of the viewing stand, as close as 21 inches
and as far as 276 inches from the eye. Standard targets
were mounted on the right side of the viewing stand lk inches
from the 8‘s eye. The track was calibrated in one inch
intervals.

On the 3 side of the viewing stand was a chin rest and
nose slot to assure each S had his head and eye in the
same position in relation to the two targets. Both targets

could be seen at the same time. An eye shield for either
left or right eyed viewing assured monocular viewing and
prevented the S from seeing the track.

 

 

 

 

 

Fig. 2

Picture of the Apparatus

10

Both the standard and comparison targets were at the same
level no matter where the comparison target was on the track.

The S was seated in an ordinary chair so that he could
comfortably place his chin in the_chin rest. No attempt
w as made to limit the peripheral vision of the S.

The targets were viewed against a flat black background.
The experimental room was lit by four equally spaced in,
candescent bulbs with diffusion bowls which illuminated the
targets equally anywhere on the track.
PROCEDURE: Each 3 was brought into the experimental room,
seated at the apparatus, and given the following instructions

to read:

This is an eXperiIent in distance judgment.
You will be presented a small standard target on
the right and several comparison targets on the
left. IOu are to place your chin in the chin-rest
with your nose in the slot and view these targets
with one eye (the eXperimenter will tell you which
eye to use.) By turning the handle on the left
side of viewing stand you are to move the comparison
target to such a distance that the man in the stans
dard and comparison targets appears the same dis-
tance from you. Ybu will make five practice judgments
and then 50 the experimenter will record. Are
there any questions?

If the 3 had any questions concerning the procedure,

the E clarified them by paraphrasing the original instructions.
Each S used his dominant eye for viewing the targets.

Eye dominance was determined in the following way. The B

told the S to line up his finger, using both his eyes,

with a vertical line some ten feet away. He was then told to

close his left eye being sure not to move his head or eyes.

11

If the finger remained lined up, he was considered right eye
dominant and used his right eye in making judgments. If
the finger did not remain lined up, he was considered left
eye dominant and used his left eye to make judgments.

The practice comparisons were made with he same standard
that was used first in making the recorded judgments.
The comparison targets were presented starting with the
smallest and ending with largest for these practice judgments.
The standard used first varied randomly from S to 8.

After the five practice judgments, the E answered any
further questions the S had concerning the procedure. The
S then made twenty-five judgments, five comparisons with
each of the comparison targets, for each standard. .A cross
comparison between Front and Back Targets was not made.
Each 8 completed all his judgments with one standard before
making any judgments with the other. The comparison targets
were presented in a random order with the exception that
the same comparison target was never presented twice in
succession.

After each judgment had been completed, the E moved
the rider on the track to a different position so that the
8 never moved the target just away from or just toward
him to make judgments.

The assumption in this procedure was that an 8 would
be able to move a comparison target on the track to a pos-
ition.that would equalize the apparent distance of the man
in the two targets.

The measure of apparent distance was the distance in
inches from the S to the comparison target.

The E recorded each judgment on a data sheet. The 8
had no knowledge of his results.

12

RESULTS

Fig 3 shows the mean distance in inches for each
target for the two sets of five targets, Front and Back,
for 18 Se. The distance predicted by the law of visual angle
for each target size is also plotted in this figure. The
two curves drawn for the observed distances were fitted by
inspection.

In Table l the observed mean distance for each target
and the predicted distance by the law of visual angle for
these targets are tabled. These means are based on the
median of five judgments of each of 18 88 for each target.

Of the 18 Se, 11 were right eye domimnt and 7 were left
eye dominant. The two groups were combined for analysis
after an inspection of the data indicated that the only
difference between the two groups would be in absolute
distance for each target. The mean distance for the left eye
dominant 88 for most targets was greater than for the right
eye dominant Ss. This could be accounted for on the basis
that the left eye Ss viewed the standard target from a slightly
greater distance and the comparison target from a slightly
lesser distance than the corresponding targets for the
right eye dominant Ss. In view of the influence of viewing
distance on the apparent distance of objects in pictures, the
results obtained were the results expected. The conclusions
drawn for the two groups separately would be no different

than for the two groups combined. (2)

14

Graph Showing the Relationship Between.Target Size
and Phenomenal Distance and the Racing of the Man

in the Target and Phenomenal Distance.

Phenomenal

Distance is Indicated by the Distance in Inches Each

Target was Moved.

