4pm.:

P

1
2,
?
1

 

H, .Trfrr »»
,,

., . A.
....,..m~rw-- 1'3"
a ,. 7w” . -v

-pp
v

"w

.. I w I”,
,... , m”)... ,.,‘,.,..,.‘. :v ,, yv
C . -

 

. .qw. :1.
_ -»-‘;"‘: _.. fl—

" Ju-r ""
”yarn A?" "’2.
.. ,-.-—r

rrfiff: : ‘
rm— rrrrrv- r 'W'"";' ' ("in
”wry-.- «#54,, "

,r

.. c’I""| ”v—

‘ a: -.- --

”5" v‘ ~~-J—4~‘~,
-;3,Wr«

w'rrIr--:" "'
.2: wra'f:l_':::'

. r4 f m "“'
v l-t-v'r : r I r-
.. 77H rurpr-twrrr'r'vt r “ " I r r
lit...“ ‘ Frr‘wrrr" :2 WV“ Anir’:;r:~ r Fr“
1"er my”. ,. rr’r rr-n-rrvryrrv "’“‘"‘""""g
7“ ('1'. .17 ”r. rl rr—vrrr “~rrrn‘:::l'w *
/‘ .m-rI-p'nr'v r"""’,,-""
Maw-p vrrrn-v-r. yaw-V; "" ’" ,urfi‘tn
Prf-FJ- ,W ' -..,...~e-rrr-
,. rum-rvvr-I‘" ‘
*1, Vw' "re-rv'r'r'“
, "0.-....

wry-p VI'
{r "my-44v?» rrr'
n- l-r rw X55". nrv-I- "raw-r1 r r:'V‘"".':';-'
'5’ rhriu ' 941‘ 1,.“ ',,.-,,,.yr1r: rr-r:r—-V
r, ,,,,.~...r.r}- "W
, '. ,: .,.
'.’ 5:- m;
"a." '1'}. . yr
'rrrr VII,
,,. yr» w)" ' “"m
’r I!“ V: ”fly-3%";
’24-. m “7;":
fl. ,hzri v ,.',«>r ‘ '
:.,“"ru’.p-v rr' 1"” ””01
:1? u .5 ,,z.'-'.'>'.:":“"" ' "-r. .1. 2
p"?",; cm! ,.,.v rr
'1” v- ~“u:_r' , nua- u u
‘ ~ ”"',”,r'np¢pul'”f‘r"
.
.:

,, w.
w-AlrIr : f .h I
N, mm. .:"£'”’”""'"r'r' ’
w , .r.r.,,..mz¢¢1rm-
fry-i . fart—f“
3:31;,"

 

 

This is to certify that the
thesis entitled
THE DEVELOPMENT AND TESTING OF A SIMULATED

DYNAMIC DRIVING HAZARD PERCEPTION
TEST

presented by
Raymond Olaf Swensen

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Secondary Education
and Curriculum

{/C’M‘UWCO i read—

Major professor

 

 

 

Date May 19, 1976

 

0-7639

ABSTRACT

THE DEVELOPMENT AND TESTING OF A SIMULATED
DYNAMIC DRIVING HAZARD PERCEPTION
TEST

By

Raymond O. Swensen

The most important elements of the driving task
are those processes related to perception. Driver per—
ception is an all inclusive term, and the process, because
of its complexity, is difficult to determine and measure.
One facet of driver perception that can be measured is
the recognition of driving hazards.

The primary purpose of this thesis was to develop
an instrument to measure a phase of that process called
dynamic driver perception, primarily that of driving haz-
ard recognition. A secondary purpose was to determine how
drivers respond to the instrument.

To accomplish these purposes two tasks were under-
taken: (1) a motion picture film sequence instrument was
developed to be used as a test in evaluating the ability
of drivers to recognize driving hazards, and (2) a means
to evaluate the manner in which drivers responded to the

instrument was accomplished.

Raymond O. Swensen

The instrument consisted of fifteen, 16mm film
clips or sequences depicting various types of driving haz-
ards called the Simulated Dynamic Driving Hazard Percep-
tion Test (SDDHPT) which was developed from six driving
simulation films provided by Allstate Insurance Company.
Experts representative of traffic safety disciplines con-
firmed the hazards and established limits by which sub-
jects' response time was measured by means of a specially
developed counter attached to a motion picture projector.

The sample consisted of fifty-eight drivers from
the Lansing, Michigan area; twenty-nine were license
renewal applicants, who had no accidents or violations on
their driving records. The other twenty-nine were drivers
who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent
(chargeable) accidents within the past two years.

The results of the analysis of the data revealed:

1. When measured only by recognition of hazards
a t-test of significance indicated no significant
difference at the .05 level between the mean score of
the renewal and improvement groups.

2. When measured by recognition of the hazards
within the time set and validated by traffic safety experts»
a t-test of significance indicated no significant differ-
ence at the .05 level between the mean scores of the R and

I Groups.

Raymond O. Swensen

3- Findings from each sequence indicate that
except for sequence 2 there was no significant differ-
ence with respect to the film location at which the

two groups identified the hazard.

4. With respect to the identification of the haz-
ard within the experts' limit there was no significant
difference between groups in all sequences.

5. The majority of subjects in Groups R and I
were unable to identify the hazards confirmed by the
experts in six of twelve sequences.

6. Recognition of hazards within the experts'
limit for both subject groups was poor as defined by the
experts.

Recommendations for further research:

1. A study should be undertaken to determine if
persons identified as problem drivers differ in visual
perception capability from recognized safe commercial
drivers as measured by the SDDHPT.

2. Research should be undertaken in which a cor-
relation of the SDDHPT with another measure of visual per—
ceptual capabilities is made.

3. A replication of this investigation using
another visual perceptual measure should be done.

4. Research should be instituted in which a cor-

relation of the SDDHPT with the Michigan State University

Raymond O. Swensen

Driver Performance Measurement instrument should be eval-

uated.

5. A correlation of the SDDHPT with the Project
METER (Machine, Examination, Teaching, Evaluation and Re-
education) should be conducted.

6. To establish norms applicable to the general
population, large groups of matched subjects should be

tested by the SDDHPT.

THE DEVELOPMENT AND TESTING OF A SIMULATED
DYNAMIC DRIVING HAZARD PERCEPTION
TEST

By

Raymond Olaf Swensen

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Secondary Education and
Curriculum

1976

CDCopyright by
RAYMOND OLAF SWENSEN

1976

ACKNOWLEDGMENTS

This research would not have been possible except
for the help and cooperation of a number of people and
several organizations. Success in completion of this dis-
sertation as well as the degree for which it was a require-
ment was in large part due to the advice and assistance
of Dr. Robert Nolan, major advisor and guidance committee
chairman.

Acknowledgment is also due Dr. Dale Alam, Dr. James
Page and Dr. Donald Smith who served as members of the
guidance committee.

Gratitude is extended to Mr. Paul Denton for the
production of the figures; to Dr. Robert Ludeman, Dr.
Robert Moon, Mr. Richard Claflin, Mr. Martin Dolan, Mr.
Richard Hilderley, and Mr. Fred Vanosdall for their
special assistance and advice; Ralph Jackson and Allstate
Insurance Company for the use of the driving simulation
films; and to Mr. Allen Bard and the Bureau of Driver and
Vehicle Services, Michigan Department of State for making

available subjects for the test.

ii

Special thanks is also due to the administration
of Andrews University without whose encouragement and
financial support the work would never have been under—
taken.

Most of all the greatest thanks goes to my wife,
Shirley, and son, Reginald, whose untiring help and

encouragement brought this project to fulfillment.

iii

TABLE OF CONTENTS

ACKNOWLEDGMENTS
LIST OF TABLES.
LIST OF FIGURES

Chapter
I. THE PROBLEM.

Introduction

Purpose. .

Statement of Problem

Subordinate Problems

The Instrument . .

The Traffic Safety Experts

Nature of Variables. .

Sample

Basic Assumptions

Limitations. .

The Hypotheses .

Definition of Terms.
Driving Hazards.
Driver Hazard Perception
Search ...
Allstate Good Driver Simulator Films
Simulated Dynamic Driving Hazard

Perception Test. . . . .

Film Clips or Sequences.
Experts' Limit . . .
Experts' Frame Number.
Recognition Score.
Time Limit Score
R Group.
I Group

Organization of. Study.

iv

Page
ii
vii

ix

\DKDKOCD 0000\I\I\I\I\JO\O\O\U1U1-L\4-\UOH H

Chapter Page

II. LITERATURE REVIEW . . . . . . . . . . . . . . . 12
Summary . . . . . . . . . . . . . . . . . . . 43
III. DESIGN AND METHODOLOGY. . . . . . . . . . . . . 45
The Simulated Dynamic Driving Hazard Per-
ception Test. . . . 45
Simulated Dynamic Driving Hazard Perception
Test Development. . . . . . . . . . . . . . 45
The Sample. . . . . . . . 49
Test Equipment and Development Des-
cription. . . . . . . . . . 50
Description of Test Procedures. . . . . . . . 51
The Null Hypotheses . . . . . . . . . . . . . 53
Analysis of Data. . . . . . . . . . . . 54
Other Questions Considered. . . . . . . . . . 54
Summary . . . . . . . . . . . . . . . . . . . 55
IV. ANALYSIS OF DATA. . . . . . . . . . . . . . . . 57
Differences Between Group Means on Basis
of Recognition Only . . . . . . . 57
Difference Between Group Means on Basis
of Recognition and Time . . . . . . . . . . 58
Other Questions Considered . . . . . . . . . 59
Sequence 2 . . 60
Sequence 4 65
Sequence 5. . . . . . . . . . . . . . . . . . 70
Sequence 6. . . . . . . . . . . . . . . . . . 75
Sequence 7. . . . . . . . . . . . . . . . . . 79
Sequence 8 83
Sequence 9. . . . . . . . . . . . . . . . . . 87
Sequence 10 . . . . . . . . . . . . . . . . . 91
Sequence 11 . . . . . . . . . . . . . . . . . 95
Sequence 12 . . . . . . . . . . . . . . . . . 99
Sequence 13 . . . . . . . . . . . . . . . . . 103
Sequence 15 . . . . . . . . . . . . . . . . . 107
Summary . . . . . . . . . . . . . . . . . . . 111
V. SUMMARY, MAJOR FINDINGS, CONCLUSIONS, RECOM-
MENDATIONS, AND DISCUSSIONS . . . . . . . 112
Summary . . . . . . . . . . . . . . . . . 112
Major Findings . . . . . . . . . . . . . . . 114
Conclusions . . . . . . . . . . . . . . . . . 115
Recommendations . . . . . . . . . . . . . . . 116
Discussion. . . . . . . . . . . . . . . . . . 117

BIBLIOGRAPHY .

APPENDIX A .
APPENDIX B

vi

Page
121
137
141

Table

10.

LIST OF TABLES

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 2

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 4

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 5

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 6

Groups R and I driving hazard perception as
measured by recognition and experts‘ frame
number when applied to sequence 7

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 8

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 9

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 10.

Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 11.

Groups R and I driving hazard perception as

measured by recognition and experts' frame
number when applied to sequence 12.

vii

Page

63

68

73

77

81

85

89

93

97

101

Table

11. Groups R and I driving hazard perception as
measured by recognition and experts' frame
number when applied to sequence 13. . .

12. Groups R and I driving hazard perception as

measured by recognition and experts' frame
number when applied to sequence 15. .

viii

Page

105

109

10.
11.
12.

Hazard
Hazard
Hazard
Hazard
Hazard
Hazard
Hazard
Hazard
Hazard
Hazard
Hazard

Hazard

LIST OF FIGURES

Recognition
Recognition
Recognition
Recognition
Recognition
Recognition
Recognition
Recognition
Recognition
Recognition
Recognition

Recognition

of
of
of
of
of
of
of
of
of
of
of

Subjects
Subjects
Subjects
Subjects
Subjects
Subjects
Subjects
Subjects
Subjects
Subjects
Subjects

Subjects

ix

in
in
in
in
in
in
in
in
in
in
in

in

Sequence
Sequence
Sequence
Sequence
Sequence
Sequence
Sequence
Sequence
Sequence
Sequence
Sequence

Sequence

11.
12.
13.
15.

\OCX)\IO\U14>N

Page
64
69
74
78
82
86
90
94
98

102
106
110

CHAPTER I

THE PROBLEM

Introduction

 

”. . . a universally agreed-upon definition of.per-
ception simply does not exist, . . . a working definition
of perception can be considered as an awareness of our
environment."1

The literature related to perception tends to be
theoretical in nature because some theories account for
limited areas of perceptual phenomena while others have
been extended with the intent to integrate all of psy-
chology. Beyond this, additional theories have stimu-
lated new research with the hope of revealing new percep-
tual relationships.2

"In the field of perception, there are many exam-

ples 0f innate perceptual responses in lower forms."3

 

1Herschel W. Leibowitz, Visual Perception (New
York: The Macmillan Company, 19697, Preface.

2Daniel J. Weintraub and Edward L. Walker, Perce -
tion (Belmont, California: Brooks/Cole Publishing Company,

I966), p. 4.

3Leibowitz, op. cit., p. 10.

”As we ascend the phylogenetic scale, learning becomes

increasingly important, and instinctive behavior is pro-

. , 4
gre531ve1y less consp1cuous.”

There is no doubt that perception is learned for

” [p]erception would be of little use to the organism if
it were not modifiable by experience."5

Attention, or selective perception, is an
extremely important concept in the study of percep-
tion, because it determines what we are aware of at
the moment.

The appropriateness of our attention may at

times be rather important, if not critical, to

our adjustment. The student in class who is think-
ing about something other than the lecture topic
may as well not be in class. The driver whose
attention wanders from the road in front of him
may become involved in an accident.

The driver whose attitude or mood is anxious or
critical could very well incorrectly perceive a truly
dangerous driving hazard.7

According to James E. Aaron and Marland K.
Strasser in Driver and Traffic Safety Education, ”Patterns
of behavior must be the result of selecting correct stim-
uli from the environment and discarding those that are

. 8
1ncorrect.”

 

41bid., p. 16.
51bid., p. 27.
61bid., p. 28.
71bid., p. 96.

8James E. Aaron and Marland K. Strasser, Driver
and Traffic Safety Education (New York: The Macmillan
Company, 1966), p. 75.

 

In view of the fact that approximately 90 per
cent of human discriminations are based on what is
seen, it is especially significant that these in-
coming data be interpreted properly by our perceptual
processes.

Perception, by definition, relates to both seeing
and understanding. It is necessary, therefore, for one
to identify a multitude of stimuli in the driving view
and then proceed to interpret such stimuli. The most
important elements of the driving task are those processes
related to perception.10

Driver perception is an all inclusive term, and
the process, because of its complexity, is difficult to
determine and measure. However, one facet of driver per-

ception that can be measured is the perception of driving

hazards.

Purpose

The primary purpose of this thesis was to develop
an instrument to measure a phase of that process called
dynamic driver perception, primarily that of driving haz-
ard recognition. It is believed the instrument could also
be of value in improving the perceptual process of driving
hazard recognition of drivers. A secondary purpose was to

determine how drivers respond to the instrument.

 

91bid., p. 74.

lOIbid.

 

 

 

Statement of Problem

 

It might be said that driver hazard perception is
important to defensive driving. Drivers should have the
ability to recognize situations that are not seemingly
dangerous, but indeed do have potential danger. The quick-
ness with which drivers recognize a situation that can
develop into a very severe hazard if action is not taken
promptly is also important. The novice driver does not
recognize some situations where potential danger exists
in time to take corrective action. In fact, he may not
be aware of them at all. For instance, a potentially
dangerous situation may not result in a problem. In others
it can result in fatalities (e.g., a child playing near
the road may dart out).

The objectives of this study were to (1) develop a
motion picture film sequence instrument to be used as a
test in evaluating the ability of drivers to recognize
driving hazards, and (2) to evaluate the manner in which

drivers respond to the instrument.

Subordinate Problems
Related to the above objectives the following
specific questions were examined: (1) Will the instrument
discriminate among drivers? (2) How promptly will the
drivers recognize a hazardous situation? Or will they

recognize it at all?

 

The Instrument

An instrument was developed consisting of fifteen,
16mm film clips or sequences depicting various types of
driving hazards. The instrument includes film clips which
have driving hazards that are felt to be obvious, others
which have driving hazards that some drivers may recognize
and others will not, and finally those which show no driv—
ing hazards. All the sequences were selected from driver
education simulator training films produced by Allstate
Insurance Company. These film sequences, individually
numbered, were connected in a continuous widescreen film
of approximately 300 feet.

The film was shown to individuals after traffic
safety experts, as indicated below, had viewed, evaluated,
and validated both the hazard content of each clip as well

as confirming the grading system.

The Traffic Safety Experts

 

The traffic safety experts were people represent—
ing a variety of disciplines in traffic safety: a police
officer from the Michigan Department of State Police; a
specialist in motor vehicle administration from the High-
way Traffic Safety Center, Michigan State University; a
representative from the Bureau of Driver and Vehicle Ser—
vices, Michigan Department of State; a full—time driver

education teacher from the Lansing Public Schools; and a

 

’ IVE?» .u .129. A . .1 .. .I .u . . . a‘vnl «5.0. . . 7 .

 

driver education consultant from the Michigan Department

of Education.

Nature of Variables
The variables are of two types: (1) the recogni-
tion vs. non-recognition of hazards in the film clip, (2)

at what point in time the subject perceived the hazard.

Sample

The sample consisted of fifty-eight drivers,
twenty-nine of whom were license renewal applicants, often
referred to as the R or renewal group, who had no accidents
or violations on their records. The other twenty-nine,
often referred to as the I or improvement group, were
drivers who were required to appear at the Driver Improve-
ment Bureau for reexamination because of three negligent

(chargeable) accidents within the past two years.

Basic Assumptions
This thesis was based on the following assumptions:
Drivers being tested can respond to driving hazards they
see in the driving simulation film by pressing a stop
button on a hand held control. Driver perceptual capa-
bility as related to driving hazard recognition can effec-
tively be measured using the simulated dynamic driving

hazard perception test developed in this study.

 

Limitations

 

The findings and the conclusions of this study
are limited to the population tested. Because of a lack
of experience in viewing simulation films, some subjects
being tested may not have recognized hazards in the driv-
ing environment to the rear as depicted in mirrors shown

on the screen.

The Hypotheses

 

The hypotheses to be tested in this study were:

H1: When measured by recognition of the driving
hazards as validated by traffic safety experts, there will
be a significant difference at the .05 level between the
renewal and improvement groups.

H When measured by recognition of the driving

22
hazards within the limit set and validated by the traffic
safety experts, there will be a significant difference at

the .05 level between the renewal and improvement groups.

Definition of Terms

 

DrivingiHazards

 

Driving hazards are driving situations that are
dangerous or potentially dangerous as judged by the panel

of traffic safety experts employed in this study.

Driver Hazard Perception

 

Driver hazard perception is that process that a

driver is involved in when making decisions based upon the

 

 

identification of stimuli in the driving view and the in-
terpretation of that stimuli to recognize or identify and

respond to driving hazards.

Search

"An observable behavior in which the driver looks

systematically toward possible sources of traffic infor-

mation.”11

Allstate Good Driver Simula—
tor Films

 

Allstate Good Driver Simulator Films are 16mm
color, wide-screen motion pictures depicting a variety of
driving situations as seen from the driver's seat. These
films were written, directed, and produced by Allstate
Training Division to be used with the training unit pro-
duced by Link Division of General Precision, Inc.
Simulated Dynamic Driving
Hazard Perception Test

The simulated dynamic driving hazard perception
test consists of fifteen film clips or sequences each
numbered and connected in a continuous 300—foot, 16mm
silent wide—screen motion picture. Two of the fifteen

sequences are controls with no hazards and the first one

 

1]'T.W. Forbes and others, "Driver Performance
Measurement Research.” A Technical Report, Michigan State
University, Vol. I, February 1973, p. xiv.

 

 

was used as a sample, thus leaving twelve sequences to be

scored. This simulated dynamic driving hazard perception
test, described more fully in Chapter III, was developed
as a specific part of this thesis. It was given to the
sample populations.

From this point on the simulated dynamic driving

hazard perception test is referred to as SDDHPT.

Film Clips or Sequences

Film clips or sequences refer to a silent wide-
screen motion picture of short duration lasting from ten
to thirty seconds selected from the Allstate Good Driver
Simulator Films. From this point on the term film clip
or sequence will be used interchangeably. Fifteen of

these clips were used to make up the SDDHPT.

Experts' Limit

The experts' limit is the location in each film
sequence referred to by frame number, at or before which
the subject should have stopped the projector and cor-
rectly recognized the hazard. Because of mechanical lim-
itations the experts‘ limit takes into account a possible

variation of five frames in the projector's stopping point.

Experts' Frame Number
The experts' frame number is a term used hereafter

interchangeably with the experts' limit as defined above.

 

 

 

 

10

Recognition Score

 

The recognition score for the SDDHPT represents
the total number of sequences for which the subject recog-

nized the proper hazard and stopped the projector.

Time Limit Score
The time limit score for the SDDHPT is a subset
of the recognition score for which the subject stopped the

projector within the experts' limit.

R Group

The R Group consists of that portion of the sample
who were license renewal applicants having no accidents or

violations on their driving record.

I Group

The I Group was that portion of the sample who were
required to appear at the Driver Improvement Bureau for re-
examination because of three negligent (chargeable) acci-

dents within the past two years.

Organization of Study

 

The general overview of this study is to present
in Chapter II a review of the literature of visual percep-
tion with primary emphasis on visual perception in driving
as related to dynamic driver perception tests.

