Teaching Students to Assess Safety
for Crossing Streets Which Have
No Traffic Control
Dr. Mary-Maureen Snook Hill and Dona Sauerburger
1996 -- International Mobility Conference No. 8
Tambartun National Centre,
This paper will address the issue of crossing streets where there is no traffic control -- that is, where there is no control signal or stop sign for traffic on the street being crossed.
To cross these streets safely without
relying on drivers to stop and avoid you, there must be two conditions:
1. There
must be gaps between the vehicles that are long enough to allow you to
cross. In other words, the traffic must
not be too busy.
2. You
must be able to hear or see the traffic well enough to recognize when there is
nothing coming that could reach you during your crossing. In other words, if there were a vehicle
coming that could reach you before you finish your crossing, you would be able
to see or hear it approach. Under this
condition, whenever you hear or see nothing coming, it is clear enough to
complete a crossing.
Our students need to recognize those
situations in which one or both of these conditions are not present, because in
those situations there is considerable risk in crossing. That is, they need to know when the traffic
is too busy, and they also need to know when they can't see or hear the
approaching traffic well enough to know when it's clear to cross.
Seven years ago, the second author
developed techniques for assessing condition #2. One technique, the Timing Method for
Assessing the Speed and Distance of Vehicles (TMASD), is used to provide
students with feedback to assess and develop their ability to judge when the
traffic is far enough or slow enough to allow them to cross. This procedure is easy to use, and is
described in Sauerburger (1989 and 1995) but time
will not permit us to explain it in this session.
The other technique, called the Timing
Method for Assessing the Detection of Vehicles (TMAD), is designed to
assess whether or not a person can hear or see vehicles well enough to know
when it's clear to cross. The steps for
this method are listed at the end of the paper.
One question which needs to be answered is
how we can know whether we have observed one of the "worst
cars." That is, in a given
situation, after we have observed a number of approaching vehicles, we need to
know whether there could be any other vehicles in that situation which cannot
be heard or seen as well as the ones we already
observed and, if there are, whether those "worst cars" can still be
detected with enough warning. This is an
important issue because some people feel that it is not possible to ever
observe enough vehicles to draw conclusions about any situation (Bennett,
1991).
When Sauerburger
first developed the TMAD, she thought it could be used by blind people and
O&M instructors to assess the safety of crossing streets which are
questionable. However, we cannot judge
any street or intersection as being safe or unsafe to cross, because the situation
may change from day to day and even moment to moment. Sounds can be affected by snow, wet pavement,
the temperature of the air, masking noises and objects that block the sound,
head colds, clogged ears, hats, and so forth, and vision can be affected by
various lighting situations, such as glare, and light that is too bright or
dim. Because these various conditions
can affect the ability of students to detect the traffic, the students must be
able to assess the situation each time they consider crossing, rather than
assess particular intersections. For
example, if the student plans to cross a familiar street where he normally can
hear the traffic with enough warning but the sound of the traffic is now blocked
by a parked truck or masked by a distant lawnmower, the student must be able to
judge whether the traffic can still be heard well enough in those new
conditions.
Thus it is important that our students be
able to routinely and accurately recognize whether they can detect traffic well
enough to know when it's safe to cross.
A proposed procedure to provide feedback with the TMAD to assess and
improve the ability of students to make this judgment is outlined at the end of
the paper.
Although Sauerburger
(1995) had noted some success with using this procedure with the TMAD to
illustrate the safety issue to students at uncontrolled street crossings, it
remains to be determined whether the TMAD can provide accurate information
about the limit of the students' ability to detect vehicles, and whether use of
the TMAD increases the accuracy of students' judgments of their own ability to
detect traffic under any given conditions.