Distance Predicted by Law of Visual

Angle is also Plotted

 

an! -
b}
‘e/m a
k
g /7.9’-*
\\
{R My -

6

Mia

£30 -

i
c,
J

5/57/7082” 7? cage/way 736’
b. \I N h
B .
‘1‘ . ‘7‘ .

a
?

 

3;

-—-—~ fey/77' /
>é‘””€x. .AQQKZK /’/

/
X /
o———+ [6/571 i/Ké'li //
/

 

 

1 I I I
are 31.? 44 éxé

7773657' i/Zi

iii

Flge 3

 

15

The data (using the median of each S for each target as
a score) were tested for homogeneity of variance by Bartlett's
test and.were found to be heterogeneous,.x2 = 58.15, df = 9,
PC .01. The data were transformed by the square root trans-
formation and again tested for homogeneity of variance by
Bartlett's test and were found to be homogeneous,.x2 3 8.55,
df - 9, P)».30. (8) The transformed data were analyzed by
analysis of variance and a summary of the results is presented
in Table 2. (2,15)

As can be seen, the statistical analysis of the data
supports the graphic presentation. .As target size increased
the apparent distance of the man in the target decreased
and the target was placed farther away from the S. The effect
of target size was significant, F a 671.31, df - fl and 68,
P<.01, (the S x 0 interaction was the error term). The
significant interaction of Sizes with Observers, F I 3.25,
df = 68 and 68, P<.Ol (the error term was the S X F X 0
interaction), indicates there was a differential effect of
target size for different Ss. Although the ranking of the
targets was the same for all Ss, the distance betweenthe
targets was not.

The difference betweennthe two sets of targets, Front
and Back, was also significant, F = 5.50, df 8 1 and 17,
P{.05 (the F X 0 interaction w as the error term). The
man with his back to the Ss was seen.as closer and therefore

moved farther away from the S than was the man facing the S.

TABLE 1 , 16

Obtained Bean Distance in Inches and Distance Predicted
By Law of Visual Angle for Each Comparison Target

 

 

 

Target Size Front Back Visual Angle
2 X .2 53.1 58.? 28
3 X 3 82.5 88.3 92
b X it 108.9 116.1 56
6 X 6 156.5 167.6 81+
8 x18 6 201.0 205.5 112

W

TABLE 2

Summary of Analysis of Variance for the Transformed Data
Based on the Median of Each 8 for Each Target

W

 

Source , df HS 1"
Sizes ll 261.81 671.31"
Pacings 1 M40 5.50*
Observers 17 17.1!!! 1&5.ijM
8 X F 4 0.06 0.50

S X 0 68 0.39 3.25**
1? x o 17 0.80 6.67“
3 X F X 0 68 0.12

Total 179

*fiignificant beyond the '1 percent point.

*Significant beyond the 5 percent point.

~
\

q"

a .

u-oc.

v

 

 

17
The significant interaction of Facings with Observers,
F 3 6.67, df =- 17 and 68, P< .01 (the S X F X 0 interaction
was the error term), indicates that not all the 8s saw the
Front and Back targets in the same way. An inspection of
the data reveals that ten 33 placed the Front targets closer
to them than the Back. Two Ss placed the Front targets
farther from them than the Back. Four 88 placed four
of the five Front targets closer than the Back and one 3
placed four of the five Front targets farther than the Back.
The other 8 was not consistent in.which target was placed
closest.

The interaction between Target Size and Facings which
was not significant, F = 0.50, df’8 4 and 68, indicates
that the curves plotted are essentially parallel. (The
S X F X' 0 interaction was the error term.)

The significant difference between Observers, F = 1n5.33,
df 8 l? and 68, P<L.01 (the SLX FIX O interaction.was the error
term), indicates that the 3s differed in their absolute
response to the targets.

In Table 3 the confidence interval, P< .01, for the
mean of each target is tabled along‘with the distance
predicted by law of visual angle for the transformed data.
This confidence interval was computed using the Sizes by
Facings by Observers interaction as the error mean square. (2)
It can.be seen by this table that the confidence interval for
each target does not include the distance predicted by the

law of visual angle but is greaterwthan.that distance.