Chapter III contains a description of the inception

and development of the SDDHPT from the Allstate Good Driver

 

 

ll

Simulator Films, an account of how the test was used,
those who were tested, and a description of the statisti-
cal techniques applied to the test results. The results
of the analysis appear in Chapter IV along with the sub—
ject's comments and a look at the relationships within
the individual sequences.

Chapter V contains the summary, major findings,
conclusions, recommendations for further study, and a

discussion.

 

 

 

CHAPTER II

LITERATURE REVIEW

The literature related to driving hazard percep-
tion tests is limited; however, literature concerning per-
ception is profuse. A computer search of perception in
psychological abstracts reveals sixteen thousand entries.
A few early investigations in the field of visual percep-
tion have been cited, which in some way relate to the
development of the Second World War military pilot selec-
tion as well as aircraft identification programs. The
selected literature review will then proceed to works of
more recent researchers in perception as related to per-
formance in varied situations including sight, athletics,
driver performance, and driver perception. Finally, some
driving hazard perception tests will be included.

The ability of people to recognize or perceive and
then reproduce digital patterns or multiple digit numbers
has been of interest to researchers for a number of years.
According to Samuel Renshaw in the article "The Visual Per-
ception and Reproduction of Forms by Tachistoscopic

Methods" published in the Journal of Psychology, 1945,

12

l3

Kroh made the first recorded studies of the exposure times
necessary for students without special training or talent
to perceive and reproduce digit patterns of differing
lengths.

A precision tachistoscope was designed and built
at Ohio State University in the early thirties which could
produce exposures as short as .5 milliseconds with as
little as two percent error. With this instrument experi-
ments were performed in which people were trained to read
large digit numbers with shorter and shorter exposure.

Renshaw concluded that recognition training was
best obtained with high speed presentations and that per—
ceptual learning of wholes is better than verbal learning
of parts.2

Because of the great need for rapidly training
large numbers of pilots, the army air force early in the
Second World War (June 1941) established a Psychological
Research Agency ("The first psychological tests were given
to aviation cadets in October, 1941, at Maxwell Field

Alabama").3

 

lSamuel Renshaw, "The Visual Perception and Repro-
duction of Forms by Tachistoscopic Methods,” The Journal
of Psychology, 1945, pp. 217-232.

2

3Staff of the Psychological Branch, Office of the
Air Surgeon Headquarters Army Air Forces. "The Aviation
Psychology Program of the Army Air Forces, ngchological
Bulletin, 40 (1943), p. 760.

 

Ibid.

 

 

14

A battery of tests evolved for the purpose of
determining a person's aptitude and ability as a pilot,
bombardier, or navigator. By October 1943, the battery of
tests were grouped into four general categories:

1. Tests of intelligence, judgment, and pro-

ficiency.

2. Tests of alertness, observation, and speed of

perception.

3. Tests of personality, temperament, and inter-

ests.

4. Tests of visual-motor coordination.

From the four categories of tests three scores were deter—
mined, a pilot aptitude score, a bombardier aptitude score,
and a navigator aptitude score. Up to this point no test
involved motion pictures.

From Renshaw's experiments many variations evolved
in teaching aircraft identification to military pilots in
training. Some schools used short exposures such as 1/50
of a second while others used exposures up to one second.
Because of these variations Lester Luborsky performed an
experiment in which

Four equated groups, of 8 per-aviation (V-S)

students each, were taught aircraft recognition in
a standardized manner. The major experimental

 

4

5Lester Luborsky, "Aircraft Recognition: I. The
Relative Efficiency of Teaching Procedures,” Journal of
Applied Psychology, Vol. 29, No. 5 (October 1945), p. 386.

Ibid., p. 764.

 

 

 

l5

variable in each group was as follows: Group I,

l/50" exposure time; Group II, only three views of
each plane; Group III, 1" exposure time; and Group

IV, presentation of the entire syllabus in almost half
the usual time, followed by a review emphasizing con-
fused planes.

The following conclusions were obtained.

1. The use of 1/50” exposures as part of train-
ing in which an approximately equal number of
longer exposures are given has no ascertain-
able advantages over 1” exposures similarly
given with longer exposures.

2. Restriction of teaching materials to only
three views of each plane results in learning
which generalizes poorly to views of planes
other than those taught and is, in this sense,
inefficient.

3. Rapid teaching followed by review of confused
planes is probably the most efficient of the
procedures tested.

During the Second World War it was deemed import—
ant to use motion picture tests which were considered more
closely resembling the real world.

The effort was made to get away from the purely aca-
demic type of examination which puts a premium on
verbal memory and to test the performance of the
student in a situation having the sequence, the
tempo, and the continuous change of the real situa-
tion with which he will have to deal.

The advantages of motion picture tests especially
for group testing, as pointed out by J. J. Gibson and sum-
marized by Benjamin Fruchter and William W. Mahan, are:
(l) movement, where both objects can move as well as the

camera, (2) temporal sequence, where the projector serves

as an exposure device in presenting material to a group

 

6Ibid., p. 397.

7J. J. Gibson (ed.), Motion Picture Testing and
Research Re ort, No. 7, U.S. Army Air Force (Washington,
D.C.: Government Printing Office, 1947), p. 100.

 

l6

simultaneously, (3) pacing, in other words, items can

be presented for just the length of time desired and the
exposure can be lengthened or shortened as desired by

the test maker, and (4) realism, compared to verbal des-
criptions or still pictures, motion pictures make things
more real. ”These advantages are especially important for

tests of perception since much greater control of the per—

ceptual process is possible."8

Gibson further emphasizes the use of motion pic-
ture tests by stating

Human behavior, and the capacities latent in it,
also involves motion, order, tempo, and the experi-
ence of reality. It is reasonable to suppose, there-
fore, that the motion picture makes available to the
test designer not only a special method of measuring
known factors of human ability but also gives him
access to new and unnamed functions not accessible
to conventional methods of test construction.

The functions are as follows: discrimination of
visual motion and locomotion; perception of space and
distance, particularly during flight; maintaining
orientation during locomotion; ability to learn a
procedure; ability to react to a changing situation;
ability to perform during emotional stress.

Early in the war the army had set up a Perceptual
Research Unit to construct aptitude tests for new aircrew
candidates. In addition, research in the general field of
perception was conducted. Part of this research included

motion pictures. In order to improve and intensify this

 

8Benjamin Fruchter and William W. Mahan, "Some
Perceptual Factors Measured by Motion Picture Tests,”
Journal of Educational Psychology, XXXXIII (1952), pp. 430—
435.

9Gibson, op. cit., p. 20.

 

 

 

 

17

motion picture development of perceptual testing a Psy-

chological Test Film Unit was activated with its primary

function of developing motion picture perceptual tests.

A number of motion picture tests were developed.

One of the important hypotheses employed for

selecting the motion picture medium for test construction

was that

It was assumed at the beginning that tests re-
quiring discriminations of time and events—in-sequence
were necessary to supplement the motionless type of
visual discriminations required by paper and pencil
tests already in use. Plans were made, therefore,
for the development of film tests which make possible
a wide range of judgments involving motion, and hence
are closer to the realities of aircrew performance.

Instead of describing all the perceptual tests

constructed a description of one representative motion

picture follows:

One of the first test films planned was Speed
Estimation Test III, Estimation of Relative Velocities,
one of three tests requiring complex judgments of
speed and motion. The intention was to construct a
motion picture test requiring a complex judgment of
the relations between two velocities, measured by
estimating the imagined point at which the faster of
two moving spots (planes) would overtake the slower.
The paths of the two motions were parallel so that
the judgment could not be reduced to a merely geo-
metrical one.

Two animated planes are seen on the screen against
a skyline background, one overtaking the other. Be-
fore the overtaking point is reached, both planes dis—
appear behind a cloud and a five point scale appears
superimposed on the cloud. The testee must project
the two velocities and judge at which of the five

 

lOStaff, Psychological Test Film Unit, "History,

Organization, and Research Activities, Psychological Test
Film Unit, Army Air Forces,” Psychological Bulletin,
XXXXI (1944), p. 459.

 

 

18

points the two planes would coincide. The test con-
sists of 50 trials, ten at each of the five points,
in a random order.11

During this era a total of eleven tests using the
motion picture medium were developed:

Three, including the one described, are tests re-
quiring estimation of visual velocities of various
types. One requires reproduction, on a standard
answer sheet, of briefly exposed visual patterns and
is intended as a test for speed of perception. Two,
entitled "Flexibility of Attention” and "Integration
of Attention" respectively, are efforts to measure
the ability to take account of different events occur-
ring in the perceptual field at the same time without
becoming confused. Two more tests have to do with
the complex ability to form unitary perceptual exper-
iences from a series of successive partial impressions.
Two require discriminations of, respectively, just
noticeable movement and just noticeable direction of
movement (drift), in the field of a bombsight sche-
matically presented on the screen. One is a test of
perceptual judgment of a sort which is required in
aircraft landings, consisting of shots of a runway
photographed at varying moments during an approach
glide, with the requirement that the trainee estimate
at which of five lettered points the present angle of
glide intersects the runway.

A considerable number of tests measuring a variety
of perceptual functions were given to aviation trainees.
It was found that a wide spread of individual differences
occurred as to perceptual abilities of individuals existed
even in the homogenous population available.13

The literature is limited concerning motion pic~

tures in testing perceptual abilities of pilots or drivers.

 

lllbid., p. 463.

121616., pp. 463-464.

l3lbid., p. 465.

 

19

One such test was developed and conducted by Robert Spicer
in Honolulu, Hawaii, as a part of a study of the human
factors approach in traffic accidents. In his research
Spicer investigated the human factors of attitude, frus-
tration response, problem solving, and visual perception.
In the process of investigating visual perception he de-
veloped ”an eleven-scene 16mm silent motion picture film
depicting typical city and highway traffic conditions.”14
Visual perception was the only one of the variables tested
which seemed to show some consistent relationship to driv-
ing and traffic mishaps. Evidently there are visual cues
related to vehicle operation that are peculiar to driving
and different from general visual alertness.

Further emphasis is placed on driver perceptual

skill by the following statement:

The factor of perception has been advanced as an
inherent element of the driving situation, and the
degree of visual alertness or perceptual skill an in-
dividual possesses should affect to a considerable
extent, his success in operating a vehicle without
mishap. To an even greater extent, his ability to
utilize the information provided by his perceptivity
(effective problem-solving following perception of
stimuli) should be a determinant of accident status.
Conversely, a motorist may still find himself re-
peatedly involved in accidents, despite his effi—

ciency at problem-solving, if he has never learned
to look for critical cues on the highway.

 

14Robert A. Spicer, ”Human Factors in Traffic Ac—
cidents," Research Grant AC-55, U.S. Public Health Service,
Department of Health, Education and Welfare (Washington,
D.C.: U.S. Government Printing Office, 1963), p. 14.

lSIbid., p. 11.

 

20

Eugene D. Carney describes the many facets of
perception from the standpoint of the driver education
teacher. One of his important ideas is that

Greater perceptual efficiency is achieved when the
driver realizes that the highway circumstances or
developments that require significant physical response
result very frequently from a change in the traffic
scene. For example, stationary vehicles suddenly
begin to move, moving vehicles suddenly stop or change
direction, pedestrians and animals suddenly stop,
start or change direction, or vehicles, pedestrians
or animals suddenly appear on the highway from an area
where the driver's vision was restricted. Consequently,
it is important for the driver to realize that he must
continually recycle his perceptual stimuli so that he
is immediately aware of any need to respond.

Visual identification activates the perceptual
skills in all but a few instances.

Admitting that the task of driving a car is not a
simple one in modern traffic, T. W. Forbes as far back as
1959 made the following observations concerning perceptions,
judgments, and responses:

In analyzing the driver's task, we find that per-
ception of rapidly changing situations, judgments based
on these perceptions plus background knowledge, and
responses adequate for each situation are the essentials.
The judgments made by the driver are based upon in—
formation about the highway situation and about pos-
sible hazards which may be so well learned that it is
automatically rather than consciously used. Such in-
formation is nonetheless very vital.

I should like to suggest here that two kinds of
specific, factual information affect very importantly
(I) the difficulty of the driver‘s task and (2) the
accuracy and effectiveness with which he responds in
driving modern highways. These two kinds of informa-
tion are: (1) what the driver does and doesn't know
about the best procedures for operating the vehicle
on modern highways and (2) what the highway designer

 

l6Eugene D. Carney, "Another Look at Perception,"
Journal of Traffic Safety Education, XIX, 2 (1972), p. 32.

 

21

and traffic engineer knows (and applies) about human
limitations and operator abilities. Even though tech—
nical engineering personnel may not be able to delve
into human factor research, specialists in this area
can obtain and supply information to them if given an
opportunity.17

Ross A. McFarland indicates in ”Human Factors in

Highway Transport Safety” that a number of tests and

studies have been conducted concerning the selection of

bus and truck drivers.

Parameters for a successful psychological testing

program were outlined as follows:

(1

v

(1) determine adequate criteria of the job require—
ments in successful driving, (2) devise specific

tests for professional drivers and do not simply use
measures which have been developed for other purposes,
(3) carry out adequate follow-up studies through a
period of years so that new items can be added and
poor ones discarded, and (4) select a large group of
drivers to study. (5) The group should be relatively
stable, homogeneous in experience background, and uni-
form in exposure to risks. Also each driver should
have a reasonably prolonged period of exposure to his
job so that individual differences will be revealed. 13

The results of a study to evaluate the psychologi-
characteristics of successful truck drivers revealed
generally that

drivers tended to show normal variability in the
following traits and interests: (a) appreciative, (b)
quiet, (c) submissive, (d) cold, (e) scientific, (f)
persuasive, (g) literary.

(2) Group characteristics indicated by the test
items were: (a) higher mental ability than average,

 

17T. W. Forbes, "Human Factors in Highway Safety,"
8.

. Traffic Safety, IV (March 1960),

18Ross A. McFarland and Alfred L. Moseley, ”Human

Factors in Highway Transport Safety,” (Boston, Massachu-
setts: Harvard School of Public Health, 1954), p. 30.

 

22

(b) composure, (c) optimism, (d) sympathetic tendency,
(e) objectivity, (f) impulsiveness, (g) appreciative—
ness, (h) quietness, (i) submissiveness, (j) coldness,
(k) somewhat high interests (though in average range)
in respect to l) mechanical, 2) artistic, 3) musical,
4) social service, 5) clerical areas; 1) low
interests in ten computational, 2) scientific, 3)
persuasive and 4) literary areas.

(3) The scores describe a 'person' who can work
alone within a limited working space, who can be
quick to respond in emergency situations, and whose
interests are so lacking in extremes as to foster 9
acceptance of the limiting aspects of the occupation.

A car operator's mood can obviously affect his per-
ception and ability to control a car. Norman W. Heimstra,
Vernon S. Ellingstad, and Arlan R. DeKock performed an
experiment in which subjects were first given a mood check
list and then tested in a simulated driving task. It was
found that ”$3 scoring high on factors of aggression,
anxiety, and fatigue performed more poorly on various tasks
than Ss who scored lower on these factors.”20

In a synopsis paper presented at the S.A.E. Golden
Anniversary Transportation Meeting in St. Louis, Missouri,
on November 2, 1955, McFarland in discussing analysis of
driver activities and critical incidents states on page
733:

When the task of driving itself is employed as

the logical, starting point, a rational analysis in—

dicates that anticipation, foresight, and continued
alertness on the part of each driver are required to

 

lgIbid., p. 37.
20Norman W. Heimstra and others, ”Effects of Oper-

ator Mood on Performance in a Simulated Driving Task,”

Perceptual and Motor Skills, XXV (December 1967), p. 729.

 

23

avoid accidents. Thus, any feature which may dis-
tract attention, interfere with perception, lengthen
reaction times, or hinder the completion of required
responses may directly compromise safety. Other fea-
tures, in themselves less critical, may however result
in reduced efficiency if they force the driver to oper-
ate near, or beyond, the limits of his normal capa-
cities. Such features may also accumulate in their
effects, adding to fatigue.

In a paper written by Lawrence E. Schlesinger and
Miriam A. Safren which was read to the Annual Meeting of
Highway Research Board in January 1964, the authors
attempted to develop a unified comprehensive model of
the driving task. According to the authors

In the model, the major tasks for the driver are
the perceptual organization from moment to moment of
a field of safe travel (a region in which the car can
move unimpeded): a minimum stopping zone (the smallest
region through which the car must move to come to a
full stop), and a comparison of these two fields. The
driver's organization of these two fields, or the field-
zone ratio is a control stimulus guiding the control
actions to the vehicle. That is, the driver varies
the speed and direction of movement of the vehicle to
maintain a safe field-zone ratio; that is, one in which
the field is greater than the zone.

 

Visual perception is then the underlying principle
in safe driving of which driving hazard perception is a
part.

H. Laurence Ross studied forty-three ordinary traf—
fic accidents intensively. He concluded that the most com-

mon circumstance was that the driver was not aware that he

 

21Lawrence E. Schlesinger and Miriam A. Safren,
”Perceptual Analysis of the Driving Task" (paper read at
the Annual Meeting of Highway Research Board, January
1964), p. 11.

 

24

was on a collision course. Two reasons for his unaware-
ness were delayed perception and erroneous prediction.

Delayed perception occurred because of impediments
to vision and inattentiveness. In only one accident
was the impediment to vision due to eye defects; in
seven it related to vehicle conditions such as ob-
scured windshields and passenger interference; in 21
cases it arose from view obstructions outside the
vehicle, the most common being roadside structures and
the next most common, parked cars. Inattentiveness
contributed to delayed perception in almost the same
number of cases as impediments to vision. It contri-
buted in the form of inadequate scanning habits, pre-
occupation, distraction within the vehicle, distrac-
tion outside vehicle, and inattentiveness to rear, the
latter being the most common. :

Erroneous predictions resulted when drivers ex- J
pected other vehicles to continue moving as they had
been, failed to receive a signal from the other driver
of his intent to turn or slow, depended on others to
observe signs and signals, or did not expect the unex-
pected from other drivers in numerous traffic situa—
tions. The reasons for failure of expectations seemed
to be insufficient cues as to behavior of the other
driver, action on his part that may suggest any of
several actions, and actions of his that differ from
that prescribed by law or based on custom.

Nick Rackoff investigated the effects of aging on
driver performance. He found among other things that per-
ceptual skill goes down with age. However, he states:

What this study does demonstrate is the large
variation in performance capabilities from one aged
subject to the next. Some subjects could hardly be
distinguished from young while others showed large
deviations. . . . Chronological age alone is a very
poor indicator of an individual's capability to drive}23

 

22H. Laurence Ross, "Ignorance of Collision Course
as a Factor in Traffic Accidents.” Traffic Institute,
Northwestern University, Evanston, Illinois, 1960, pp. 23-
24.

23

Nick Joseph Rackoff, "An Investigation of Age-
Related Changes in Driver's Visual Search Patterns and

25

Raymond Reilly and his associates attempted to
determine the manner in which certain items of perceptual
information are used by the driver in beginning and guid-
ing the braking response in the process of bringing the
vehicle to a safe stop behind a lead vehicle under night
conditions.

Measures of the distance from the display at
which deceleration began and the magnitude of the
braking response were then examined to discover
whether these responses were systematically related

to variations in the visual cues under study.

The three visual clues studied were taillight
area, brightness, and angular velocity.

It was concluded that a driver uses angle velocity
and vehicle velocity information to make the decision to
stop. Also, a vehicle with bright wide-spread taillights
produce the best braking response.

The complexity of the perceptual processes is
emphasized by the following comments:

. There are, to be sure, many aspects of driv-
ing behavior which appear to be considerably more com-
plex than initially envisioned and which do not con-
form to earlier predictions. In all instances, how-
ever, a very definite regularity is noted in the data.
The decision to decelerate and to exert a certain
magnitude of braking force seems to be a carefully
ordered response by the driver to the visual world

 

Driving Performance and The Relation of Tests of Basic
Functional Capacities" (Doctor's dissertation, Ohio State

University, 1974), p. 204.

24Raymond E. Reilly and others, "The Translation
of Visual Information into Vehicular Control Actions,"
Report Biotechnology, Incorporated, Arlington, Virginia,

October 1965, p. 5.

 

 

 

26

around him and to his expectations as to how this
world should behave.

In general, these results mediate against the
treatment of human responses in mechanistic terms.
Simple functional relationships do not exist between
human behavior in a given situation and isolated per-
ceptual variables which might be present. The human
operates within a 'framework.‘ In the present in—
stance, this framework is influenced by such aspects
as his insight into the purposes of the experiment,
his knowledge of the speed of his vehicle, and his
awareness of the consequences of a mistake while
traveling at a given speed.

Richard Lucas investigated a method to improve the
accuracy of drivers' last moment passing judgments. A
motion picture test was developed in which subjects would
indicate the last possible time in the film that they could
pass a car. One group of subjects were given the motion
picture training with feedback as to their accuracy while
the other group was given no feedback. Both groups took a
behind-the-wheel pre and post car passing test. The results
indicated the feedback group improved while the no feed—
back group showed no improvement. The feedback group did
"improve in the ability to make last minute passing judg-

ments"26

while the no feedback group did not improve.
Neither group performed superiorly on the final test. This
is another attempt at trying to check the perception abil-

ity of drivers on a specific task.

 

251bid., p. 38.

26Richard Lawrence Lucas, ”Development and Eval-
uation of a Part Task Film Simulation Technique for Train-
ing Drivers on Critical Passing Skill” (Doctor's disser-
tation, University of South Dakota, 1970), p. 73.

 
 

 

 

27

Albert Burg studied driver visual acuity both
static and dynamic and compared the findings with driving
records. His conclusions were that a relationship between
static visual acuity and the driving record could not be
supported. The study seemed to indicate a positive
relationship between dynamic visual acuity and driving
record; however final confirmation or rejection must await
a repeat using a greater sample. He states that "DVA
(dynamic visual acuity) performance is potentially a more
valid predictor of driving performance than is static
acuity score.”27

No studies were found which indicated that hazard
recognition would be more valid if conducted under
dynamic conditions.