18

TABLE 3

Confidence Interval at the One Percent Level for the Mean
Of Each Target Based on the Trans formed Data and the
Square Root of the Distance Predicted by the
Law of Visual Angle for Each Target

 

 

Target Confidence Interval Visual Angle
2 X 2 Front 6.99 -- 7.41 5.29

2 X 2 Back 7.37 -- 7.79

3 X 3 Front 8.80 -- 9.22 6J8

3 x 3 Back 9,10 -- 9.52

l} X b Front 10.11! - 10.56 7.1!8

4 X ~ Back 10.1-15 - 10.87

6 X 6 Front 12.25 - 12.67 9.17

6 X 6 Back 12.61) - 13.06

8 X 8 Front 13.85 - 1#.27 10.58

 

DISCUSSION

The results of this experiment supports both of the
hypotheses stated in the first part of this paper.

a. As picture size increases the apparent distance of
objects in the picture will decrease.

b. A man in a picture with his back turned toward
the viewer is seen as closer than a man in a picture facing
the viewer.

In.a very consistent manner 18 untrained Ss placed
each target progressively farther from their eye as the
target size increased. The distances obtained were sig-
nificantly greater than would be predicted by the law of
visual angle. ‘

The decrease in apparent distance with the increase in
target size are in keeping with.the results obtained by
Smith, Smith, and Hubbard although a different method was
used in each study. These investigators also found their
results significantly different from visual angle predictions.

It would be expected that the results obtained inthis
experiment and the results obtained by Bartley and Adair
would be consistent since almost the same method and targets
were used in both studies. The general conclusion of
decrease in apparent distance with an increase in target
size is the same in both studies. The results of the

two studies differ in the absolute distance at which the

targets were placed.

20

Bartley and Adair obtained distances that were less than
those predicted by the law of visual angle while the results
in the present study are significantly greater than visual
angle predictions.

There are several differences in the two studies. One
is the distance used for the standard target. Bartley and
Adair used 16, 24, and 32 inches, the present study, 1h inches.
From the results mentioned earlier concerning the effect of
viewing distance on the apparent distance of objects in
the picture, it would be eXpected that the Se in Bartley
and Adair's study would place their targets even farther
than the Se in this study.

Another variable that was different in the two studies
was the number of comparison targets used.

In both studies there were significant differences between
Se and the S variability was great. These may be the factors
that account for the difference in.the results.

As was mentioned before, the generalization of the
relation of picture size to apparent distance in pictures
is not changed by the different results in the two studies.

Nor does the result of this study contradict the
analysis Bartley and Adair gave their data concerning the
law of visual angle and the law of constancy. The results
of this study do not fit the law of visual angle nor do

they fit the law of constancy, i. 3., “metric indifference'.

21
If the law of visual angle had been followed, the 88 would
have placed their targets near the points predicted by this
law. Only one S's data was close to these predictions.
The other 83 made judgments greater than the law of visual
angle predicts.

If the 33 had been following the law of constancy,
their results would have indicated not one consistent
distance but I'... that no matter where the comparison
target was put, or no matter what its size, the scene in
it always looked as far away as the same scene in the fixed
reference print.‘ (5) The Se in this eXperiment did not
show this 'metric indifference.“ They were able to make
the judgments in a very systematic way.

Regardless of what the crucial factors are in explaining
the results obtained in these two studies, it is apparent
that one is not a compromise between.the two laws nor does
the law of constancy predominate.

The finding regarding the second of the two hypotheses
is quite in line with.the introspective analysis of Gaffron
mentioned earlier. That is, a person with his back turned
toward the viewer is seen as closer than a person facing the
viewer in a picture. Because the results are based on only one
picture, any generalization would be premature although
it might be suggestive for more research with this variable.

The object that was reversed in the present study
occupied a dominant place in the picture as a whole. ibuld
the same relationship hold for objects that do not occupy

such a dominant place?

22
Gaffron has suggested that the difference in the front and
back in pictures is due to the viewer's sense of identifi-
cation with the person depicted. would this sense of
identification hold for animals or inanimate objects?
Further research is necessary before any generalization
about the front and back in pictures can be made with any
confidence.

To summarize briefly, the following factors have been
found to enhance the nearness of objects in pictures.

a. Having the object in the left foreground.

b. Viewing the picture from close rather than far away.

c. Using a large picture rather than.a small picture.

d. Show the rear view of anobject rather than a

front view.

A variable that has not been investigated but one that
might well be a determinant is the size of the object in
relation to the rest of the picture. In the present study
the size of the man increased in the same amount as the whole
photograph. What would be the effect of increasing the size
of the man, holding picture size constant?