From Burg's work it would seem that the use of the
SDDHPT, which is a dynamic test, would be a reasonable
approach. Only further research will give more definitive
data.

R. L. Newsome conducted a study of peripheral
vision in which two experiments were performed. In the
first experiment square blocks were placed in the peri—

pheral vision of subjects so as to look the same size as

 

27Albert Burg, "An Investigation of Some Relation-
ships Between Dynamic Visual Acuity, Static Visual Acuity
and Driving Record" (Doctor‘s disseration, University of
California, Los Angeles, 1964), p. 95.

 

 

28

a block straight ahead. The greater the angle of peri-
pheral vision the closer the block had to be moved to the
subject.

In the second experiment the subject stood facing
a man at a definite distance on a playing field. The sub-
ject was asked to stop a walking man in the subject's
peripheral view when both men appeared to be the same
size.

From the above experiments it was concluded that
objects from a side road may appear to be much further
away than the real distance. The perception of peripheral
objects may or may not be a hazard to drivers.2

An interesting conclusion from a peripheral visual
field study by A. T. Slater—Hammel was that ”an athlete's
reaction time to stimuli directly perceived will appar-
ently provide an excellent relative index of his reaction
speed to stimuli along the peripheral visual field."29

Albert Burg in 1964 made an extensive study of the
literature regarding the role that vision plays in driving
and found that there was a discouraging lack of any con-
clusive findings. This work cited in 1967 indicates that

the situation had not changed appreciably. He states:

 

28L. R. Newsome, "A Perceptual Factor That Could
Contribute to Road Accidents," Road Research Laboratory
Ministry of Transport, RRL Report LR 135, 1967.

29A. T. Slater-Hammel, "Reaction Time to Light
Stimuli in the Peripheral Visual Field," Research Quarterly,
XXVI (March 1955), pp. 82—87.

 

 

 

29

To summarize our current state of knowledge, it
is obvious that at the present time there is no widely
recognized evidence that vision is related to driving.
Everyone will readily admit that there must be a re-
lationship, but as yet there is no way of translating
this 'feeling' into valid and useful practice, such
as the development of practical vision-test standards
for driver-screening agencies.

Burg gives some interesting factors that he thinks
may be reasons why attempts to relate vision and driving

have been inconclusive.

l. Accidents are rare and complex events, usually
involving a number of factors. Vision is only one of
these factors, and while it is undoubtedly important,
vision's contribution may be small enough to make it
difficult to show a significant relationship.

2. 'Driving Ability,’ 'Driving Performance, and
'Driving Record' are three separate concepts. 'Abil-
ity' is the one we would like to be able to measure,
but cannot with any degree of confidence. 'Perfor-
mance' is an approximation of 'Ability,' but cost,
time, and legal factors make it impossible to obtain
any truly representative sample of driving performance
for the large sample of drivers essential to research
of this type. This leaves us with 'Driving Record'
which we can ascertain, but without any assurance that
we have obfained an accurate or complete estimate of
the driver's on—the-road performance, over the speci-
fied period of time, because of the lack of consis-
tency and completeness in reporting accidents and
traffic citations.

Vision is very likely an important factor in both
driving ability and driving performance. Driving
record, however, is influenced strongly by the element
of chance or luck and thus is, at best, an uncertain
indicator of true driving ability. It seems that we
are in a 'heads you win, tails I lose' sort of situa-
tion since we do not, as yet, have the means to mea-
sure true driving ability, and thus are forced to use
driving record (which is two steps removed from driv-
ing ability) as the criterion. At present we have no
alternative.

 

30Albert Burg, ”Some Preliminary Findings Concern-
ing the Relation Between Vision and Driving Performance,"
Journal of the American Optometric Association, XXXVIII, 5
(May 1967), p. 372.

 

30

3. The vision tests used may not be valid. That
is, they may not be related to the visual functiOns
used in driving. These functions, needless to say,
have yet to be established in any conclusive fashion.

4. The reliability of the vision test and/or of
the measure of driving ability used (e.g., driving
record) may be low. It is well known, for example,
that the correlation between driving records in adja-
cent periods of time is low.

5. What the person is capable of seeing and the
use he makes of this capability in driving may differ
535stantially. In other words, there may be consider-
able disparity between an individual's physiological
vision and his functional vision. (A complicating
factor, here, is that the individual may, to a certain
extent, be able to compensate for his visual short-
comings, thus making unsafe vision in driving).

6. Using drivers as our subjects, as we must, we
necessarily are dealing with a restricted range of
visual capabilities.3l

 

At the time Burg stated that there was no means of
measuring true driving ability, the Michigan State Univer-
sity Driver Performance Measurement Research project had
not yet been undertaken. However, that research has been
completed and the first two paragraphs of the preface best
describe what this research intended to accomplish in
terms of a valid and reliable measure of driving perfor-

mance .

This is a two volume final report of research
carried on over a period of more than two years, to
develop a reliable and valid method of measuring safe
and good driving applicable for small group research
to improve driver education methods. Problems caused
by the many unknown and variable factors in accident
records had been shown by state of the art studies,
from which the need for an immediate, usable method
of measuring safe and good driving had become clear.

None of the previous research, however, had shown
how much a reliable measurement of safe and good

 

3lIbid., pp. 372—373.

 

 

 

 

31

driver behavior could be accomplished. In this proj-
ect, therefore, the research team had to first con-
sider fundamentals; then develop the method, refine
it, try it out, evaluate it, train others to use it,
and then evaluate their results.32
Other disciplines have developed visual perception
tests of various sorts. James Wise developed and pre-
tested a visual perception test using slides of different
shapes and colors projected on a screen by a tachistoscope.
Intentions of the test were to serve as a screen-
ing device in determining preschool children's visual per-
ceptual capabilities. It did differentiate between sub-
jects who were performing at teacher expectations and sub-
jects who were not.33
Harriet Williams measured perceptual ability of
athletes by projecting a tennis ball with a tennis ball-
boy machine. Subjects were to visually judge the flight
path, then immediately move to a point of interception.
A canvas, suspended four to five feet above the head of
the subjects, prevented actual physical contact with the
ball. Movement time, reaction time, and measures of spa-
tial accuracy were recorded.

Three main purposes of the study were:

(i) to assess the effects of systematic variation
in velocity and direction of ball flight upon visuo—

 

32

33James Elton Wise, "Development and Pretest of a
Visual Perceptual Screening Test for Use in Preschool Edu-
cation” (Doctor's dissertation, University of South Caro-
lina, 1973).

Forbes and others, p. i.

 

. .51...
. In .. .
_..._.

32

perceptual judgments made about moving objects in space;
(ii) to determine whether or not highly-skilled and
poorly-skilled performers differ in their ability to
visually judge the flight of a moving object in three—
dimensional space, and (iii) to assess the effect of
age or maturity level of the individual upon the ipeed
and accuracy of such visuo-perceptual judgments.3

The results are summarized as follows:

(1) The speed and accuracy with which the flight
of a moving object was judged was, to a large extent,
dependent upon the specific set of visual cues involved,
that is, upon the particular speed, horizontal and/or
vertical direction in which the object was moving at
the time it was being judged. This of course suggests
that if we are to enhance the degree of success ex-
perienced by the individual in learning and/or perform-
ing certain gross motor skills, we need to begin to
identify more specifically the kinds of visual cues
involved in the performance of motor skills and to eval-
uate them in terms of the kinds of demands which they
place upon the sensori-perceptual apparatus of the in-
dividual.

(2) Individuals classified as highly-skilled were
significantly superior to individuals classified as
poorly-skilled in visually judging the flight of a
moving object in space. Such findings tend to support
the notion that the highly-skilled performer may, in
fact, possess a sensori—perceptual mechanism that is
superior to that of the unskilled performer. If such
differences in the visuo—perceptual capacities of the
highly-skilled and the poorly-skilled sports performer
do exist, it is important that we begin to establish
whether or not such differences are innate ones or if
the 'potentially' unskilled individual can be trained
to use his visuo-perceptual apparatus more effect1ver
through properly planned and appropriately timed
perceptual-motor experiences.

(3) Age, as represented by a sample of junior high,
high school and college age males, had little or no
effect upon the speed and accuracy with which the in-
dividual judged the flight of a moving object in three-
dimensional space.

 

34Harriet G. Williams, ”The Effects of Systematic
Variation of Speed and Direction of Object Flight and of
Skill and Age Classifications upon Visuo-perceptual Judg-
ments of Moving Objects in Three-dimensional Space" (Toledo
University, Ohio), report no. BR—6-8102, January 1968, p. I.

351bid., p. II.

 

33

Football players were trained by C. Frazier Damron
in defensive formations using two and three dimensional
slides exposed tachistoscopically. The group trained with
two dimension slides performed higher in correct responses
than the three dimension trained group. He concluded that
the entire training method had great potential for teach-
ing football defense recognition in a short time period
and in a classroom environment.36

Ben R. Londeree divided football players into two
groups and trained them for play recognition from the
defensive end position. One group was trained using motion
pictures while the other group was trained with flash cards.
When tested in a live situation for the quickness of recog-
nition, the motion picture—trained group recognized the
football plays in significantly less time than the flash
card-trained group. Here again the process involved per—
ception of different plays where motion of the players
obviously was of great value in early recognition.37

Charles Thiffault, in a study designed to evaluate

how tachistoscopic training effects the perceptual ability

of ice hockey players, discovered the results to support

 

36C. Frazier Damron, ”Two— and Three-Dimensional
Slide Images Used With Tachistoscopic Training Techniques
in Instructing High School Football Players in Defenses,"
Research Quarterly, XXVI (1955), pp. 36-43.

37Ben R. Londeree, Jr., "Effects of Training with
Motion Pictures Versus Flash Cards Upon Football Play
Recognition," Research Quarterly, XXXVIII (May 1967), pp.
202-207.

 

   

 

34

his working hypothesis that tachistoscopic training can
improve visual perception in ice hockey players. The
tachistoscopic training did improve the players' ability
to appraise and react to tactical situations in ice
hockey.38

A number of studies in driver hazard perception

have been undertaken. Many have included 35mm slides of
driving hazards rather than motion pictures. The test
most commonly used by researchers at Illinois State Uni-
versity has been The Perception of Traffic Hazards Test
(PTHT) which was developed at that institution.

Francis Kenel says,

Safe drivers must depend upon the development of
accurate judgments and correct decisions for determin—
ing their mechanical control over the vehicle. These
judgments and decisions are, in turn, based primarily
on the driver's perception of the traffic scene.39

Kenard McPherson using the PTHT endeavored to find

how IQ and the perception of traffic hazards were related.
Three groups were given instruction with traffic simulator
films selected from Aetna and Allstate. As a result of
his work he concludes:

The findings of this study suggest that training in

driving simulators using programmed films will improve
an individual's ability to perceive traffic hazards.

 

38Charles Thiffault, "Tachistoscopic Training and
Its Effects Upon Visual Perceptual Speed of Ice-Hockey
Players." (Doctor's dissertation, University of Southern
California, 1974.)

39Francis C. Kenel, "Employment of Teaching Assis-
tants in Driver Education," Safety, April 1970, p. 14.

 

I: ffl‘lfitrialflnflflmuu luv. "Ii

 

35
These findings appear to be relatively constant with
each of the three classifications studied.4

Gerald M. Streeter conducted special classroom
visual training with instruments similar to the tachisto-
scope for an experimental group of driver education stu-
dents prior to the laboratory instruction. He used the
PTHT in a test-retest method to measure perceptual growth.
It was concluded that

Visual perceptual ability as indicated by the recog-
nition of potentially hazardous traffic events in a
filmed situation presented in a classroom environment
can be improved without employing sophisticated costly
apparatus.

Allen Robinson administered the PTHT in a test-
retest method to two groups of high school driver educa-
tion students. The control group received conventional
simulator instruction while the experimental group received
the same instruction in a classroom setting. As a result
of this research he states:

The findings of this study indicate that visual
perception of traffic hazards and events can be devel-
oped to comparable levels through the use of a
conventional traffic simulator laboratory or by use
of the classroom simulation method when utilizing

programmed instructional films and active teacher
instruction.

 

 

 

40Kenard McPherson, "Perception of Traffic Haz-
ards: A Comparative Study (Master's thesis, Illinois State
University, 1966), pp. 40-41.

41Gerald M. Streeter, "A Classroom Visual Percep-
tion Program for Beginning Motorists," (Master's thesis,
Illinois State University, 1968), p. 30.

42Allen Robinson, "The Influence of Programmed In-
structional Films on Perception of Traffic Hazards" (Mas-
ter's thesis, Illinois State University, 1968), p. 44.

 

 

36

Ralph 0. Johnson trained culturally and econom-
ically deprived driver education students using visual
perceptual training. This consisted of 35mm slides of
digits and traffic signs as well as actual traffic events
flashed on the screen for short periods of time to develop
fast recognition. The PTHT was administered as a pretest-
posttest to measure perceptual ability. According to
Johnson,

This visual perception program using a 'static'
training medium proved positive and appropriate for
disadvantaged youth; therefore this instructional
method may be employed to enhance the visual percep-
tual capabilities of the deprived student.

Ronald L. Thomas divided a driver education class
in half giving a traditional course to the entire group.
However, with one half the class special training sessions
were conducted using 35mm slides containing digits, traffic
signs and traffic situations. The class was measured
before and after treatment with the PTHT as well as the
Streeter Numbers and Signs Test. Thomas concluded that

The results of The Perception of Traffic Hazards
Test indicate that visual stimuli recognition level
of a student can be significantly raised with a class-
room visual training program of this type.

The inferences derived from the test results ob—

tained from the Streeter Numbers and Signs Test indi-
cate that these students did not significantly improve

 

43Ralph 0. Johnson, ”A 'Static' Visual Percep-
tual Training Program for Minority Ethnic Groups" (Master's
thesis, Illinois State University, 1970), p. 31.

37

in their ability to identify numbers and signs as
presented in this test instrument.44

Donald Louis LaFond gave a control group of driver
education students a traditional driver education course
of thirty hours class instruction and six hours behind-
the-wheel instruction. The experimental group received
the same training with the addition of dynamic visual
training films integrated into the classroom phase.
LaFond's conclusions were thus:

1. The ability of beginning drivers to identify
visual stimuli as measured by The Perception of Traffic
Hazards Test is increased through the use of dynamic
training films.

2. Traditional classroom and laboratory experi-
ences also appear to increase the ability to identify
visual stimuli as measured by The Perception of Traffic
Hazards Test but to a lesser degree.

3. There is a significant relationship between
dynamic visual training of beginning drivers and the
drivers' manipulative skills of steering and braking
as measured by the Drivometer.

4. There is a significant positive relationship
between dynamic visual training and gross eye movements
as recorded by a video tape recorder.45

 

 

Robert S. Lazarewicz investigated the relationship
between behavior characteristics and visual perception of

young drivers using the Mann Inventogy and the PTHT. He

 

found no significant difference between behavioral

 

44Ronald L. Thomas, "A Comparison of Two Types of
Classroom Presentation of Visual Stimuli Recognition in
Driver Education" (Master's thesis, Illinois State Univer-
sity, 1970), pp. 26-27.

45Donald Louis LaFond, "The Effect of Dynamic
Visual Training on Manipulative Driving Skills" (Master's
thesis, Illinois State University, 1970), pp. 50-51.

 

38

characteristics and visual perceptual ability among
drivers.46

Dale W. Goby's study was similar to Lazarewicz's,
but used rural students. He states that

The findings of this investigation tend to indi-

cate that no significant difference exists between
visual perceptual abilities and behavioral types
among beginning motorists as measured by the test
instruments employed.

W. Laurance Quane's study was of greater magnitude
than either Lazarewicz's or Goby's including subjects from
suburban, inter city and city residential areas. This
greater magnitude and cross cultural approach allowed for
all ethnic, social, cultural, and economic backgrounds to
be included. Even so the results of the study showed no

direct relationship between personality factors and visual

perception capabilities as measured by the instruments

used.48

Jerrold Glassman developed a paper—pencil test to

measure knowledge of correct response in selected hazardous

 

46Robert S. Lazarewicz, ”The Relationship Between
Behavioral Characteristics and Visual Perception" (Master's
thesis, Illinois State University, January, 1970), p. 36.

47Dale W. Goby, "The Relationship Between Visual
Perceptual Abilities and Behavioral Categories Among Begin-
ning Motorists” (Master's thesis, Illinois State University,
1970), p. 39.

48W. Laurance Quane, ”The Relationships of Visual
Perceptual Capabilities as Measured by The Perception of
Traffic Hazards Test and Behavioral Categories as Measured
by the Mann Inventory" (Doctor's dissertation, Michigan
State University, 1970).

39

driving situations administered to three groups of driver
education students: (1) those who had been instructed in
the conventional way, (2) those who received programmed
instruction, and (3) those who had received seeding pro-
grammed instruction. The results indicated the control
group did as well as the programmed groups, but did not
retain the information as long as evaluated by a second
posttest. No difference appeared in the study between the
seeded and nonseeded program instruction.49
After giving three groups, novice, student, and
experienced drivers, tachistoscopic instruction in recog-
nition to collision—producing situations using two- and
three- dimensional color slides, Keith Barenklau tested
the three groups by tachistoscopic means. He found that
the instruction significantly improved their recognition
and responses to collision—producing situations. However,
"the results of responses obtained from the use of three-
dimensional slide images as opposed to those obtained from
the same slides shown in two dimensions”50 was not signif-

icantly different. Barenklau went to great length

 

49Jerrold Glassman, "The Effectiveness of A Teach-
ing Machine-Program as Compared With Traditional Instruc-
tion in the Learning of Correct Responses to Hazardous
Driving Situations" (Doctor's dissertation, New York Uni-

versity, 1965).

50Keith Edward Barenklau, "Improving Driver Recog-
nition of and Response to Collision Producing Situations
Through Tachistoscopic Instruction" (Doctor's dissertation,
The University of Wisconsin, 1972), p. 69.

 

40

producing the slides. All were developed using pictures
of scale models built by him. All situations were
developed in conjunction with the National Safety Coun-
cil's records to exact specifications.

James R. Adams developed a different driving haz-
ard perception test in which he researched methods to mea-
sure driver hazard perception ability by the use of photo-
graphs of traffic hazard situations. Each photograph is
covered with blocks. The subject, by removing blocks,
uncovers the hazard picture. Evaluation is based on the
number of blocks removed, the number of hazard cues des-
cribed, as well as consistency of response style. The
scores were compared with voluntary driving records fur-
nished by the subjects. Subjects tested were employees of
an insurance company, teachers in training, and peace
corps volunteers. It was found that smaller stimulus size
scores were associated with higher accident indexes (as
reported by subjects).51

Sanford Weinstein conducted a study that closely
parallels Adams, however, a small ll-inch by l4-inch rear
view projection screen was used in projecting color slides.
These pictures were covered with a small portion of the

picture being revealed at one time. The scores were

 

51James R. Adams, ”Measurement of Hazard Judgment
by a Stimulus Accretion Technique" (unpublished Safety
Research and Education Project, Teachers College, Columbia
University, February 1968).

 

 

41

determined in a manner similar to Adams. Volunteers from
a real estate company served as subjects. Two hypotheses
were tested. The first hypothesis was intended to measure
the subject's capacity or ability to recognize the hazards
presented in the slides and said that it would be inversely
related to an accident involvement score. This was not
confirmed; therefore, there was no statistical confirma-
tion between hazard recognition and accident involvement.

The second hypothesis indicated that those subjects
who were more cautious about making decisions when looking
at the hazard pictures also had low accident involvement.
This second hypothesis was confirmed. The author says
the "confirmation of hypothesis 2 indicates that a lack of
cautiousness is associated with accident involvement."52
However, it seems to this writer that those who took a
long time to decide on the hazards in the still picture
would not do well in a dynamic situation. Therefore, one
could come to just the opposite conclusions.

Weinstein states:

A major problem in the study of driver behavior

is the lack of adequate and reliable criteria for
measuring driver performance. The use of inadequate

criteria has led to confusion rather than knowledge
in the field of traffic safety.53

 

52Sanford Arthur Weinstein, "Recognition of Photo-
graphed Traffic Hazards as an Indicator of Automobile Ac-
cident Involvement" (Doctor's dissertation, Columbia Uni-
versity, 1970), p. 68.

531616., p. 16.

 

42

The Michigan State University Driver Performance
Measurement Research Project is a major step in develop-
ing the criteria for measuring driver performance.

Project METER conducted in the State of Washington
included among other tests a motion picture of traffic
events as viewed from the auto. The filmed traffic situa-
tions were used in conjunction with a simulator in an
experimental driver examination program. It was decided
that the simulator test could not be used to replace the
road test; however, improved film tests could provide
potentially useful road test supplements.54

At the end of an extensive literature review of
motion pictures and perceptual research, Jacqueline Herko-
witz states: ”The use of film in perceptual testing is a
relatively recent innovation. There has been little done
to date, but the future looks promising."55

Visual perception is a major factor in the avoid-
ance of traffic hazards. Why then has there not been more

work done in the development of evaluation instruments to

measure the phenomenon of traffic hazard perception?

 

54State Department of Motor Vehicles, "Project:
Machine, Examination, Teaching, Evaluation and Re-education'
(evaluation prepared under contract #Fh-11-6832 with the
U.S. Department of Transportation, Federal Highway Adminis-
tration, Washington, D.C., 1971), p. 35.

55Jacqueline Herkowitz, "Filmed Test to Assess
Elementary School-Aged Children's Perception of Embedded
Figures Which Appear to Move Away From Stationary Back-
grounds" (Doctor's dissertation, Purdue University, 1971),
p. 68.

 

43

This study is an attempt to start such a process
by the development of a Simulated Dynamic Driving Hazard
Perception Test and the administering of the instrument to

a sample population.

Summary

Literature relating to driving hazard perception
tests is limited, but literature dealing with perception
is plentiful.

A few early investigations in the field of visual
perception were cited such as the works of Samuel Renshaw
and Lester Luborsky which relate to the development of the
Second World War military pilot selection and aircraft
identification programs.

A few works that discussed the Army's Perceptual
Research Unit and Psychological Test Film Unit along with
a description of a sample motion picture perceptual test
were reviewed.

After a brief description of the research of Robert
Spicer in visual perception, a number of articles were
reviewed that dealt with human factors and perception.
Several works dealing with vision as related to perception
both in drivers as well as some studies in athletes were
included. Albert Burg indicated that there was a surpris—
ing lack of conclusive findings regarding the role of vision

in driving. He also gave some opinions as to the causes.

 

 

 

44

Next a variety of works using visual perception
tests of various sorts were reviewed, concluding with
driving hazard perception tests. Only a few driving haz-
ard perception tests have been developed. One has been
used by a number of researchers at Illinois State Univer—
sity.

Driving hazard perception tests using motion pic-
tures have not been pursued to any extent, however, it is
the belief of this author that efforts should be put forth

in this direction.

 

CHAPTER III

DESIGN AND METHODOLOGY

The Simulated Dynamic Driving Hazard
Perception Test

 

 

The Simulated Dynamic Driving Hazard Perception
Test was developed as a part of this project. It is
believed that no such test of this nature has been pre-
viously developed using wide-screen simulator motion pic-

tures .

Simulated Dynamic Drivinngazard
Perception Test Development

 

 

The idea first evolved in discussion with Dr.
Robert Nolan of the Michigan State University Highway Traf—
fic Safety Center as to the possibility of doing further
work with the study that had been done by William Laurance
Quane in the unpublished theses, ”The Relationship of Vis-
ual Perceptual Capabilities as Measured by the Perception
of Traffic Hazards Test and Behavioral Categories as Mea-
sured by the Mann Inventory.”

It was decided that a different approach to the
visual perception test might be a useful tool. A series
of discussions lead to the idea of using simulation films
for the purpose of developing some sort of dynamic driving

hazard perception test. 45

 

46

At first it was felt that a group of subjects
could view a short sequence of the film and then answer
questions on the test. This idea was similar to the
method used in the perception test developed at Illinois
State University. That test used 35mm slides rather than
16mm motion pictures. They also lacked the cinemascope
or wide-screen effect.

The group-showing idea was relinquished in favor
of a system designed to measure an individual's ability to
respond as he/she viewed the motion picture. A device was
developed that allowed the person being tested to stop the
projector when he/she saw a hazard or potential hazard
that in his/her opinion needed some action on the part of
the driver, such as acceleration, deceleration, braking,
steering right, steering left, etc. Through such a system
driver perception was measured.

To test the feasibility of the project an old sim-
ulator film was used from which was taken short sequences.
The frames of each sequence were numbered individually in
order to distinguish at what point the person being tested
had stopped the projector. Using this film a small pilot
study was conducted. It was evident from the success of
this trial that the endeavor should be continued.

Contact was made with Mr. Ralph Jackson of Allstate

Insurance Company who in a personal interview held in

 
  

47

Northbrook, Illinois, supported the proposed study and
agreed to provide Allstate Simulation films for the proj—
ect .

After reviewing the entire series, six films were
selected. The titles were: Critical Situations, Hazard-
ous Situations, Driving in Review, Control, Expressways
are Different, and Complex Traffic.

Sixty-five clips or sequences were cut, spliced,
and numbered. Sequences were selected because they showed
a variety of hazardous driving situations including city,
residential, rural, highway, and limited access highway
conditions and insofar as possible were representative of
the driving events included in the behavioral-environmental-
traffic-situational-sequences as depicted and defined in
the Michigan State University Driver Performance Measure—

ment Research project.

This was a process of selecting and diagraming
locations typical of everyday driving and describing
possible traffic sequences and driving behaviors to be
expected. The term 'BETSS' (Behavioral-Environmental—
Traffic-Situational-Sequences) resulted from a long
series of observations and discussion. This term em-
phasizes the dynamic interrelationships of behaviors
and changing traffic situations and helps to avoid a
tendency to think in terms of physical location char—
acteristics only.

1. A BETSS is defined therefore, as a Behavioral-
Environmental-Traffic-Situational-Sequence which in-
volves both changing traffic situations at a selected
location and possibly driving behavior patterns re-
lated to these.1

 

1Forbes and others, p. 57.

 
 

n-.|.l

 

 

 

48

A process to reduce the number of sequences to a
practicable number began. In the first evaluation, which
was conducted with volunteer students of the instructional
media class at Andrews University, driving hazard content
of the sequences were rated by each individual on a basis
of one to five, with five being the highest score. In the
second evaluation both student and faculty volunteers
from the Industrial Education Department of Andrews Uni-
versity rated the sequences from one to ten. Subsequently
the sixty-five were reduced to twenty-three sequences. At
this point a pilot study was conducted of twenty volunteers
selected from the student body, faculty, and staff of
Andrews University. They were tested individually, which
was the procedure intended to be used with the final
instrument.

Results from the pilot study included changes in
the instructions given to the participants, establishment
of the hazards to be named for each sequence, as well as
the film frame number to be used for scoring the sequence.
Also a number of sequences were deleted and two new ones
added to insure that sufficient sequences were available
with no apparent hazards.

The instrument was then ready for review by the
experts to validate the hazards and the scoring system.
They were asked to view each sequence of the instrument,

note the hazard, confirm its existence in that sequence,

49

and approve the frame number by which the examinee should
have stopped the projector in order to receive credit.

The recognition score was composed of the number
of sequences for which the subject stopped the projector
and correctly identified the hazard. The time limit score
consisted of the subset of the above sequences for which
the subject stopped the projector within the experts'
limit.

Both hazards and frame numbers had already been
worked out, but it was important to have the experts' con-
firmation as well as recommendations where adjustments
were to be made. When this was completed, the sequences
were renumbered and put in final form consisting of fifteen
clips, individually numbered and connected in a continuous
film of three hundred feet.

The first clip was used as a sample so that the
subject would have a full understanding of how he was to
respond to the instrument. Numbers 3 and 14 are controls
with no hazards. The possible score was 12 for the recog-

nition as well as the time limit score.

The Sample

 

Mr. Allen E. Bard, administrative assistant for
the Bureau of Driver and Vehicle Service, Michigan Depart-
ment of State, who served as one of the validation experts,
suggested that his department would be interested in having

the perception test used on two groups of people.

 

 

 

50

Subjects were grouped as follows: (1) Drivers
who had had three negligent accidents within the past two
years and were required to appear for reexamination at the
Driver Improvement Bureau. (2) Drivers who were renewing
their license and had had no accidents or violations on
their driving records. All subjects were from Lansing,
Michigan, or the immediate area and had been licensed
operators for a minimum of four years.

Rural country roads to cross-town expressways and
rural limited access highways as well as urban streets
are all represented in the Lansing area. Testing was con-
ducted for three consecutive weeks in November 1974, on
Tuesday, Wednesday, and Thursday. Each subject was tested
individually. The test took approximately fifteen to
twenty minutes.

Those who were required to appear for reexamination
did so by appointment. More time was given to these
individuals to allow time for the perception test in addi-
tion to the regular procedures of the reexamination.

Test Equipment and Development
Description

 

Equipment requirements demanded: (l) A 16mm cin-
emascope projector that could be stopped within two to
three frames by the subject pressing a stop button on a

hand-held control box. (2) A method of measuring and

 

51

recording the exact frames at which the film was stopped.
(3) A method of recording the subject's audible comments.

A Singer-Graflex 16mm projector especially modi—
fied by the researcher to include an electrically attached
stop button on a hand-held control box was used. In add-
ition a portable wide—screen was constructed as no com-
mercial screen was available.

Two systems were developed to measure at what
point the subject stopped the film. The first method was
by actually writing the frame numbers between each sprocket
hole on the film. The second system was a special frame
counter which was attached to the projector. This frame
counter was reset for each subject at a predetermined
point on the film.

A master control box was connected in such a man-
ner that when the subject stopped the projector the attend-
ant could deactivate the projector and at the same time
activate a room light and tape recorder to record all

audible comments.

Description of Test Procedures
The Driver Improvement Bureau provided a room in
which the equipment could be set up to conduct the test.
They also made appointments for those who were required to
appear for reexamination.
The criteria for selecting license renewal appli-

cants was a driving record free of violations and

 

52

accidents. When a subject met these initial requirements,
he/she was directed to participate in the SDDHPT which
replaced the normal renewal examination. As soon as a sub-
ject completed his/her test the next available applicant
meeting the selection criteria was referred to the study
examiner. This seemed to be the best substitute in that a
totally random procedure could not be used. It was assumed
that once the criteria were met the selection procedure

did not relate to driving factors. Therefore, there were
no factors related to the applicant's driving ability

other than the criteria that affected the selection process.
Those selected were asked to take the perception test in
place of the renewal test.

The selected applicant was escorted to the testing
room where he/she was asked to write his/her name, address,
license number, and years of driving experience on a test
answer sheet. Written instructions were given to him/her.
He/she was seated in a chair near the projector facing the
wide screen and given the control box. The subject was
asked to listen to some instructions from a tape recorder.
(These instructions were identical to the written instruc-
tions ) He/she was then told that his/her comments would
be recorded.

After viewing sequence number 1, which was the
sample, the subject was asked to listen to an explanation

of that event on the tape recorder. Thereafter each time

53

the subject stopped the projector the attendant recorded
the frame number and continuous frame counter number.
This procedure was done while the subject's comments were
being recorded.

In addition to the comments described above, each
subject was asked for his/her reaction to the test. These
comments did not seem appropriate to the data and there-
fore were not included in the research. At the end of the
test the subject was returned to the Driver License Bureau
or Driver Improvement Bureau.

Two types of data were gathered from each subject:
(1) verbal comments used to determine whether the subject
identified the proper hazard, and (2) the frame number at
which he/she stopped the projector which was used to eval-
uate whether or not he/she recognized the hazard in each
sequence within the experts' limit. The tape comments of

the examinees are discussed in Chapter IV.

The Null Hypotheses
The following is a restatement of the hypotheses
of this study in the null form, i.e., stating that no
significant relationships between the variables exist for
the purposes of the statistical treatments.

H 1: When measured by recognition of the hazards

0
as validated by traffic safety experts there will be no

significant difference at the .05 level between the

renewal and improvement groups.

 

54

H02: When measured by recognition of the hazards
within the limit set and validated by the traffic safety
experts, there will be no significant difference at the

.05 level between the renewal and improvement groups.

Analysis of Data

 

The data obtained were analyzed using a t—test of
significance.

A t-test of significance was employed to determine
the difference between the mean of the test scores on the
SDDHPT when measured only by recognition of hazards of the
individuals in the R Group and of individuals in the I
Group. An .05 level of significance was used to determine
the acceptance or rejection of the hypothesis H01.

A t-test of significance was employed to determine
the significance of the difference between the mean test
score on the SDDHPT when measured by recognition of the
hazard within the limit set and validated by the traffic
safety experts of the individuals in the R Group and of
individuals in the I Group. An .05 level of significance
was used to determine the acceptance or rejection of the

hypothesis H02.

Other Questions Considered
In addition to the hypotheses tested it was con-
sidered of value to look at the relationships between the

two groups Renewal and Improvement in each film sequence

 

55

and also compare the individual group means with the
experts' frame number in the respective film sequences.
Tables and figures were constructed to show these relation-
ships and to assist in the clarification of their discus-
sion. Finally a summary and discussion of the taped com-
ments for each sequence was included with each table of

the twelve sequences.

Summary

The Simulated Dynamic Driving Hazard Perception
Test was developed from six driving simulation films pro-
vided by Allstate Insurance Company. Selection of
sequences was based, when possible, on driving events
included in the behavioral—environmental-traffic-
situational—sequences as depicted and defined in the
Michigan State University Driver Performance Measurement
Research Project. Initially sixty-five sequences were cut
and spliced. Two evaluations reduced this number to twenty-
three. A pilot study was conducted and further reductions
and adjustments were made. A group of experts representa-
tive of traffic safety disciplines recommended final
adjustments confirming both the hazards and scoring system.

The sample consisted of two groups: (1) Drivers
who had three negligent accidents within the past two
years, and (2) drivers who had no accidents or violations

on their driving record.

 

56

The subject while viewing the wide-screen motion
picture test was required to press a stop button on a
hand-held control box when in the subject's opinion a
traffic hazard appeared in the film.

Two types of data were gathered from each subject:
(1) His/her verbal comments, and (2) the frame number at
which the subject stopped the projector. A t-test of
significance was employed to determine the difference
between the mean of the test scores of the two groups for
both the recognition of the hazard as well as recognition
of the hazard within the limit set by the experts. An
.05 level was used to determine significance.

In Chapter IV the results of the analysis appear
along with a summary of the subject's comments on the
individual sequences as well as a look at the relation-

ships within the individual sequences.

 

 

CHAPTER IV

ANALYSIS OF DATA

The results of the analysis of the data are pre-
sented in this chapter. These analyses include: (1) the
difference between the mean of the two groups on the basis
of recognition of hazards, and (2) the difference in the
mean of the two groups on the basis of recognition of the
hazards within the limit set by the traffic safety experts.

In addition other questions were considered such
as individual sequence relationships between groups and
comparisons of group means with the experts' frame number
in all sequences.

Differences Between Group Means on Basis
of Recognition Only

The following is a restatement of the null hypoth-
esis:

H01: When measured by recognition of the hazards
as validated by traffic safety experts there will be no
significant difference at the .05 level between Group R and
Group I.

A t-test of significance was employed to determine

the difference between the mean of the recognition scores

57

 

58

on the SDDHPT when measured by recognition of hazards of

the individuals in the R Group and of individuals in the I
Group. The mean of the recognition scores for the R Group
was 7.31 sequences and for the I Group was 7.10 sequences.

The difference between the means was .21 with a
confidence interval of i 1.07. Since this confidence
interval spans zero using a .05 level of significance,
the null hypothesis was not rejected.

Because the null hypothesis was not rejected there
is no indication of a statistically significant difference
between the two sample groups in their ability to recog-
nize the hazards in the SDDHPT.

Difference Between Group Means on Basis
of Recognition and Time

The following is a restatement of the null hypoth-
esis which was tested:

H02: When measured by recognition of the hazards
within the limit set and validated by the traffic safety
experts, there will be no significant difference at the
.05 level between Group R and Group I.

A t—test of significance was employed to determine
the significance of the difference between the mean time
limit scores on the SDDHPT when measured by recognition of
the hazards within the limit set and validated by the traf-

fic safety experts of individuals in the R and I Groups.

59

The mean of the time limit scores for the R Group was
2.79 sequences and for the I Group was 2.86 sequences.

The difference between the means was -.07 with a
confidence interval of i 1.04. Since this confidence
interval spans zero using a .05 level of significance
the null hypothesis was not rejected.

Since there was no statistical difference indicated
between the R and I groups in the recognition of the driving
hazards within the limit set by the traffic safety experts

in the SDDHPT, the null hypothesis was not rejected.

Other Questions Considered

To portray group relationships within individual
film sequences tables were constructed showing the follow-
ing information: (1) the number of subjects who stopped
the projector some time during the sequence and named the
proper hazard, (2) the number of subjects who stopped the
projector within the experts' frame number and indicated
the proper hazard, (3) the mean film frame number and
standard deviation for each of the above when applied to
the subject within each group, and (4) the data necessary
for determining the significance of the difference between
the mean for the R Group and the mean for the I Group.

A separate figure shows relationships expressed in
percent of the total number of subjects who stopped the
projector some time during the sequence and named the

proper hazard as well as the number of subjects who stopped

 
     

I . . .. W
i . . .. . 1-1.131... If“... . 511..-? 5... . . --.-.,

 

60

the projector within the experts' frame number and indi-
cated the proper hazard.

A summary of the subjects' comments as well as a
discussion of the data for each of the twelve sequences
follows. Please note that no table or figure is included
for the sample (Number 1) or for the control sequences 3

and 14 (which had no hazards).

Sequence 2

 

As the test car approaches a van parked at the
curb on a residential city street, a person exits the van

stepping into the path of the test car.

Table 1 shows that of the twenty-nine subjects in
each group only one from the R Group and two from the I
Group failed to recognize the hazard in sequence 2.

The mean frame number for stopping the projector
and recognizing the hazard was 159.43 frames into the
sequence for the R Group and 152.07 frames for the I Group,
the difference being 7.36 frames. This mean difference of
7.36 frames was larger than the calculated confidence
interval of 7.01 frames making a significant difference at
the .05 level with respect to the film location at which
they identified the hazard, thus the I Group did signif-
icantly better than the R Group.

Eight subjects in the R Group and twelve subjects

within the I Group recognized the hazard at or before the

 

61

experts' frame number which was 152 for this sequence.

The data show that the mean frame number for recognizing
the proper hazard within the experts' limit was 147.63
frames for Group R and 146.33 frames for Group I, the dif-
ference being 1.30 frames. The calculated confidence
interval was 5.88 frames which means that there was no
significant difference between the two groups where
identification of the hazard within the experts' limit

was concerned.

Figure 1 depicts graphically the fact that most
subjects in both groups were able to identify the hazard.
The data show that 97 percent of the R Group and 93 per-
cent of the I Group were successful.

Of interest, however, were the relatively few sub-
jects from each group who were able to identify the haz-
ard at or before the experts' frame number. Here the data
shows that only eight subjects or 28 percent of the twenty-
nine subjects in Group R and twelve or 41 percent of the
subjects in Group I correctly identified the hazard at or
before the established frame number.

The primary theme of the comments from both the
R and I Group centered around the person exiting the van.
The majority from each group saw and recognized this haz-
ard; however, they did not see it and decide it was a haz—
ard as soon as they should based on the experts' cut-off

point. There was group consensus that action should have

62

been taken sooner. The question is: Could they have
avoided the dangerous situation at the particular time
that they stopped the projector? It is assumed that those
who acted after the experts' limit (152) could not have

safely avoided the hazard.

 

63

.mocmonchHm mmumoHch

Amo. mHmsco mcaH<V Hm>pwucH mocopwwcoo mucmmmuaom

H mom a mdoouu w>HuoOQmou :H coHuwH>mp pumpcmum mucomouawm

H mam m aaouu w>Huomdmou CH meQOum mmB nobowmopa LoHSB um nonEoc wEmuw some mucmmouawm
H mom a masouo m>Huooamou CH muommnom Ho umpESC mucomouamm

.smouom ucmEo>oudEH um>Huo onu um coHumcHmeoon pow umoaam cu pouHsvmu muoz 0:3 mumwx
o3u coma ecu CHLuHB mucooHoom ucowHwa: mouse :uH3 muo>Hup ocHCuzucoBD mo mumwmcoo

.mucmoHoom paw mcomumHoH>
mo wouu puoowu wCH>HuU w LuHB wucmoHHddw Hm3ocou owcooHH ocHCuxucmBu Ho mumwmcou

.umo ummu osu Ho some osu OucH mafiadoum cm> wnu wuHXw COmuma m
.uoouuw xuwo Hmwucopwmou m co Luau ocu um pmxoma cm> w meromouddm umo ammo msu w<

b

.8
Ham .mom
Hx .mx

Hz .mz
H dsouu
m a50po

N mocwsvmm

”zoz

 

ANmH u .oz mEmuH .muuwaxmv
mw.m+ m.H no.H mm.qu NH HH.m mm.qu w .02 msmuw .muumdxm chuH3
common mo GOHuHcmooom

 

mocwscmm wcausp wewuzc<

 

 

 

 

 

 

 

 

 

 

*Ho.nH om.m ON.mH no.NmH mm om.o mq.mmH mm pumum: «0 coHuchoowm
mo Hm - mm Ham Hm. Hz mom mm mz
mm H muoomQSm Hmuoe mm H muowmnom HmuOH
H macho. m asouo

 

 

 

 

 

 

N mocmsvmm cu vaHoam
cogs nonfisc osmum .muuoaxm paw coHuchooou >3 pmusmmos mm coHuamouma comma; wcH>Hup H paw M masouo H mHan

 

64

100

90

80

70

60

50

PERCENT

40

NUMBER OF SUBJECTS

30

20

10

0
R 8: I SUBJECTS GROUPED FOR RECOGNITION AND TIME

 

FIGURE 1

HAZARD RECOGNITION OF SUBJECTS IN SEQUENCE 2

KEY
Sequence 2: As the test car approaches a van parked at the curb on a residential city
street, a person exits the van stepping into the path of the test car.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sample Grouping , _ . , _
Violations on their drivrng records.

R I

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past

two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

 

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

65

Sequence 4

 

On a two-lane road‘in a rural area an oncoming car
is being passed by another car forcing the test car off

the road.

Table 2 shows that all twenty-nine subjects in
each Group R and I recognized the hazard in Sequence 4.

The mean frame number for stopping the projector
and recognizing the hazard was 91.79 frames into the
sequence for the R Group and 91.28 frames for the I Group,
the difference being .51 frames. This difference was less
than the calculated confidence interval of 13.52 frames
which indicates no significant difference at the .05 level
with respect to the film location at which Groups R and I
identified the hazard.

Of the twelve subjects in the R Group and fourteen
subjects within the I Group who recognized the hazard at
or before the experts' frame number, which was 87 for this
sequence, data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 69.75
frames for Group R and 68.86 frames for Group I, the dif-
ference being .89 frames. This difference was less than
the calculated confidence interval of 10.58 frames which
indicates no significant difference at the .05 level
between the two groups within the experts' limit at which

they identified the hazard.