SUMMARY

An experiment was carried out to determine (1) the
effect perceptually of increasing photograph size on the
apparent distance of objects in the photograph, and (2)
if there was a phenomenal difference between the front and
back view of a man in a picture.

Eighteen observers were asked to compare a small fixed
standard picture with each of five larger comparison pictures,
enlargements of the small standard, (in which the crucial
object in the picture was a man standing facing the observer,
or standing with his back turned to the observer.) The
observers could move the larger comparison targets toward or
away from them by turning a crank and in this way set them at
a distance perceptually equal to the standard. In each com-
parison, the scenes were identical except for the size of the
pictures.

The results indicate the following conclusions can be,
drawn:

(1) As picture size increases the apparent distance of
objects in the picture decreases.

(2) A.man in.a picture facing the viewer is seen as
farther away than a man in a picture with his back turned
toward the viewer.

(3) The distance obtained for each target does not
follow either the law of visual angle or the law of constancy--
“metric indifference“.

A summary of the variables influencing the apparent distance
of objects is pictures was given together with.some suggestions

for future research.

BIBLIOGRAPHY

1. Adair, H. J., and Bartley, S. H. Nearness as a Function
of Lateral Orientation in Pictures. Percept. and 41.
Skills, 1958, 8, 135-141.

2. Allen, T. Personal Communication.

3. Ames Jr., A. Depth in Pictorial Art. g3; Bulletin,
Sept. l925-Jun. 1926, 8, 5-24.

1}. Bartley, S. H. Pringiples of Perdgptiog, New York:
Harper 8: Brothers, 1958, 181-187.

5. Bartley, S. H. What is Perceptual “Constency?“ In
manuscript.

6. Bartley, S. 8., and Adair, H. J. Comparison of Phenomenal
Distance in Photographs of Various Sizes. 11. _9_f_ Psycholu
1959, “7, 289-295-

7. Bartley, S. H., and Thompson, R. W. A Further Study of
Horizontal Asymmetry in Perception of Pictures. Percept.
mg 1193;. Skillg, In Press.

8. Edwards, A. L. Eggrimental Design in Psycholggica;
ngearch, New York: Rinehart and Company, Inc,, 1950, 195-202.

9. Gaffron, H. Right and Left in Pictures. 51;: Quarterly,
1950, 13, 312-331.

10. Schlosberg, H. Stereoscopic Depth From a Single Picture.
Aggy. .1. Psychol., 19111, 51!, 601-605.

11. Smith, '0. H. Depth in a Photograph Viewed From Two Distances.
Percgpt. sad 593;. 1151.115, 1958, 8, 79-81.

12. Smith, 0. H. Judgments of Size and Distance in Photographs.

m0 £0 MO: 19589 71: 529-538e

1..

13.

19.

15.

Smith. 0. W., Smith P., and Hubbard, D. Perceived Distance

as a Function of the Method of Prepresenting Perspective.
Ignaz..§. Psychol., 1958, 71, 662-674.

Smith 0. H., and Gruber, H. Perception of Depth in Photographs.

Percept. 599 not. Skills, 1958, 8, 307-393.
Walker, H. H., and Lev, J. Statistical Inference, New York:

'Henry Holt and Co., 1953, 309-373.

AFFENDIX

Mean Distance in Inches for Each Target
For Right and Left Eye Dominant Observers

 

 

 

Target Right Left

2 X 2 Front 53.91 52.61
2 x 2 Back 59.36 57.6#
3 x 3 Front 82.07 83.11
3 X 3 3901‘ 33.57 37.35
b X it Front 106.1!8 112.71
I! X It Back 110.82 124.96
6 X 6 Front 151.50 169.93
6 X 6 Back 164.23 171.50
8 X 8 Front 189.70 212.29
8 X 8 Back 201.73 209.1!3

 

 

V. '. V
m_—~ - ‘ - - - .M . .
U n . '-~ ' -
L
\ .
. my —' I -‘ > - a -
.
‘ I
. .
Q ‘
, ..
‘ V
. .
'e l
a .
Q . '
.
‘
C
m . '-
.
e " '
z .
. .
‘ I.
‘ s ' "
e "1
, \

‘_--

-O-_ ‘

4 H
5‘ |
.
.O
‘\
‘ .
v
, .
,\
.Q
.
x
i.‘
.
e.
_ ‘0

\
l
s
4
—.-""

“C'l’fli’fi'fifl'fl @fl'flflflfl'fl'fiflt '1'“