 

66

Figure 2 depicts graphically the fact that all
twenty-nine subjects in each group recognized the hazard.
This is the only sequence in which 100 percent of the
subjects did recognize the hazard, however, less than half
of the subjects from each group were able to identify the
hazard at or before the established frame number. Here
the data show that twelve subjects or 41 percent of the
twenty-nine subjects in Group R and fourteen or 48 percent
of the subjects in Group I correctly identified the hazard
at or before the established frame number.

There was a large difference in stopping points
for the individuals within each group. Even though each
person recognized the hazard, not all subjects recognized
the hazard in sufficient time to avoid serious consequences.

Subjects who recognized the hazard within the
agreed upon limit did so well below the required limit
which indicates that they found it early, undoubtedly in
time to take corrective action to avoid serious danger.

All subjects either stated or inferred that an
oncoming car was in their lane as a result of performing a
passing maneuver. Most of the subjects suggested at least
one or all of the following alternatives as a driver:
speed up, slow down and take the ditch to the right. Other
frequent comments referred to an oncoming car passing on a

curve with insufficient space for such a maneuver. The R

 

67

and 1 Groups were not notably different in their verbal

reactions to this film clip.

68

Amo. mHmavo deH<V Hm>ooucH monopwmcoo mucmmmudmm mo

H one x mmsouo o>Hoooamou CH coHomH>op pumpcmum mucowmuaom Ham .mom
H van m maoouu w>Huowamwu CH poddOum mp3 uOuuomoua LUHLB um Mensa: wEmuw some mucmmmuaom Hx .mx
H pom m maoouo w>Huooamou CH muowmpsm Ho umnEoc monommudwm Hz .mz

.smouom ucmEo>ouQEH uo>HuQ ago on cOHumcHmemou pom unmade ob pooHDUmu mpmz 0:3
momma oau ummd ozu :quH3 mucopHoom usumHHwoc woucu LDH3 wum>pr mchizucwzu wo wumHmcoo H dsouo

.mucmpwoom paw mcoHuwHoH>
Ho meow puooou wcH>Hup m LuHB mucmoHHddm Hmzocmu omcooHH ocHCrmucozu Ho mumwmcoo m asouo

.pmou ecu Hmo you been mnu
mcHouoH one uozuocm >2 common wcHop mH poo wcHEooco cm .moum Hmusu m CH pmou mcmHioau n so q mucosvwm n%mz

 

Hum n .02 mEmum .muumaxmv
mm.oaH mm. wo.~H ow.wo «H mo.qH mm.mo NH .02 oEmuH .muuoaxm GHLDHS
phonon Ho coHuchoomm

 

~92“ S. S. 2 mm. a 8 5.3 2. :6 R 85:86 9526 653.94
phonon Ho coHuchoomm

 

 

 

 

 

 

 

 

 

 

8 Hm - am Ham 5% Hz mam am «2
mm H muommnam HmuOH aw n muoonnsm HmuOH
H QDOHU m anouo

 

 

 

 

 

 

q deconvom Cu poHHaam
cons “mass: oEwuH .muuodxo paw coHuchoomu mp pouommos mm coHuaoouoa punnmn wcH>Hup H use M maoouu N mewH

 

69

100

90

80

70

60

50

PERCENT

40

NUMBER OF SUBJECTS

30

20

IO

0

 

R & I SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 2

HAZARD RECOGNITION OF SUBJECTS IN SEQUENCE 4

K EY
Sequence 4: On a two-lane road in a rural area, an oncoming car is being passed by
another car forcing the test car off the road.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sample Grouping . . . . .
Violations on their drivmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

 

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard

   

70

Seqoence 5

 

A police car pursues another vehicle. Both cars

pass the test car on a two-lane rural road.

Table 3 shows that of the twenty—nine subjects in
each group, eighteen from the R Group and nineteen from
the I Group recognized the proper hazard in sequence 5.

The mean frame number for stopping the projector
and recognizing the hazard was 251.67 frames into the
sequence for the R Group and 215.79 frames for the I Group,
the difference being 35.88 frames. This difference was
less than the calculated confidence interval of 53.56
frames which indicates no significant difference at the
.05 level with respect to the film location at which Groups
R and I identified the hazard.

Eleven subjects in the R Group and fifteen from the
I Group recognized the hazard at or before the experts'
frame number which was 260 frames for this sequence. The
data show that the mean frame number for the eleven recog-
nizing the hazard within the experts' limit was 190 frames
into the sequence for the R Group and 194.2 frames for the
individuals in Group I, the difference being 4.20 frames.
This difference was less than the calculated confidence
interval of 43.15 frames which indicates no significant
difference at the .05 level between the two groups within

the experts' limit at which they identified the hazard.

 

71

Figure 3 depicts graphically the fact that more
than half of the subjects in both groups were able to iden-
tify the hazard. The data show that eighteen subjects or
62 percent of the R Group and nineteen subjects or 66
percent of the I Group were successful.

Fewer subjects were able to identify the hazard
at or before the established frame number. The data indi-
cates that eleven subjects or 38 percent of the R Group
and fifteen or 52 percent of the subjects in the I Group
correctly identified the hazard at or before the estab—
lished frame number.

There are two interesting developments that should
be noted. The mean for Groups R and I were both better
than the experts' limit and data indicate too many subjects
stopped late into the sequence.

More subjects of the I Group specifically mentioned
police or cop rather than emergency vehicle than those of
the R Group. This might suggest that the I Group were
more conditioned to look for the police rather than other
types of emergency vehicles, or they interpret all flash-
ing lights as being the police instead of different emer-
gency vehicles. Many of the subjects in both groups
failed to see the flashing light of the emergency Vehicle
in their rear view mirror. These people interpreted the
problem to be the car that passed them in its effort to

avert the police. This indeed was a hazard, but they

 

72

should have identified the flashing light of the police
vehicle earlier and taken into account the problems which

the pursuit was precipitating.

 

'73

Hmo. mHmsqo mzaH<v Hm>poucH mocmpHmcoo mqummpamm

H paw m mdsooo o>Huomdmou CH coHunH>wp pumpcmum mucwmouowm

.H pom m masouo w>Huomammu CH pmdQOum mm3 noboonoua Loch um amass: wEmuw some mucmmouamm
H pom m maoouo w>Huooamou CH wuoomnpm Ho popes: monommuaom

.owousm uCoEm>ouaEH nm>Huo ecu on coHumcHmeowu poH umoddm ou pmuHovou cums 0:3
when» ozu ammo ecu cHLqu mucopHoom ucowHmec mouse :uHB mum>Hup ochizucmBD Ho mumHmcoo

.mucmpHoom can mcoHomHoH>
mo meow puoomu wcH>Hup m :uH3 mucmoHHaam HmBocwu omcooHH ochimucozu Ho momecoo

no
Ham .amm
HX .MX

Hz .mz
H asouo

m dsouo

.pmou House ocmHiosu m so one umou one mmma memo cuom .oHoHLo> uwsuocm mosmuoa one moHHoa < m oucoovmm

”mom

 

Aoom i .02 mEmuH .muumaxmv

women: we coHuchooom

 

pummmn Ho coHuchooom

 

 

 

 

 

 

 

 

 

 

ma.meu o~.q- am.m6 N.eaa m4 so.om .oaa an .62 pence .muppaxp gasps;
6m.mnH mm.mm NN.N6 ak.mHN as so.sa s6.HmN ma pocpsemm waspsp peausa<
no Hm . am How Hm Hz mom mm az
om n muuwmpom Hmuoe mm H muownpSm Hmuow
H asouo ,m asouo

 

 

 

 

 

 

m deconvom cu poHHaam c653

pagans osmuu .muuodxo pan coHuchoowu an venomous mm :oHuaoouod pupae: wcH>Hup H man a «macho

m mHan

 

74

100

90

80

70
(I)
.—
U
“H“ 60
g
j—
‘0 z
1 5° 23‘
O c:
a“.
5 40
an
2
D
2 30
20
in
o

 

R & I SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 3.

HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE 5

KEY ,
Sequence 5: A police car pursues another vehicle. Both cars pass the test car on a two-
lane rural road.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sample Grouping . _ . _ .
Violations on their drivmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Time: Number of subjects also expressed in percent who stopped projector within
experts‘ limit and indicated the proper hazard.

 

 

75

Sequence 6

 

A city street has two lanes in both directions.
The test car is in the center lane approaching an inter-
section where an oncoming truck makes a left turn. A
pedestrian appears from the left pushing a shopping cart

in the crosswalk in front of the vehicle.

Table 4 shows that of twenty-nine subjects in
each group R and I, five recognized the proper hazard in
sequence 6.

The mean frame number for stopping the projector
and recognizing the hazard was 96.4 frames into the
sequence for the R Group and 111.6 frames for Group I, the
difference being 15.2 frames. This difference was less
than the calculated confidence interval of 67.83 frames
which indicates no significant difference at the .05 level
with respect to the film location at which Groups R and I
identified the hazard.

One subject in each Group R and I recognized the
hazard at or before the experts' frame number which was
75 for this sequence.

The frame number for recognizing the hazard within
the experts' limit was 69 frames into the sequence for
the subject in Group R and 60 frames for the subject in
Group I.

Figure 4 depicts graphically the fact that few

subjects in both groups were able to identify the hazard

76

of the truck making a left—hand turn in on-coming traffic.
The data show that only five subjects or 17 percent in
each group actually recognized the hazard.

Note that almost all subjects from each group
were unable to identify the hazard at or before the estab-
lished frame number. Here the data show only one subject
or 3 percent from each group correctly identified the haz-
ard at or before the established frame number.

Interestingly five subjects from each Group R and
I recognized the hazard and commented about the truck mak-
ing a left turn in front of the test car. The remaining
comments in both groups almost exclusively named the
pedestrian with a shopping cart near the end of the
sequence as the hazard. To most subjects, however, the
truck apparently did not appear to be close enough for

the driver of the test car to take any action.

 

'77

Amo. mHmsvm mcaH<v Hm>umuCH oucwpwmcoo mqumwuamm

H cam m masouo m>Huomamou cw :oHumH>mp pumpCmum mucmmwuaom

H cam m masouo o>wuowammu a“ cwaQOum mmz pouomnoua nows3 um Menage wEmuw cams mucommuawm
H cam m masouo w>wuomammu CH muowmbam Ho umbEsc mucmmmuawm

.smmuam ucmEm>ouaEH um>HpQ msu um coHumcHmewop u0m umoaam ob nmpHsvmp mumB 0:3
mumwz ozu umma mcu cflfiuHB mucmwfluom ucwwHwac mops“ LuHB mum>Hup mchizucm3u mo mumflmcoo.

.mucopmoom paw mcoflumHoH>
we owuw ppoomu wsH>Huv m Luwz mucmoHHaam Hmzmcou mmcmoHH mcwcixuco3u mo mumwwcoo

.wHoHLm> mnu mo ucouw CH me3mmouo cab CH undo mafiaaonm m mafismsa ummH onu Eoum mummmmm
cmwuumovoa < .cuSu umoH m mome xosuu mGHEooco Cm mpmn3 cofiuoomuwucfi cm wcHLomouaaw

so
How .mam
HX .mx

Hz .mz
H anouo

m mzouu

mcmH Housmo osu CH ma umu ummu onH .msofluoouwp Soon a“ mwcmH ozu mm: uomuum muwo < o mommsvwm

nzmx

 

Amn fl .oz oEmuw .wuuoaxmv
.m 00 H mo H .02 mEmuw .muuoaxw awnuwz
pumumc mo COHUchoomm

 

mucosvmm mcHusc wefluxc<

 

 

 

 

 

 

 

 

 

mm.no+ N.mH - NO.H© s.szz m so.NN e.oo m wanna; mo aouuucwoomm
mo Hm . mm Ham Hm Hz mom mm mz
om u muoombsw Hmuoe mm u muoombsm Hmuoa
H asouo m asouo

 

 

 

 

 

 

o pudendum Cu poHHaam cons

ambasc oEmuw .muuoaxo paw cowuwcwooou an wousmmme mm cowumwouoa pumnmn wcH>Huv H van m masouo

a anmH

 

 

78

100

90

80

ID

60

50

PERCENT

40

NUMBER OF SUBJECTS

30

20

ID

 

D

R Bi I SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 4

HAZARD RECOGNITION OF SUBJECTS IN SEDUENCE 6

KEY Sequence 6: A city street has two lanes in both directions. The test car is in the center
lane approaching an intersection where an oncoming truck makes a left
turn. A pedestrian appears from the left pushing a shopping cart in the
crosswalk in front of the vehicle.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sample Grouping . . _ . _
Violations on their drivmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

79

Sequence 7
On a busy city street a car is waiting to exit an

alley near a traffic light as the test car approaches.

Of the twenty-nine subjects in each group, five
from the R Group and seven from the I Group recognized the
proper hazard in sequence 7 as shown in Table 5.

The mean frame number for stopping the projector
and recognizing the hazard was 175.4 frames into the
sequence for the R Group and 194 frames for the I Group,
the difference being 18.6 frames. This difference was less
than the calculated confidence interval of 30.79 frames
which indicates no significant difference at the .05 level
with respect to the film location at which Groups R and I
identified the hazard.

One subject in each Group R and I recognized the
hazard at or before the experts' frame number which was
157 for this sequence. The frame number for recognizing
the hazard within the experts' limit was 140 frames into
the sequence for the subject in Group R and 151 frames for
the subject in Group I.

Figure 5 graphically displays the fact that few
subjects in either Group R and I were able to identify the
hazard by showing that five subjects or 17 percent of the

R Group and seven subjects or 24 percent of the I Group

were successful.

80

Once again note that most subjects from each group
were unable to identify the hazard at or before the
established frame number. The data show only one subject
or 3 percent from each group correctly identifying the haz—
ard at or before the established frame number.

While few subjects from each group did recognize
the car in the alley to the right as the immediate problem,
most subjects referred to the line of cars stopping ahead

as the hazard.

 

81.

Amo. mHmzcm msdH<v Hm>umuCH mocmpfimcoo mucwmwuamm

H can a museum w>Huowdmmu CH :oHumH>wp wumvcmum mucmmmuaom

H cam m masouo m>Huowmmmu CH aquacum wmz nobomnoua £0H£3 um umbezc wEmuH :mmE mucwmmuamm
H paw m masouo w>Huowamwu CH muowmbsm mo umbEsc mucwmmuamm

.Dmmusm ucmEo>ouaEH uw>HuQ mnu um :oHumcHmeowu pow unmade cu pouHsvou whoa one
munch o3u umma mLu CHLuHB musopwoom ucwwHmec woucu LuHB mum>Hup ochihquzu Ho mumHmcoo

.mqupHoom paw mcoflumHoH>
mo mmum whoomu wcH>Huv m :uH3 mucmoHHadw Hmzmcmu mmaoHH mCHcizucwzu mo mumHmcoo

.mwzomouaam umo

no
Ham .mom
Hx .mx

HZ .MZ
H macho

m asouo

 

 

 

 

 

 

 

 

 

 

 

 

ummu wsu mm uLwHH onwmuu m uwoc zmHHm cm uon ob wCHuHm3 mH umo m poouum zuHo mmsb m co m mocmsvwm ”sex
AmmH u .02 oEmpm .muumdxm

Hz - HmH H qu H .02 mamuu .muumaxm sagas:

puwmmfi mo sowuwconmm

. I. . . . . mommadmm wcHusv meuxc<

on om+ oo wH i no «N qu m mm Hm q mmH m champ: Ho cowuwcmoomm
no Hm . am How Hm Hz mom mm mz
mm H wuoom95m Houow mm H muoombnm kuoH

H anouu m asouu

 

 

 

 

 

 

m mucosvom cu umHHaaw conz

o kebabs oEmum .muuwaxm paw cowuwcwoomu %b powsmmoe mm coHuamuuma pupae: wcH>Hpn H paw M masouo

m magma

  

82

100

90

BO

70

60

50

PERCENT

40

NUMBER OF SUBJECTS

3O

20

ID

0

 

R & I SUBJECTS GROUPED FDR RECOGNITION AND TIME

FIGURE 5

HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE 7

KEY
Sequence 7: On a busy city street a car is waiting to exit an alley near a traffic light as
the test car approaches.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sample Grouping . . . . .
Violations on their drivmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

 

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

83

Sequence 8

 

The four-way flashers are operating on a double-
parked car in a congested two-lane residential city street.
When the test car nears the double-parked vehicle a child
runs out from behind the vehicle across the path of the

CESt car .

Of the twenty—nine subjects in each group sixteen
from the R Group and thirteen from the I Group recognized
the proper hazard in sequence 8 as shown in Table 6.

The mean frame number for stopping the projector
and recognizing the hazard was 258.94 frames into the
sequence for the R Group and 262.92 frames for the I Group,
the difference being 3.98 frames. This difference was less
than the calculated confidence interval of 67.52 frames
which indicates no significant difference at the .05 level
with respect to the film location at which Groups R and I
identified the hazard.

Thirteen subjects in the R Group and ten from the I
Group recognized the hazard at or before the experts' frame
number which is 317 frames for this sequence.

The data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 223.15
frames into the sequence for the R Group and 226.5 frames
for Group I, the difference being 3.35 frames. This dif-
ference was less than the calculated confidence interval

of 46.32 frames which indicates no significant difference

 

84

at the .05 level between the two groups in their ability
to identify the hazard within the experts' limit.

Figure 6 depicts graphically that 55 percent of
the R Group and 45 percent of the I Group were successful
in identifying the hazard.

Approximately as many subjects were able to iden-
tify the hazard at or before the established frame number.
The data show that thirteen subjects or 45 percent of the
R Group and ten or 34 percent of those in Group I correctly
identified the hazard at or before the experts' frame num-
ber.

It is interesting to note that both groups stopped
well before the time limit, however, there was a great
difference in the stopping points.

Comments from both groups were primarily concerned
with two events, those who recognized the stopped vehicle
and its flashing lights as a hazard and, second, those who
considered the child darting across the road from the
stopped car as the hazard. Several subjects volunteered
advice as to what actions should be taken rather than nam-

ing the hazard itself.

 

855

Amo. mHmsvm mLQH<v Hm>uwucH wucmbwwcoo mucmmwuamm

H can m masouo m>Huomdmmu :H coHumH>wc bumbcmum mucmmwuawm

H vac m mdsouo m>Huomamwu CH boaaoum mm3 uOuommoua noH33 um nomads oemum cmoE masomoumom
H cam m maaouo m>Huumamwu CH muumnbsm Ho umbEDC mucmmmuamm

.smmusm quEw>ouaEH um>Huo mcu um coHumcHmemmu DOM umoaam ob bmuHsvmu ouoz 0:3

munch o3u ummd wzu CHLDH3 muCoUHoow ucwwHHwoc owusu :DHB muw>Hu© mchimucmBD mo wumHmcoo.

.mucobHoom can mcoHumHoH>
Ho omuw buoowu wcH>Hup w LuHa mucmoHHdam Hm3wcwu wmcmoHH mchimucw3u Ho mumfimcoo

.umo ummu msu Ho Luca onu mmouom wHoH£o> ecu vcHLmb Eouw uno mcpp
nHHLo m .oHoHLo> poxpmainbsov mcu mumps poo umou oLu co£3 .umouum quo HmHqumemu
mcmHio3u bmumwwcod m CH umo bmxumainbsoc m co wcHumuuao mum wumcmme mmziuaow mLH

no
How .mom
Hx .mx

Hz .mz
H asouo

m macho

w oocwsvom

"mmx

 

AHHM u .02 mEmuH .muumaxmv

wumums Ho coHuHcmoowm

 

bumumn wo cowuchoomm

 

 

 

 

 

 

 

 

 

mm.oqfi mm.m i oo.om n.0NN OH om.qq mH.mmm mH .oz mEmuw .muuwaxo awnuHB
~m.nsu wo.m - m~.am No.NoN ma mo.mw as.wm~ as mocmsamm manage asauxc<
no Hm . am How Hm Hz mam mm mz
mm H muoombsm HmuoH om u muommbsm Hmuoe
H nacho .m asouu

 

 

 

 

 

 

m mocwsvom cu boHHaam Coca

ywbssc oEmuw .muuomxm use coHuchoumu ma bmuSmme mm :oHuamouma wanna: mcH>Hup 4

ll

cam m masouu

o pomH

 

    
 
  
  

KEY

Sample

 

86

100

90

BO

70

BO

50

PERCENT

4O

30

20

IO

0

R & I SUBJECTS GROUPED FDR RECOGNITION AND TIME

FIGURE 6

HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE 8

(I)
p—
Q
n.
a
an
3
(D
u.
D
G
LU
m
2
D
2

Sequence 8:

Grouping R:

I:

Rec:

Time:

 

The four-way flashers are operating on a double-parked car in a congested
two-lane residential city street. When the test car nears the double-parked
vehicle, a child runs out from behind the vehicle across the path of the

test car.

Consists of 29 drivers license renewal applicants who had no accidents or
violations on their driving records.

Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Number of subjects also expressed in percent who stopped projector 'within
experts' limit and indicated the proper hazard.

 

87

Sequence 9

 

A two-lane country road has an intersection desig-
nated in advance by a sign. A car is waiting to cross the
intersection from the right. .As the test vehicle approaches,

a car from the left speeds through the intersection.

Table 7 shows that of the twenty-nine subjects in
each group eleven from the R Group and five from.the I
Group recognized the proper hazard in sequence 9.

The mean frame number for stopping the projector
and recognizing the hazard was 331.91 frames into the
sequence for the R Group and 309.2 frames for the I Group,
the difference being 22.71 frames. This difference was
less than the calculated confidence interval of 77.25
frames which indicates no significant difference at the
.05 level with respect to the film location at which
Groups R and I identified the hazard.

Four subjects in the R Group and five from the I
Group recognized the hazard at or before the experts'
frame number which was 337 frames for this sequence.

The data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 253.25
frames into the sequence for the R Group and 274 frames
for Group I, the difference being 20.75 frames. This dif-
ference was less than the calculated confidence interval

of 97.75 frames which indicates no significant difference

 

88

at the .05 level between the two groups identifying the haz-
ard within the experts' limit.

Figure 7 shows that a limited number of subjects
were able to identify the hazard. The data show thirty-
eight percent of the R Group and 17 percent of the I Group
were successful.

Few subjects were able to identify the hazard at
or before the established frame number. The data show
that only four subjects or 14 percent of the subjects in
Group R and three or 10 percent of the subjects in Group
I correctly identified the hazard at or before the estab-
lished frame number.

The majority of the comments directed themselves
to the car crossing the intersection from the left without
making a stop. A few subjects referred to the car on the
right waiting to cross the intersection. Some subjects
did suggest that the intersection ahead was indicated by
a sign along the road. Members of the R Group mentioned

this part more than those of the I Group.

 

89

Amo. mHmsvm 0:QH<V Hm>umucH wocwwacoo mucmmouamm

H can m masouo 0>Huooamwu CH cowumw>00 bumvcmum mucmmmuamm

H 0cm x manouo 0>Huuwammu CH vmaaODw mm3 nobommoua :0H:3 um um:6:: macaw amps mucmmmuaom
H can m masouo 0>Hu00dm0u CH muommbsm m0 um:E:c mucmmmpamm

.smwusm acmem>0uaEH u0>Huo 0:0 um coHumcHEQXmmu now “momma Cu pmuwnvmu mums 0:3 mumm»
03p umma 0:0 GH:DH3 mucwbfioom bamwfiwac 00p:u :0H3 muw>Hub ochi%0003u H0 mumwmcov

.mucobHoom 0cm mGOHumH0H>
mo 00pm 0u000u wcH>Huw m :uHa mquoHHadm H03000u mmcmoHH mewcizucw3u Ho mumHmcoo

.cowuowmpoucH 0:0 :wsou:u mpwoam ume 0:0 Eouw umo
m .m0:0m0uamm mHoH:0> umwu 0:0 m< .u:wHu 0:u 80pm cofluoomumucH 0:0 mmOpo 00 wcHuHmz

so
How .mom
Hx ix

Hz .mz
H dsouu

m asouo

0H umo < .cme m H: mocm>0m CH bmumcwwmmv :0HuommuwDCH cm mm: bmou mpucsoo mcmHiozu < a mosesvmm

”hex

 

Ammm u .02 mEmuw .muumaxmv
ma.nmfl mx.om - NH.sq «AN m H3.Nm mN.mmN e .02 «swam .muuaaxa canons
pumum: H0 coHuchooom

 

m~.n~H HA.~N + mN.mm N.mom m ma.as Ho.smm HA accesses weapon menuxc<
cummm: H0 COHDchoomm

 

 

 

 

 

 

 

 

 

 

m0 Hm.- mm Ham Hm. Hz mam mm mz
¢N u muoomnsm Hmuoa am n muommbsm Hmuoa
H asouo m anouo

 

 

 

 

 

 

m wocmsvmm Ou vmeaam cw:3

ambasc wswum .muuoaxw 0cm cOHuchooou »: wouSmon mm coHuaoouoa pumnm: wcH>Huv H 000 a maaouo

n memH

 

90

100

90

BO

70

60

50

PERCENT

40

NUMBER OF SUBJECTS

30

20

IO

0

 

R & I SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 7

HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE 9

KEY Sequence 9: A two-lane country road has an intersection designated in advance by a
sign. A car is waiting to cross the intersection from the right. As the test
vehicle approaches, a car from the left speeds through the intersection.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sam Ie Grouping , , ,
p Violations on their drivmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

91

Sequence 10

 

On a country road the test car approaches a dis-
abled car barely off the road, its four—way flashers oper—
ating. A car is approaching in the opposite lane when
suddenly children dash out from in front of the disabled

vehicle.

Table 8 shows that of the twenty-nine subjects in
each group nineteen from the R Group and twenty-one from
the I Group recognized the proper hazard in sequence 10.

The mean frame number for stopping the projector
and indicating recognition of the hazard was 211.05 frames
into the sequence for the R Group and 228.81 frames for the
I Group, the difference being 17.76 frames. This differ—
ence was less than the calculated confidence interval of
33.57 frames which indicates no significant difference at
the .05 level with respect to the film location at which
Groups R and I identified the hazard.

Three subjects in the R Group and two from the I
Group recognized the hazard at or before the experts'
frame number which was 155 frames for this sequence.

The data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 111.0
frames into the sequence for the R Group and 107.5 frames
for Group I, the difference being 3.5 frames. This dif-

ference was less than the calculated confidence interval

 

92

of 67.84 frames which indicates no significant difference
at the .05 level between the two groups within the experts'
limit and at which they identified the hazard.

Figure 8 portrays that the majority of subjects
were able to identify the hazard. It indicates 66 percent
of the R Group and 72 percent of the I Group were success-
ful.

Few subjects from each group were able to identify
the hazard at or before the experts' frame number. Figure
8 indicates that only three subjects or 10 percent of
Group R and two or 7 percent of the subjects in Group I
correctly identified the hazard at or before the experts'
frame number.

A considerable number of subjects in the R and I
Groups recognized the disabled car and commented about it.
Some subjects referred to the car while others mentioned
only the flashers. Many criticized its location without
commenting that it might also be a hazard. The majority
of subjects remarked about the oncoming vehicle and the
children dashing into the roadway, however, many subjects

recognized it too late and reserved their criticism for

the children in the road.

 

933

H 000 m 003000 0>00000000 00 0000000 003 00000n000 :00:3 00 000630 06000 0006

:000 0000H0:0 0H000030 00:3 000H 00000000 0:0 00 w00:0000000 00 000 <

H 000 m 003000 0>00000000 00 000000>00 00000000

H 000 m 003000 0>00000000 00 0000nb30 00 000630

Amo. 0H0300 0:0H<V H0>00000 0000000000

0000000000
0000000000
000000000m

000000000m

.30003m 00060>0006H 00>00Q 0:0 00 000000060x000 000 000000 00 00003000 0003 0:3 0000»
030 0000 0:0 00:003 000000000 000w0Hw00 000:0 :003 000>000 0000i>00030 00 00000000

.000000000 000 000000H00>

00 0000 000000 w00>000 0 :003 000000H000 H030000 000000H 0000i>00030 00 00000000

.0H00:0> 00H:0000 0:0 00 00000 00 6000 030
.m00000000 000:00H0

00
000 .000
00 .mx

HZ .mz
H 03000

m 03000

 

 

 

 

 

 

 

 

 

 

 

>03i0300 000 .0000 0:0 000 0H000: 000 00H:0000 0 00:0000000 000 0000 0:0 0000 0000300 0 00 OH 00003000 ">00
AmmH u .02 06000 .00000me
0w.00+ m.m 00.0H m.00H N 0N.mm o.HHH m .oz 06000 .00000x0 000003
00000: 00 000000w000m
00.nm0 00.00 - «0.00 00.000 00 0.00 00.000 00 00005600 000050 000006<
00000: 00 000000w000m
00 Hm - mi 000 0m Hz 000 0M 02
mm H 0000mb30 H000H mm H 00000030 H0009

000630 06000

 

 

H 03000

 

 

a 03000

 

 

.00000x0 000 000000w0000 >0 00030006 00 0000000000 00000: w00>000 H 000 0 003000

OH 00003000 00 000H000 00:3

w 0H00H

 

 

KEY

Sample

 

94

100

90

80

70

60

50

PERCENT

40

30

20

IO

0

R 81 I SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 8

HAZARD RECOGNITION OF SUBJECTS IN SEQUENCE 10

U)
.—
C.)
u.
a
a:
D
U)
LL
D
I
LU
an
2
2
2
Sequence 10:

Grouping R:

Rec:

Time:

 

On a country road the test car approaches a disabled car barely off the road,
its four-way flashers operating. A car is approaching in the opposite lane
when suddenly children dash out from in front of the disabled vehicle.

Consists of 29 drivers license renewal applicants who had no accidents or
violations on their driving records.

Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the pr0per hazard.

 

95

Sequence ll

 

The test car drives up a snow-covered, slippery
hill in a residential neighborhood. At the top of the
hill a car partially obscured by high snow banks is back-

ing from a driveway.

Table 9 shows that of the twenty-nine subjects in
each group twenty-eight from the R Group and twenty-five
from the I Group recognized the proper hazard in sequence
11.

The mean frame number for stopping the projector
and recognizing the hazard was 175.39 frames into the
sequence for the R Group and 188.36 frames for the I Group,
the difference being 12.97 frames. This difference was
less than the calculated confidence interval of 15.74
frames which indicates no significant difference at the
.05 level with respect to the film location at which
Groups R and I identified the hazard.

Six subjects in the R Group and two from the I
Group recognized the hazard at or before the experts'
frame number which is 161 frames for this sequence.

The data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 132.67
frames into the sequence for the R Group and 159.0 frames
for Group I, the difference being 26.33 frames. This dif-
ference was less than the calculated confidence interval

of 74.84 frames which indicates no significant difference

 

96

at the .05 level between the two groups within the experts
limit at which they identified the hazard.

Figure 9 shows that most subjects in both groups
were able to identify the hazard. Ninety-seven percent of
the R Group and 86 percent of the I Group were successful.

Of interest, however, were the relatively few sub-
jects from each group who were able to identify the haz—
ard at or before the established frame number. Here the
data show that only six subjects or 21 percent of the sub-
jects in Group R and two or 7 percent of the subjects in
Group I correctly identified the hazard at or before the
established number.

All or most of the subjects, experts' limit not-
withstanding, talked about or referred to the car backing
out between the snow banks. Many mentioned the snow and
slippery road conditions. One subject was concerned about
the slippery hill before the backing car was even visible.
Most of these comments relative to braking on ice came

from the R Group.

 

€97

Amo. mHmsvm m£QH<V Hm>umucH mocmpHHcoo mucmmmuamm

H mom a mdsouo m>Huowammu CH :oHuwH>mp pumpcmum mucmmouamm

H paw m mdsouo o>Huowamwu CH poQQOum mmz acuoomoua noHLB um pagans wEmuH :me mucmmmuamm

H paw m wmaouo m>Huomammu CH muommnsw Ho amass: mucmmmuaom

.smwuzm ucmEm>0paEH pm>HuQ mLu um coHumcHmemmu ho“ ummaam cu pmuHscmp muoz 0:3 mumoa

so

osu umma wsu :quHB mucmpHoom ucmeHwo: mouxu nuH3 mum>Huv ocHsuhuCoau Ho mumwmcoo

.mucmpHoom pcm mcoHumHoH>
mo wmum puoumu wcH>Hup w LuH3 mucmoHHadm Hmamcmp mmcoUHH mcHCuxucm3u Ho mumHmcov

.xm3w>Huu m EouH wconmn wH mxcmn 30am :wH: >3 pmusomno AHHmHuumd umo m HHHL o£u mo mob mnu

no
How .mom
HX .mx

HZ .mz
H anouo

m macho

 

 

 

 

 

 

 

 

 

 

 

 

u< .poozuonstmc HwHucmvamu m :H HHH: %uodaHHm .pmuo>oou3ocm m as mm>pr umo ummu m£H HH wocwsvmm ”xox
I AHoH n .02 mEmpH .muummxmv
cm.q~+ mm.o~ u mm.~ .omH N oo.oq mo.NmH 0 .oz wEmuH .muuoaxo CHLqu
pumum: Ho coHuHcmoowm
. l . . . . . mosmsvmm wcHusv oeHumc<
on mH+ mm NH mm mm om me mm mm mm mm mnH mm vumwm: Ho coHuHcmoumm
mo Hx - ax Ham Hx Hz mam mm mz
mm a muownnsm HmuoH an n wuuonnsm Hmuoe
H asouu -m msouo

ouonEsc mamum .muuoaxo cam coHuchooou zn

 

 

 

 

 

 

HH mucosvwm cu pmHHaam cons

cwusmmme mm coHuaoouwa pumumn wcH>Hup H vcm m masouu

a ”Hams

 

KEY

Sample

 

98

100

90

80

70

BO

50

PERCENT

4O

30

20

IO

0

R & l SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 9

HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE II

(I)
.—
C.)
u.
a
co
3
(D
u.
C
C
u
a:
E
D
2
Sequence 11:

Grouping R:

I:

Rec:

Time:

 

The test car drives up a snow-covered, slippery hill in a residential neighbor-
hood. At the top of the hill a car partially obscured by high snow banks is
backing from a driveway.

Consists of 29 drivers license renewal applicants who had no accidents or
violations on their driving records.

Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

99

Sequence 12

 

As the test car drives down a freeway during a
rainstorm, another car attempting to merge comes care-

lessly close to the front of the test car.

Of the twenty-nine subjects in each group twenty-
eight from the R Group and twenty-five from the I Group
recognized the proper hazard in sequence 12 as shown in
Table 10.

The mean frame number for stopping the projector
and recognizing the hazard was 221.68 frames into the
sequence for the R Group and 228.6 frames for the I Group,
the difference being 6.92 frames. This difference was less
than the calculated confidence interval of 24.5 frames
which indicates no significant difference at the .05 level
with respect to the film location at which Groups R and I
identified the hazard.

Sixteen subjects in the R Group and twelve from the
I Group recognized the hazard at or before the experts'
frame number which is 240 frames for this sequence.

The data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 193.06
frames into the sequence for the R Group and 187.17 frames
for Group I, the difference being 5.89 frames. This dif-
ference was less than the calculated confidence interval

of 23.35 frames which indicates no significant difference

 

100

between the two groups in their ability at the .05 level
to recognize the hazard.

Most subjects in both groups were able to identify
the hazard as shown in Figure 10. Data indicate that 97
percent of the R Group and 86 percent of the I Group were
successful.

Approximately one half of the subjects from each
group were able to identify the hazard at or before the
established frame number. Here the information reveals
that sixteen subjects or 55 percent of Group R and twelve
or 41 percent of the subjects in Group I correctly iden-
tified the hazard at or before the established frame num-
ber.

Most subjects in both groups referred to the car
merging from the right. Many were eager to describe how
to handle the situation. Some suggested increasing their
speed while others would slow down. The lane on the free-
way to the left of the test car was available. A large
number suggested a lane change. About an equal number in
both groups commented about the rain. Several subjects
in the improvement group referred to the merging lane as
a cross or side street.

It is interesting to note that the mean frame num-
ber for stopping the projector for both groups was below

the experts' film frame number of 240.

 

Amo. mHmsvo mLaH<v Hm>uoucH mocoprcoo moCmmmudmm mo

H can m masopo o>Huowamop CH coHumH>op numpcmum mucomouaom Ham .mom

H can m mQDOpU w>Huooammu CH poddOum mp3 wouocmOpd soch on pecans oEmuH some mucomwuawm Hx .mx

H pcm m masouu o>Huooamwu CH muoomLSm Ho umbEDC mucmwouaom Hz .mz
.smouam quEo>ouaEH uw>Huo osu um aoHumcHwaomu pow unmade Ou pouHsvmu ouoz 0:3

whom» o3u ummd ecu cHsqu mucouHoom ucwwHHwoc moucu LuH3 mpo>Hup ochizucmzu Ho wumwmcou H asouo
.mucopHoom bum mcoHumHoH>

Ho mouw puooou wcH>Hup m LuHB mucmoHHddm Hmzocmu omcooHH ochi%uco3u Ho wumflmcoo m macho

.pmo umou osu Ho ucouH ecu

cu omoHo kamonumo moEoo owuoe

Ou wcHuanuum umo umcuocm .EuOumchu m wcHusp >m3ooum m GBOv mo>Hup pmo umou wLu m<

NH mucosvom ”mom

 

1(31

Aoqm u .oz mEmuH .muppdxmv
mm.m~+ mm.m mm.mm NH.mmH NH HN.©N oo.mmH 0H .02 mamum .muumaxw cucuwz
pumumn mo coHuchooom

 

monosvom wswusv mEHuhc<

 

 

 

 

 

 

 

 

 

 

m.q~H mm.o - mo.mq o.wNN mN w©.am wo.HNN mm
pumnmn Ho coHuchooom
m0 Hm I MM. HOW Hm HZ mDm MM m2
mm H muoomnsw HmuOH mm H muoomQSm HMuOH
H asouo m QDOpo

 

 

 

 

 

 

NH mucosvow Ou noHHaam Gena
umpEnc oEmuw .muuoaxo can aoHuHcmooou an cousmmoe mm coHuaoopoa bumwms wcH>Hup H paw m masouo OH oHan

 

102

100

90

BO

70

BO

50

PERCENT

40

NUMBER OF SUBJECTS

30

20

10

O

 

R & l SUBJECTS GROUPED FOR RECOGNITION AND TIME

FIGURE 10

HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE 12

KEY . .
Sequence 12: As the test car drives down a freeway during a rainstorm, another car

attempting to merge comes carelessly close to the front of the test car.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sam le Grouping . , , , ,
p Violations on their drlvmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

 

103

Sequence 13

 

The test car is in the center lane of a three-
lane limited access highway with a car fast approaching

from the rear.

In Table 11, of the twenty-nine subjects in each
group thirteen from the R Group and sixteen from the I
Group recognized the proper hazard in sequence 13.

The mean frame number for stopping the projector
and recognizing the hazard was 290 frames into the
sequence for the R Group and 228.24 frames for the I Group,
a difference of 61.76 frames. This difference was less
than the calculated confidence interval of 68.7 frames
which indicates no significant difference at the .05 level
with respect to the film location at which Groups R and I
identified the hazard. In fact one subject of Group R
and four subjects of Group I recognized the hazard at or
before the experts' frame number which was 175 for this
sequence. The one subject's frame number was 174 while the
mean for the four individuals in Group I was 154.75 frames,
the difference being 16.25 frames. This difference does
not span the calculated confidence interval of 50.38
frames which indicates no significant difference in recog—
nizing the hazard of the two groups within the experts'
limit.

The data in Figure 11 shows the limited number

of subjects in both groups who were able to identify

 

104

the hazard, 45 percent of the R Group and 55 percent of
the I Group.

Very few subjects were able to identify the hazard
at or before the established frame number. Considering
this percentage wise, only 3 percent of the R Group and
14 percent of the I Group correctly identified the hazard
at or before the established frame number of 175. More
subjects from the I Group noticed and mentioned the car
coming from the rear, however, more subjects in the R

Group were concerned about normal freeway traffic flow.

 

 

‘—

105

 

 

Amo. mHmsvo CCCH<V Hm>uouCH ooCoCHHCOo muComouamm mo

H CCC m maoouo o>Huooamou CH CoHumH>op pumpCmuw muCommuaom Ham .mam
H CCC m masouo o>Huooamou CH Coaaoum mm3 u0uoomoua CoHC3 um umnECC oEmuw Coos mqumouaom Hx .m»
H pCm M masouw o>Huooamou CH wuoonbnm Ho wonECC muCommuaom Hz .mz

.Cmousm uCuEw>ouaEH uo>HuQ «Cu uw COHuwCHmeoou uoH unmade Ou Coufisvou muo3 0:}
undo» oau umma oCu CHCuHB mucopHoom uComHHwoC woufiu CuHB mpo>Hup wCHCizquau Ho mumHmCoo H asouo

.muCopHoom CCm mCoHumHoH>
Ho ovum puOowu wCH>Hup m Cows mquoHHaam Hmzocou omCooHH oCHanuCoBD Ho mumHmCoo m Csouo

.umwu osu Eouw wawnomouaam ummm
Coo m CuHs >m3£ch mmmoom wouHEHH oCmHuoopnu m we oCmH woquo onu CH mH umo umwu och MH ooCmsvmm n%wm

 

AmNH u .02 mECCH .muuoaxmv
mm.0mH mN.©H «Hiva anNmH 4» HNH H .02 wEmum .muumaXm c.2335
vumwm: Ho COHuHCwooom

 

 

 

 

 

 

 

 

 

 

5.on 3.3 8.: «~52 S 3.2: 903 2 8533 $526 9515
cumwms Ho COHuHCwoomm
.6 HM - Cm Ham Hm Hz mam mm mz
ow u wuomnnsm kuoe am u muownnsm Hmuoa
H anouo m macho

 

 

 

 

 

 

MH ooCvaom cu onHmam Cos3
wonECC osmuw .muuoaxo va CoHuHCwooou ha wousmwoe ma COHuCoouoa CCCNCL wCH>HuC H CCC m masouo HH oHan

 

 

106

29 100

90
25

  
 
 
 
 
 
 
 
 
 
 
 
 
 
  

BO

70

 

m 20
.—
Q
g 60
m
5’ 5
O a:
‘i‘
5 40
a:
z 30
20
5
10
o o
n a I SUBJECTS GROUPED FOR RECOGNITION AND TIME
FIGURE II
HAZARD RECOGNITION OF SUBJECTS IN SEQUENCE I3
KEY

Sequence 13: The test car is in the center lane of a three-lane limited access highway with
a car fast approaching from the rear.

R: Consists of 29 drivers license renewal applicants who had no accidents or
violations on their driving records.

Sample Grouping

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

 

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

 

107

Sequence 15

 

As the test car is traveling on a congested free-
way, the car ahead reduces its speed, finally brakes,

pulling off the freeway to a stop on the emergency strip.

In sequence 15, twelve of the twenty-nine subjects
in Group R and thirteen in Group I recognized the proper
hazard as shown in Table 12.

The mean frame number for stopping the projector
and recognizing the hazard was 339.5 frames into the
sequence for the R Group and 319.69 frames for the I Group,
the difference being 19.81 frames. This difference was
less than the calculated confidence interval of 48.96
frames which indicates no significant difference at the
.05 level between groups in identifying the hazard.

Five subjects in the R Group and seven from the I
Group recognized the hazard at or before the experts' frame
number which is 335 frames for this sequence.

The data show that the mean frame number for recog-
nizing the hazard within the experts' limit was 296.8
frames into the sequence for the R Group and 276.57 frames
for Group I, the difference being 20.23 frames. This dif-
ference was less than the calculated confidence interval
of 73.42 frames which indicates no significant difference
at the .05 level between the two groups within the experts'

limit at which they identified the hazard.

 

 

108

The R Groups' 339.5 mean film frame number is only
4.5 frames above the experts' film frame number while the
I Groups' 319.95 is 15.31 frames less than the experts'
film frame number (335). I

More than half the subjects in both groups were
unable to identify the hazard. The data in Figure 12
reveals that 41 percent of the R Group and 45 percent
of the I Group accomplished the task.

In addition very few subjects from each group were
able to identify the hazard at or before the established
frame number. The data show that only five or 17 percent
of the twenty-nine subjects in Group R and seven or 24
percent of the subjects in Group I correctly identified
the hazard at or before the established frame number.

Many subjects in both groups recognized the vehicle
ahead slowing, not considering it a hazard until the brake
lights came on. However, they suggested cars approaching

in adjacent lanes from the rear as hazards.

 

 

 

Amo. mHvao CCQH<V Hm>uouCH oocopHHCoo mqummuaom mo

H pCm m manouo o>Huomammu CH COHumH>wC pumpCmum muComouamm Ham .mam
H va m maaouo o>HuooCmou CH pmaaoum mm3 Couomhoum CUHCS um wonECC osmum Coos mqumuuawm Hx .mx
H va m museum o>Huooamou CH muoomnsm Ho nonECC muComwuaom Hz .mz

.Cmousm uCoEo>ouaEH Cm>Huo onu um CoHumCHECxoou CON umomdm cu CouHsvou ouo3 0:3
mumoh o3u uwma osu CHCqu mqupHoom uCowHmeC moucu CuHa mCo>HCp wCHCimuCoBu Ho mumHmCoo H Czouo

.muCopHoom pCm mCOHumHoH>
Ho ooum buooou wCH>Hup m CuHs mquoHHaam HCBmCoC omCmoHH wCHCimuchD Ho mumHmCoo m dsouo

.aHuum zoCowuoEo oCu Co acum m Cu mmamoum osu HHo wCHHHCQ .mmxmun AHHCCHH
.nmoam muH moospmu Cause you osu .hm3oouw pwumowCoo C CO wCHHo>muu mH Cmo umou osu m4 mH woCosvmm ”aux

 

109

 

 

 

 

 

 

 

 

 

 

Hmmm u .02 mfimuw .muuooxmv
Nc.mu+ MN.0N nq.oN mm.oNN N mm.HN w.omN m .02 oEmuH .muumaxo CHCuHB
pumum: Ho CoHuHCmoomM
8.3M 32 2.: $52 2 8.? 92m .2 8833 9:58 23.92
phone; Ho CoHuHCwooom

«HQ me I mm Ham Hm HZ mam mm. m2

aN n muoonnsm HCuOH mN u muoomnam HCDOH

H asouu .M asouo

 

 

 

 

 

 

mH oUCosvom Ou CoHHaaw Con3
wonECC mawum .muuoaxo pcm CoHuchooou an wouammoe mm CoHuaoouoa pumwms wCHpHup H va m maaouo NH oHan

 

 

110

100

90

BO

   

70
so
.—
Q
3 so
2
w E
m 50 g
Q m
E
5 40
an
2
= I
z 30 I.
20
10
O
R & l SUBJECTS GROUPEO FOR RECOGNITION AND TIME
FIGURE 12
HAZARD RECOGNITION OF SUBJECTS IN SEOUENCE 15
KEY .
Sequence 15: As the test car is traveling on a congested freeway, the car ahead reduces Its
speed, finally brakes, pulling off the freeway to a stop on the emergency
strip.

R: Consists of 29 drivers license renewal applicants who had no accidents or

Sample Grouping . . . . .
Violations on their dnvmg records.

I: Consists of 29 drivers who were required to appear at the Driver Improvement
Bureau for reexamination because of three negligent accidents within the past
two years.

Rec: Number of subjects also expressed in percent who stopped projector before
end of sequence and named proper hazard.

Time: Number of subjects also expressed in percent who stopped projector within
experts' limit and indicated the proper hazard.

 

111

Summary

The results of the analysis of the data reveal:
(1) when measured only by recognition of hazards a t-test
of significance indicated no significant difference at the
.05 level between the mean score of the R and I groups;
(2) when measured by recognition of the hazards within
the time set and validated by traffic safety experts a
t-test of significance indicated no significant difference
at the .05 level between the mean scores of the R and I
groups.

To show relationships within each of the individual
film sequences, a table was constructed showing (1) the
number of subjects who stopped the projector some time
during the sequence and named the proper hazard; (2) the
number of subjects who stopped the projector within the
experts' limit and indicated the proper hazard; (3) the
mean film frame number and standard deviation for each of
the above applied to the respective groups, and (4) data
to determine the significance of the mean difference
between the R and I Groups. Also subjects recognizing the
hazard as well as those recognizing the hazard within the
set limits were expressed numerically as well as in percent
and displayed in graphical form.

A summary of the subjects' comments as well as a

discussion of the data preceded each table.

 

 

 

CHAPTER V

SUMMARY, MAJOR FINDINGS, CONCLUSIONS,
RECOMMENDATIONS, AND DISCUSSIONS

Summary

The primary purpose of this thesis was to develop
an instrument to measure a phase of that process called
dynamic driver perception, primarily that of driving haz-
ard recognition. A secondary purpose was to determine how
drivers respond to the instrument.

To accomplish these purposes two tasks were under—
taken: (1) a motion picture film sequence instrument was
developed to be used as a test in evaluating the ability
of drivers to recognize driving hazards, and (2) a means
to evaluate the manner in which drivers responded to the
instrument was accomplished.

The instrument consisted of fifteen, 16mm film
clips or sequences depicting various types of driving haz—
ards called the Simulated Dynamic Driving Hazard Percep-
tion Test (SDDHPT).

The SDDHPT was developed from six driving simula—
tion films provided by Allstate Insurance Company. Selec—

tion of sequences was based, when possible, on driving

112

 

 
 

 

113

events included in the behavioral-environmental-traffic-
situational-sequences as depicted and defined in the
Michigan State University Driver Performance Measurement
Research Project. Sixty-five sequences were cut and
spliced. Two evaluations reduced this number to twenty-
three. A pilot study was conducted after which further
modifications were made. A group of experts representative
of traffic safety disciplines recommended final adjustments

confirming both the hazards and grading systems.

 

The sample consisted of two groups: (1) Drivers
who had three negligent accidents within the past two
years and (2) Drivers who had no accidents or violations
on their driving record.

Subjects viewed the wide-screen motion picture
test on an individual basis and were required to press a
stop button on a hand-held control box when in the sub-
ject's opinion a traffic hazard appeared in the film.

Two types of data were gathered from each subject: his/
her verbal comments and the numerical scores for each
sequence.

A t-test of significance was employed to determine
the difference between the mean of the test scores of the
two groups for both the recognition of the hazard and
recognition of the hazard within the limit set by the ex-
perts. An .05 confidence level was used to determine sig-

nificance.

 

114

To see relationships within each of the individual
film sequences, a table was constructed showing (1) the
number of subjects who stopped the projector some time
during the sequence and named the proper hazard; (2) the
number of subjects who stopped the projector within the
experts' limit and indicated the proper hazard; (3) the
mean film frame number and standard deviation for each of
the above applied to the respective groups, and (4) data
to determine the significance of the mean difference

between the R and I Groups. Also subjects recognizing

 

the hazard as well as those recognizing the hazard within
the set limits were expressed numerically, in percentage,

and displayed in graphical form.

Major Findings

The following is a restatement of the null hypoth-
eses which were tested:

H01: When measured by recognition of the hazards
as validated by traffic safety experts there will be no
significant difference at the .05 level between the renewal
and improvement groups.

H02: When measured by recognition of the hazards
within the limit set and validated by the traffic safety
experts, there will be no significant difference at the

.05 level between the renewal and improvement groups.

The results of the analysis of the data revealed:

 

 

 

115

1. When measured only by recognition of hazards,
a t-test of significance indicated no significant differ-
ence at the .05 level between the mean score of the R and
I Groups.

2. When measured by recognition of the hazards
within the time set and validated by traffic safety
experts, a t-test of significance indicated no signifi-
cant difference at the .05 level between the mean scores
of the R and I Groups.

3. Findings from each sequence indicate that
except for sequence two there was no significant difference
with respect to the film location at which the two groups
identified the hazard.

4. With respect to the identification of the haz-
ard within the experts' limit there was no significant
difference between groups in all sequences.

5. The majority of subjects in Groups R and I
were unable to identify the hazards confirmed by the
experts in six of twelve sequences.

6. Data show that many subjects recognized the
correct hazards, but few recognized the hazards within the

experts' limit.

Conclusions

 

l. The SDDHPT did not distinguish between the R
and I Groups for either the identification of hazards or

the recognition of hazards within the experts' limit.

 

116

2. Often hazards were recognized too late to take
corrective action.

3. The great variation in test scores in relation
to time indicate that some drivers have acquired well-
developed perceptual skills while other drivers lack such
skills.

4. The SDDHPT did produce a variation in scores
which indicates it did distinguish among drivers which
could be due to chance.

5. The groups tested may not have been different
as drivers in their hazard perception abilities.

6. An accident and violation-free driving record
does not guarantee that an individual will recognize driv-
ing hazards when applied to this study.

7. A standard other than violations and accident
involvement for grouping subjects might produce different
results.

8. Drivers are not using systematized search pro-
cedures.

9. In general, drivers wait too long to make

decisions.

Recommendations
1. A study should be undertaken to determine if
persons identified as problem drivers differ in visual per-
ception capability from recognized safe commercial drivers

as measured by the SDDHPT.

 

 

117

2. Research should be undertaken in which a cor-
relation of the SDDHPT with another measure of visual per-
ceptual capabilities is made.

3. A replication of this investigation using
another visual perceptual measure should be done.

4. Research should be instituted in which a cor-
relation of the SDDHPT with the Michigan State University
Driver Performance Measurement instrument should be eval-
uated.

5. A correlation of the SDDHPT with the Project
METER (Machine, Examination, Teaching, Evaluation and Re-
education) should be conducted.

6. To establish norms applicable to the general
population, large groups of matched subjects should be

tested by the SDDHPT.

Discussion

 

The SDDHPT did not distinguish between the R and I
Groups in terms of recognition of hazards nor recognition
of these hazards within the experts' limit when compared
by t-tests of the means. However, it did produce a vari-
ation in raw scores which indicate that it did distinguish
among drivers. The problem is, are the two groups that
were tested different as drivers or not? If the Michigan
State University Driver Performance Measurement instru-
ment had been used on the same subjects, would those doing

well on it also do well on the SDDHPT and vice versa?

 

 

118

Even though a person's record shows no violations or acci—
dents, it does not necessarily guarantee that that indi-
vidual is a good or poor driver. A standard other than
violation and accident involvement is necessary to eval-
uate drivers. Perhaps the SDDHPT does distinguish
between good and poor drivers if in fact a reliable
standard is used to make that decision. However, its
intent was to measure only driving hazard recognition and
one's ability to recognize such hazards early.

The experts agreed upon the hazard and the recog-
nition time necessary in each film sequence. The test then
administered showed a variation of scores which indicated
that some drivers respond before others.

The mean of the time limit scores for both the R
and I Groups were extremely low, 2.79 and 2.86 sequences
respectively. Why did so many subjects have so much
trouble recognizing the hazard within the experts' limit?
Maybe the drivers were poor or the experts were too ideal-
istic in what they thought drivers could or should see and
respond to in a dynamic situation. The key to the traffic
safety problem could very well be locked up in such a dis-
crepancy between the idealistic represented by the experts
and reality represented by the subjects tested. Further

research with the SDDHPT using a large sample should prove

helpful.

 

 

 

 

 

119

Each of the studies, Adams, Weinstein, Spicer,
the METER Project, and this study differ in how they
attempt to measure driver perception. The test used in
this research falls somewhere between that which Spicer did
and the elaborate computerized system of the METER Project.
However, if it can do the job of measuring the driver's
ability in hazard recognition, why have the costly equip-
ment? Neither Adams or Weinstein used motion pictures
which in this author's opinion was an omission.

In Project METER a motion picture was used in place
of the driver's road test in a license examination. This
caused the researchers to discard the simulation film and
produce their own so as to be able to carry the subject
further into hazardous situations. Whereas in the SDDHPT
the experts desired early recognition which did not call
for the hazard to go beyond that which the simulation film
accomplished.

An extensive amount of research has been completed
pertaining to the driver and driving task. A few examples
have been cited. This study dealt with one aspect of more
extensive research, primarily that of testing driver hazard
recognition by the use of the SDDHPT. It would seem that
the above test with certain refinements could be used on
a group of drivers who were tested with the Michigan State

University Driver Performance Measurement Instrument. If

 

{‘1'

120

there was high correlation between the two tests, then
the SDDHPT could be used on a large scale at reasonable
expense.

Nearly all evaluation of drivers has been on the
basis of accident or traffic violation involvement rather
than on performance based on safe driving habits and pro-
cedures.

It is very likely that a sizeable number of drivers
who are never involved in an accident or cited for a vio-
lation of the law are in reality very poor drivers. This
may be one additional factor involved in the changing
accident population.

Part of the rationale in this study was the assump-
tion that the Allstate Simulation Films would be more real
than any other known, practical simulation system. Sub-
jects reacted to a color motion picture projected on a
wide screen which duplicated the real life driving environ-
ment. This is in great contrast to investigations like
Weinstein and Adams in which still pictures were used of
near notebook size.

One of the most valuable uses of the SDDHPT may
well be as a teaching tool for driver education teachers to
help students in their early recognition of hazards in

various driving situations.

 

 

 

 

BIBLIOGRAPHY

121

BIBLIOGRAPHY

Books

Aaron, James E. and Marland K. Strasser. Drivinngask
Instruction Dual-Control, Simulation, and Multiple-
Car. New York: Macmillan Company, Inc., 1974.

 

. Driver and Traffic Safety Education. New York:
The Macmillan Company, 1966.

 

Allen, Merrill.J. Vision and Highway_Safety, New York:
Clinton Book Company, 1970.

Ditchburn, R.W. Eye-Movements and Visual Perception.
Oxford: Clarendon Press, 1973.

Forbes, T. W. (ed.) Human Factors in Highwey Traffic
Safety Research. New York: Wiley-Interscience,
1972i

 

Graham, Clarence H. Vision and Visual Perception. New
York: John H. Wiley and Sons, 1966.

Hochberg, Julian E. Perception. Englewood Cliffs, N.J.:
Prentice-Hall, Inc., 1964.

 

Insurance Institute for Highway Safety. Driver Behavior.
Washington, D.C.: Insurance Institute for Highway
Safety, 1968.

Julesz, Bela. Foundations of Cyclopean Perception. Chi-
cago: University of Chicago Press, 1971.

Lawther, John D. The Learning of Physical Skills. Engle-
wood Cliffs, NJJ.: Prentice-Hall, Inc., 1968.

Leibowitz, Herschel W. Visual Perception. New York: The
Macmillan Company, 1969.

Sperling, Abraham P. Ps cholo Made Sim 1e. Garden City,
New York: Garden City BOOKS, 195?.

Vernon, M.D. (ed.) Experiments in Visual Perception. New
York: Penguin Books, I966.

122

 

6r:

 

 

123

Weintraub, Daniel J. and Walker, Edward L. Perception.

Willman,

Barrett,

Barrett,

Bloomer,

Bonsall,

Belmont, California: Brooks/Cole Publishing Comp-
any, 1966.

Charles W. Seeing and Perceiving. Oxford:
Pergamon Press, 1966.
Journals and Magazines
Gerald V. "Relation Between Embedded Figures

Test Performance and Simulator Behavior." Journal
of Applied Psychology, LIII, 3 (1969), 253—354.

 

Gerald V., and Thornton, Carl L. "Relationship
Between Perceptual Style and Driver Reaction to an
Emergency Situation." Journal of Applied Psy:
chology, LII, 2, (1968), 160-176.

"Relationship Between Perceptual Style and Sim-
ulator Sickness." Journal of Applied Psychology,
LII, 4 (1968), 304-308.

Richard H. "Perceptual Defense and Vigilance and
Driving Safety.” An Independent Journal for Better
Traffic, XVI, 4 (October 1962).

C. and Dornbush, R. L. "Visual Perception and
Reading Ability." Journal of Educational Psychol-
ogy. LC (1969), 294-300.

Burg, Albert. ”Some Preliminary Findings Concerning the

Relation Between Vision and Driving Performance.”
Journal of the American Optometric Association,
XXXVIII, 5 (May 1967), 372-377.

Brown, J. "Evidence for a Selective Process During Percep-

Carney,

Coan, R.

tion of Tachistoscopically Presented Stimuli."
Journal of Experimental Psychology, LIX, (1960),
176-181.

Eugene D. "Another Look at Perception." Journal
of Traffic Safety Education, XIX, 2 (1972), 31—32,
38.

J. ”Factors in Movement Perception." Journal of
Consulting Psychology, XXVIII, (1964), 394-402.

Cumming, R. W., ”The Analysis of Skills in Driving."

Australian Road Research, I, (March 4—14, 1964).

 

 

 

 

124

Damron, C. F. "Two- and Three-Dimensional Slide Images
Used With Tachistoscopic Training Techniques in
Instructing High School Football Players in De—
fenses." Research Quarterly, XXVI (1955), 36-43.

 

Dick, A. C. "Relations Between the Sensory Register and
Short-Term Storage in Tachistoscopic Recognition."
Journal of Experimental Peychology, LXXXII, (1969),
279-284.

Egeth, H. H. and Smith, E. E. "Perceptual Selectivity in
a Visual Recognition Task." Journal of Experiment-
al Psychology, LXXIV, (1967), 543-549.

 

Erikson, Charles W. "Object Location in a Complex Percep-
tual Field." Journal of Experimental Psychology,
XXXXV, (1953),Tl26-I32i

"Location of Objects in a Visual Display as a
Function of the Number of Dimensions on Which the
Objects Differ.” Journal of Experimental Psychol-
ggy, XXXIV, (1952), 56-60.

Flanagan, John C. "The Selection and Classification Pro-
gram for Aviation Cadets." Journal of Consulting;
Peychology, XI, (1942), 229-240.

 

Forbes, T. W. "Human Factors in Highway Safety." Traffic
Safety, IV, March, 1960, 8-11.

Frazier, Damron C. "Two and Three-Dimensional Slide Images
Used With Tachistoscopic Training Techniques in
Instructing High School Football Players in De-
fenses." Research Quarterly) XXVI (1955), 36-43.

 

Fruchter, Benjamin and Mahan, William W. "Some Perceptual
Factors Measured by Motion Picture Tests.”
Journal of Educational Peychology, XXXXIII (1952),
430-435.

Gallahue, David L. ”The Relationship Between Perceptual
and Motor Abilities." Research Quarterly, XXXIX
(1968), 948-953.

Heimstra, Norman W. and others. ”Effects of Operator Mood
on Performance in a Simulated Driving Task." Per-
ceptual and Motor Skills, XXV, December 1967, 729-

735.

 

 

   

125

Hochberg, Julian and Virginia Brooks. ”Effects of Pre-
viously Associated Annoying Stimuli (Auditory)
on Visual Recognition Thresholds." Journal of

Experimental Psychology, LV, 5 (19585, 490-49I.

' Johnson, H. M. "Detection and Treatment of the Accident
Prone Driver." Peychological Bulletin, XLII
(November 1946), 489-533.

 

Kenel, Francis C. ”Employment of Teaching Assistants in
Driver Education." Safety, April 1970, 14.

Londree, Ben R., Jr. "Effects of Training with Motion
Pictures Versus Flash Cards Upon Football Play
Recognition." Research Quarterly, XXXVIII, (May
1967), 202-207.

 

Low, Frank N. "Some Characteristics of Peripheral Visual

Performance." American Journal of Physiology,
(July 1946), 573-584.

 

Luborsky, Lester. "Aircraft Recognition: A Study of
Prognostic Tests.” Journal of Applied Psychology,
XXIX, 5 (December 1945), 449-457.

”Aircraft Recognition: I. The Relative Effi-
ciency of Teaching Procedures." Journal of Applied
Peychology, XXIX, 5 (October 1945), 385-398.

 

 

Mackworth, N.S. and Bruner, J. S. ”How Adults and Children
Search and Recognize Pictures." Human Development,
XIII, 1970, 149-158.

 

 

Man, Cecil W. "Visual Factors in the Perception of Ver-
ticality." Journal of Experimental Peychology,
XXXXIV, (1952), 460.

McFarland, Ross A. and Moseley, Alfred L. ”Human Factors
in Highway Transport Safety.” Boston, Mass.:

Harvard School of Public Health, 1954.

Mourant, Ronald R. and Rockwell, Thomas H. "Mapping Eye
Movement Patterns to the Visual Scene in Driving:
An Exploratory Study." Human Factors, 12:81-87,
February 1970.

 

Payne, Donald E. and Barmack, Joseph E. ”An Experimental
Field Test of The Smith-Cummings-Sherman Driver
Training System." Research Review.

 

 

 

 

126

 

Quensel, Warren P. ”Teaching Visual Percept on Driver
Education.” ADEA News and Views, 3:4, May 1963.

Renshaw, Samuel. "The Visual Perception and Reproduction
of Forms by Tachistoscopic Methods." The Journal

 

of Peychology, XX (1945), 217-232.

 

Slater-Hammel, A. T. "Reaction Time to Light Stimuli in
the Peripheral Visual Field.” Research Quarterly,
XXVI (March 1955), 82-87.

Smith, Harold L. and Cummings, John J. ”Let's Teach
Drivers How to See." Traffic Digest and Review,
IV, 2, February 1956.

Staff of the Psychological Branch, Office of the Air Sur-
geon Headquarters Army Air Forces, The Aviation
Psychology Program of the Army Air Forces, Ps -
chological Bulletin, XXXX (1943), 759-769.

 

Staff, Psychological Research Unit No. 1. History, Organ-
ization and Procedures Psychological Research Unit
No. 1, Army Air Forces. Psychological Bulletin,
XXXX (1944), 103-114.

Staff, Psychological Test Film Unit. History, Organiza-
tion and Research Activities, Psychological Test
Film Unit, Army Air Forces. Psychological Bulletin,
XXXXI (1944), 457—468.

Staff, Psychological Test Film Unit Santa Anna Army Air
Bases, Santa Anna, California, History, Organiza-
tion, and Research Activities, Psychological Test
Film Unit Army Air Forces. Psychological Bulletin,
XXXXI (1944), 463-465.

Swets, J. A., Tanner, W. P. and Birdsall, T. G. ”Decision
Processes in Perception." Psychological Review,
LXVIII, (1961), 301-340.

Thomas, E. L. "Movements of the Eye.” Scientific Ameri-
can, 218:88, August 1968.

Warrick, M. J., Kibleu, A. W., and Topmiller, D. A. "Re-
sponse Time to Unexpected Stimuli." Human Factors,

VII, 1 1965, 81-86.

Wittreich, W. J. "Visual Perception and Personality.”
Scientific American, April 1959, 56-74.

 

 

 

127

Bulletins and Reports

 

Allen, Merrill.l. Seeing and Safety, Analogy. Allstate

Insurance Company, Northbrook, I inois, 1971.

Bartlett, Neil R., Bartz, A. E., and Wait, J. V. ”Recog-

Burg, Albert and Coppin, Ronald S. "Visual Acuity and

Domey, Richard G., and Paterson, Donald. "Development of

Farber,

Farber,

Forbes,

Gibson,

Gordon,

Gordon,

nition Time for Symbols in Peripheral Vision."

Highway Research Board Bulletin 330. Presented
at the 4lst Annual Meeting, January 8-12, 1962,
Washington, D.C.

Driving Record.” Highway Research Record, Number
122. Presented at the 44th Annual Meeting, Jan-
uary 11-15, 1965, Washington, D.C.

a Vehicle Simulator for Evaluation Driver Perfor-

mance." Highway Research Board Bulletin 330. ix
Presented at the 4lst Annual Meeting, January 8- ‘
12, 1962, Washington, D.C.

 

Eugene and Silver, Carl A. "Behavior of Drivers
Performing a Flying Pass." Highway Research Rec-
ord, Number 247. Washington, D.C., 1968.

Eugene, Silver, Carl A., and Landis, Daniel.
"Knowledge of Closing Rate Versus Knowledge of On-
coming-car Speed as Determiners of Driver Passing
Behavior.” Highway Research Record, Number 247.
Washington, D.C., 1968.

T. W. and others. "Driver Performance Measurement
Research." A Technical Report, Michigan State
University, Vol. I, February 1973.

James (ed.). Motion Picture Testing and Research
Report No. 7. U.S. Army Air Force. Aviation Psy-
chology Program. Washington, D.C.: Government
Printing Office, 1947.

Donald A. "Experimental Isolation of the Driver's
Visual Input." Highway Research Record, Number 122.
Presented at the 44th Annual Meeting, January 11-
15, 1965, Washington, D.C.

Donald A. and Michaels, Richard M. "Static and
Dynamic Visual Fields in Vehicular Guidance.”
Highway Research Record, Number 84. Presented at
the 43rd Annual Meeting, January 13-17, 1964,
Washington, D.C.

128

Gordon, Donald A. and Truman M. Mast. "Drivers' Decisions
in Overtaking and Passing." Highway Research Rec-
ord, Number 247. Washington, D.C., 1968.

Gould, Lawrence N. An Optometrist Looks at Perception,
(May 1969). A paper presented at the International
Reading Association Conference, Kansas City, Mis-
souri, April 30, May 1, 1969.

Greenshields, Bruce D. "Changes in Driver Performance
with Time in Driving." Highway Research Record,
Number 122. Presented at the 44th Annual Meeting.
January 11-15, 1965, Washington, D.C.

Heimstra, Norman W. and others. ”The Effects of 'Stress
and Fatigue' on Performance in a Driving Device."
Final Report, Supported by the U.S. Department
of Commegce Bureau of Public Roads, Washington,
D.C., l9 5.

 

Heimstra, N.W., Mast, Truman, M., and Larrabee, L. L.
”Effects of Fatigue on Basic Processes Involved
in Human Operator Performance: I. Simple Vigi-
lance and Target Detection," Highway Research
Record Number 55. Presented at the 43rd Annual
Meeting, January 13-17, 1964, Washington, D.C.

A Program for Driver and Traffic Safety Education. How to
Handle Driving.Emergencies, [Highway Users Feder-
ation for Safety and Mobility]. Washington, D.C.

1970.
Jones, Howard V. and Heimstra, Norman W. ”Ability of
Drivers to Make Critical Passing Judgments." High-

way Research Record, Number 122. Presented at the
44th Annual Meeting, January 11-15, 1965. Washing-
ton, D.C.

Kaluger, N. A. and Smith, G. L. Jr. ”Driver Eye-Movements
Patterns Under Conditions of Prolonged Driving and
Sleep Deprivation." Highway Research Record,
Number 336. Washington, D.C., 1970.

King, Barry G. and Sutro, Peter J. "Dynamic Visual Fields."
Highway Research Board Bulletin 152. Presented at
the 35th Annual Meeting, January 17-20, 1956.
Washington, D.C.

Mast, Truman M., Jones, H. V., and Heimstra, N.W. "Effects
of Fatigue on Performance in a Driving Device."
Highway Research Record, Number 122. Presented

 

 

 

 

129

at the 44th Annual Meeting, January 11-15, 1965,
Washington, D.C.

McFarland, Ross A. "Human Factors in Highway Transport
Safety.” Paper presented at the ASE Golden Anni-
versary Transportation Meeting, St. Louis, Mis-
souri, November 2, 1955.

Merrill, Irving R. ”Attitude Films and Attitude Change."
Paper of Research Grant supported by the Division
of General Medical Sciences. Public Health Ser-
vice, U.S. Department of Health, Education and
Welfare. Vol. 10, No. l, January-February 1962.

"Films in Driver Education.” Michigan State
University. Highway Traffic Safety Center.

Merrill, Irving R. and.McAShan, Hildreth H. "Predicting
Learning, Attitude Shift, and Skill Improvement
from a Traffic Safety Film." Research Grant from
the Division of General Medical Sciences, Public
Health Service, U.S. Department of Health, Educa-
tion and Welfare. Michigan State University.

Michigan State University. "Brief Description of MSU:
East Lansing, February, 1972.

Miller, M. M. ”A Study of Some Accidents on Motorway M.
4 in Fog.” Crowthorne Road Research Laboratory.
1967.

 

Michaels, Richard M. and Solomon, David. "Effect of Speed
Change Information on Spacing Between Vehicles."
Highway Research Board Bulletin 330. Presented at
the 4lst Annual Meeting, January 8-12, 1962.
Washington, D.C.

Newsome, L. R. "A Perceptual Factor That Could Contribute
to Road Accidents." Road Research Laboratory Min-
istry of Traneport, RRL Report LR 135, 1967.

 

Nolan, R. O. and others. Guide for training Observer/
Raters in the Driver Performance Measurement Pro-
cedure. Final Report, Vol. 2, Michigan State
University. February 1973.

O'Day, James. Proceedings of the Second Annual Traffic
Safepy Researcthymposium of the Automobile Insur-
ance Industgy, Insurance Institute for Highway
Safety. Washington, D.C., 1968.

 

 

130

Platt, Fletcher N. "Operations Analysis of Traffic Safety."
Dearborn, Michigan, 1959.

"A Unique Method of Measuring Road, Traffic,
Vehicle and Driver Characteristics." Presented
at the IV World Meeting of International Road Fed-
eration, Madrid. October 14-20, 1962.

”A New Method of Evaluating the Effects of
Fatigue on Driver Performance." Presented at the
42nd Annual Meeting, Highway Research Board, Jan-
uary 10, 1963.

"Operations Research in Driver Behavior."
Presented at the National Automobile Week Society
of Automotive Engineers. Sheraton-Cadillac Hotel.
Detroit, Michigan. March 13-17, 1961.

 

Reilly, Raymond E. and others. "The Translation of Visual
Information into Vehicular Control Actions." Re-
port Biotechnology Incorporated. Arlington, Vir-
ginia. October, 1965.

Richards, Oscar W. "Vision at Levels of Night Road Illum-
ination.” Literature 1967-1969. Highway Research
Record, Number 336. Washington, D.C., 1970.

Rockwell, Thomas H. and others. "Drivers' Eye Movements:
An Apparatus and Calibration." Highway Research
Record, Number 247. Washington, D.C., 1968.

Ross, H. Laurence. "Ignorance of Collision Course as a
Factor in Traffic Accidents." Traffic Institute.

Northwestern University. Evanston, Illinois, 1960.

Safety Education Films, Inc. "Safer Driving." Manual for
Driving All Roads. Minneapolis, Minnesota. 1956.

Schlesinger, Lawrence E., and Safren, Miriam A. "Percep-
tual Analysis of the Driving Task." Paper read at
Annual Meeting of Highway Research Board, January
1964.

Shell Oil Company. Perception of Driving Hazards. (Film-
strip Series Part 1), Center for Safety Education.
(1967) 6.

Shell Traffic Safety Center. Perception of Driving Hazards,
Part 1, Urban and Suburban, (n.d.).

Shell Traffic Safety Center. Perception of Driving Hazards,
Part II, Limited Access Highways, (n.d.).

131

Shell Traffic Safety Center. Perception of Driving Haz-
ards, Part III, Highways and Byways, (n.d.).

Silver, Carl A., and Farber, Eugene. "Driver Judgment in
Overtaking Situations.” Highway Research Record,
Number 247, Washington, D.C., 1968.

Snider, John N. and Ernst, Ronald L. "A Study of Driver
Variability in Car Following and Open Road Driv-
ing." Highway Research Record, Number 84. Pre-
sented at the 43rd Annual Meeting. January 13-17,
1964. Washington, D.C.

Spicer, Robert A. ”Human Factors in Traffic Accidents."
Research Grant AC-55, United States Public Health
Service. Department of Health, Education and Wel-
fare. Washington, D.C.: U.S. Government Printing
Office, 1963.

State Department of Motor Vehicles. "Project: Machine,

Examination, Teaching, Evaluation and Re-education.

An evaluation prepared under contract No. FH-ll-

6834 with the U.S. Department of Transportation,

Federal Highway Administration, Washington, D.C.,
1971.

Stephens, B.W., and Michaels, R. M. ”Time Sharing Between
Compensatory Tracking and Search-and-Recognition
Tasks." Highway Research Record Number 55. Pre-
sented at the 43rd Annual Meeting January 13-17,
1964, Washington, D.C.

Steet, R. L. and others. "Annotated Bibliography and Sum-
mary of Research Needs in the Human Factors As-
pects of Driver Visual Communications." Supported

by the Texas Transportation Institute. Texas A
and M University. July, 1970.

Todosiev, E. P. and Fenton, R. E. "Velocity Thresholds in
Car-Following at Night." Highway R.R., #122.

Presented 44 A.M. January 11-15, 1965. Washington,

D.C.

Twyford, Loran C. and others. Behavioral and Factual An-
alysis: A Paper on Research Grant supported by
the Public Health Service. New York State Educa-
tion Department.

Visual Perception and the Driving Task. Highway Users
Federation for Safety and Mobility. Washington,
D.C., 1970. l3pp.

 

 

132

Weir, David H. and McRuer, Duane T. "A Theory for Driver
Steering Control of Motor Vehicles." Highway Re-
search Record, Number 247. Washington, D.C., 1968.

Williams, Harriet G. "The Effects of Systematic Variation
of Speed and Direction of Object Flight and of
Skill and Age Classifications upon Visuo-perceptual
Judgments of Moving Objects in Three-Dimensional
Space." Toledo University, Ohio. Report No. BR-
6-8102. January 1968.

Wohlauer, Valentin E. and others. Proceedings of Fifteenth
Conference of the American Association for Auto-
motive Medicine. Society of Automotive Engineers,
Inc., New York, 1972.

Unpublished Materials

 

 

Adams, James R. ”Measurement of Hazard Judgment by a Stim-
ulus Accretion Technique." A Research Paper for
Safety Research and Education Project Teachers
College. Columbia University. February, 1968.

Barenklau, Keith Edward. "Improving Driver Recognition
of and Response to Collision Producing Situations
through Tachistoscopic Instruction.” Doctor's
dissertation, the University of Wisconsin, 1971.

Barry, Mathew Everett. "Development of Visual Perceptual
Capabilities Among Emotionally Disturbed/Socially
Maladjusted Students." Master's thesis, Illinois
State University, 1969.

Bhise, Virek Dattatray. "The Relationship of Eye Move-
ments and Perceptual Capabilities to Visual Infor-
mation Acquisition in Automobile Driving." Doc-
tor's dissertation, Ohio State University, 1971.

Burg, Albert. ”An Investigation of Some Relationships
Between Dynamic Visual Acuity, Static Visual
Acuity and Driving Record.” Doctor's disserta-
tion, University of California, Los Angeles, 1964.

Butler, Brian E. "The Role of Attention in Visual Percep-
tion.” Doctor's dissertation, University of
Waterloo, 1973.

Crabb, Owen L. ”A Driver Education Classroom Media Pack—
age to Develop Visual Perception.” Master's thesis,
Illinois State University, 1970.

133

Dorner, Edmund T. ”The Effects of Active Versus Passive
Traffic Simulation Instruction on Visual Percep-
tion." Master's thesis, Illinois State University,
1968.

Glassman, Jerold. ”The Effectiveness of a Teaching Machine-

Program as Compared with Traditional Instruction
in the Learning of Correct Responses to Hazardous
Driving Situations.” Doctor's dissertation, New
York University, 1965.

Goby, Dale W. ”The Relationship Between Visual Perceptual
Abilities and Behavioral Categories Among Beginning
Motorists.” Master's thesis, Illinois State Uni-
versity, 1970.

Hagen, Roger E. ”An Assessment of Sex Difference in the
Performance of a Simulated Driving Task." Doctor's
dissertation, University of South Dakota, 1971.

Hales, Ronald W. ”A Comparison of Two Visual Perceptual
Training Methods.” Master's thesis, Illinois State
University, 1969.

Herkowitz, Jacqueline. ”Filmed Test to Assess Elementary
School-aged Children's Perception to Embedded
Figures which Appear to Move Away from Stationary
Backgrounds." Doctor's dissertation, Purdue Uni—
versity, 1971.

Johnson, Duane Reed. ”A Case Study Analysis of Motorcycle
Accidents in Three Illinois Counties.” Doctor's
dissertation, Michigan State University, 1968.

Johnson, Ralph O. "A 'Static' Visual Perceptual Training
Program for Minority Ethnic Groups.” Master's
thesis, Illinois State University, 1970.

LaFond, Donald Louis. ”The Effect of Dynamic Visual Train-
ing on Manipulative Driving Skills.” Master's
thesis, Illinois State University, 1970.

Lawson, David Cadden. "Teaching Knowledge of Correct
Responses for Emergency Driving Situations through
the Use of an Intrinsil Type Program.” Doctor's
dissertation, West Virginia University, 1969.

Lazarewicz, Robert S. "The Relationship Between Behavioral
Characteristics and Visual Perception." Master's
thesis, Illinois State University, 1970.

 

 

134

Lucas, Richard Lawrence. ”Development and Evaluation of a
Part Task Film Simulation Technique for Training
Drivers on Critical Passing Skill." Doctor's dis-

sertation, University of South Dakota, 1970.

McDaniel, Charles E. ”A Study to Compare the Relative
Effectiveness of Selected Laboratory and Classroom
Programs in Driver and Traffic Safety Education.”
Doctor's dissertation, Michigan State University,

1970.
McPherson, Kenard. "A Systematic Analysis of Traffic
Education for Beginning Motorists.” Doctor's dis-

sertation, Michigan State University, 1970.

”Perception of Traffic Hazards A Comparative

 

Study.” Master's thesis, Illinois State Univer-
sity, 1966.

Poster, Dorothy C. "The Influence of Attention and Aware-
ness on Visual Perception.” Doctor's dissertation,

New School for Social Research, 1966.

Quane, William Laurance. ”The Relationship of Visual Per—
ceptual Capabilities as Measured by the Perception
of Traffic Hazards Test and Behavioral Categories
as Measured by the Mann Inventory.” Michigan
State University, 1970.

Rackoff, Nick Joseph. ”An Investigation of Age-Related
Changes in Driver's Visual Search Patterns and
Driving Performance and the Relation of Tests and
Basic Functional Capacities." Doctor's disserta-
tion, Ohio State University, 1974.

Robinson, Allen. "The Influence of Programmed Instruc-
tional Films on Perception of Traffic Hazards."
Master's thesis, Illinois State University, 1968.

Schapiro, Frank Hillel. "Information Extraction and Fil-
tering During Perceptual Learning in Visual Search.”
Doctor's dissertation, Cornell University, 1970.

Schiff, William. "The Perception of Impending Collision.”
Doctor's dissertation, Cornell University, 1964.

Streeter, Gerald M. "A Classroom Visual Perception Pro-
gram for Beginning Motorists." Master's thesis,

Illinois State University, 1968.

135

Thiffault, Charles. ”Tachistoscopic Training and Its
Effects Upon Visual Perceptual Speed of Ice-Hockey
Players.” Doctor's dissertation, University of
Southern California, 1974.

Thomas, James Albert. ”Decision Processes and Visual
Search Behavior in a Complex Perceptual Task."
Doctor's dissertation, The American University,
1966.

Thomas, Ronald L. "A Comparison of Two Types of Class-
room Presentations of Visual Stimuli Recognition
in Driver Education.” Master's thesis, Illinois
State University, 1970.

Tyson, Paul Dean. ”Perceptual Modification in the Visual
System.” Doctor's dissertation, University of
Waterloo, 1971.

Von Buseck, Calvin R. ”Augmented Feedback in a Simulated
Driving Task." Doctor's dissertation, the Univer-
sity of Michigan, 1965.

Weinstein, Sanford A. ”Recognition of Photographed Traf-
fic Hazards as an Indicator of Automobile Accident
Involvement.” Doctor's dissertation, Columbia
University, 1970.

Welch, Thomas C. "Visual Perceptual Training in Classroom
and Laboratory Instruction in Driver Education.”
Master's thesis, Illinois State University, 1969.

Wise, James Elton. "Development and Pretest of a Visual
Perceptual Screening Test for Use in Preschool
Education.” Doctor's dissertation, University of

South Carolina, 1973.

Wood, Patricia Ann. ”The Role of Age and School Experi-
ence in Performance on Certain Visual Perception
Tests.” Doctor's dissertation, University of
Illinois at Urbana-Champaign, 1973.

 

APPENDICES

136

APPENDIX A

PARTICIPANT INSTRUCTIONS AND OBJECTIVES
FOR THE SIMULATED DYNAMIC DRIVING
HAZARD PERCEPTION TEST

137

APPENDIX A

PARTICIPANT INSTRUCTIONS AND OBJECTIVES
FOR THE SIMULATED DYNAMIC DRIVING
HAZARD PERCEPTION TEST

INSTRUCTIONS

You are about to take part in an experimental
study, the results of which may someday play a small part
in improving traffic safety. When viewing the film, assume
you are behind the wheel of the car from which the picture
was taken. The film will be made up of a variety of driv-
ing situations lasting from only a few seconds to as long
as half a minute. In each situation you are to assume you
are the driver. If any hazardous or potentially hazardous
situation develops that, in your opinion, requires you to
take some action as a driver, such as decelerate, brake,
accelerate, steer left, steer right, et cetera, you are to
immediately stop the projector and tell the attendant what
hazardous or potentially hazardous situation caused you to
act. The attendant will tell you when it is all right to
start the projector again to see the remainder of that
sequence. Not all of the situations require action and in

other situations action on the part of the driver may be

138

 

139

questionable. Be sure to push the stop button as soon as
possible, but not before you can name the hazard.

The first sequence you will see is a sample so
that you might gain a full understanding of how the
sequences work. After viewing No. 1, you will hear on the
tape an explanation of when and why you should have reacted.

Please treat the sample as a regular sequence.

PARTICIPANT OBJECTIVES

As a participant of this study I will:

1. Identify hazardous or potentially hazardous
. situations when viewing the film.

2. Stop the projector when I would normally take
some action as a driver such as decelerate, brake, accel-
erate, steer left, steer right, etc.

3. Name hazard or hazards to attendant.

4. Wait for OK from attendant to continue viewing

film.

140

APPENDIX B

SCORE SHEET FOR THE SIMULATED DYNAMIC
DRIVING HAZARD PERCEPTION TEST

141

 

APPENDIX B

Name
Address

Years as Licensed Driver

 

 

Driver License No.
Date

 

 

Score Sheet

 

Simulated Dynamic Driving Hazard Perception Test conducted
under the auspices of the Highway Traffic Safety Center,
Michigan State University, East Lansing, Michigan.

Sequence Film Frame Counter Number

 

 

1.
2
3
4
5.
6
7
8
9

10.
11.
12.
13.
14.

15.
142

 

 

 

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

TillMilli11111111111Ilium

1 6 2512