Thus a study was conducted in 1995 by the
Orientation and Mobility Department of Peabody College. One objective of this study was to determine
how one could ascertain the limit of students' ability to detect vehicles in a
given situation (that is how one could be relatively certain of having observed
and evaluated the students' detection of one of the "worst cars"). Another objective was to see if using the
TMAD to provide feedback to students would be effective in increasing their
ability to determine their relative safety in crossing streets.
|
Problem #1: In
order to use the Timing Method for Assessing the Detection of Vehicles
(TMAD), it is necessary to observe (and analyze the detection of) at least
one of the "worst cars' (which are those vehicles which are most
difficult to detect, and which reach you in the shortest time once you
detected them). It is only by making sure that you can detect even the
"worst cars'' well enough that you can be confident of being able to
detect all the vehicles sufficiently. Thus we need to know how to be
relatively certain that we have observed one of the "worst cars"
for the situation that we are analyzing. |
To determine how we can be relatively certain we
observe the detection of one of the "worst cars" when using the Timing
Method for Assessing the Detection of Traffic.
Six subjects who were O&M specialists
each used the TMAD to analyze their detection of 20 vehicles under four
different conditions. The hearing and
vision of the subjects was normal. In
about two thirds of these conditions the subjects used only their hearing to
detect the cars, and in some of these cases the conditions were varied by
providing a steady masking sound or wearing ear muffs. In about a third of the cases the subjects
used their vision.
This made a total of 24 conditions, in each of which 20 vehicles were detected. The data was analyzed to determine which of the 20 vehicles in each condition were the "worst cars" for that condition (that is, those which reached the subject in the shortest time or within a second of the shortest time from when they were heard or seen).
The results showed that for many of the
conditions, there were more than two "worst cars," some having as
many as five examples of "worst cars" among the 20 vehicles that were
observed. We found that if we started
counting the passing vehicles at the beginning of each condition as well as
immediately after the occurrence of a "worst car," in 95% of the
instances one of the "worst cars" was observed within the next 12
vehicles that passed. Thus it was
decided that when using the TMAD to test students' ability to detect vehicles,
we could be 95% certain of observing one of the "worst cars" if we
observed 12 vehicles. Note that this
means that one needs to detect 12 vehicles from each
direction, because vehicles coming from one's left are being detected in
conditions that are different from those that are coming from the right.
It would be helpful if people could learn
to recognize one of the "worst cars" whenever one passes, because it
can be very time-consuming to observe the detection of 12 vehicles in each
direction and even then, we are only 95% sure that we have observed a
"worst car." If instructors or
students could reliably recognize when they have observed one of the worst
cars, then they would not need to observe 12 cars, they could instead observe
until one of the "worst cars" passed, whether it is the first car or
the 21st car.
To try to find out if people could
recognize which are the "worst cars," four of the researchers each
observed 12 cars to see if they could recognize when they had observed one of
the "worst cars." Two of the
researchers did so accurately, but the other two thought that one of the cars
was a "worst car" when, in fact, they later observed another car
which reached them one or two seconds sooner than what they thought had been a
"worst car."
Discussion:
We determined that we could be 95% sure of
observing a "worst car" by observing at least 12 vehicles. The question that remains is whether 95%
certainty is sufficient to draw any conclusions about whether the vehicles can
be detected adequately and, if not, how much certainty is required.
Because it would be much easier to use the
TMAD if we could recognize which vehicles are the "worst cars," we
addressed the question of whether it is possible for people to be able to do
so. We found that when testing in one
condition with no practice, several people did identify the worst cars, but
whether they could do so repeatedly and in other conditions remains to be
seen. We also don't know whether the two
who failed to identify the worst cars could learn to do so with practice and
feedback. More research is needed to
answer this question.
|
Problem #2: Although
Sauerburger (1995) had noted some success with
using the Procedure to Develop Judgment of the Detection of Traffic with
students at uncontrolled street crossings, it remains to be determined
whether use of the TMAD increases the accuracy of students' judgments of
their own ability to detect traffic under any given conditions. |
Objective #2: To determine whether the "Procedure to Develop Judgment of the Detection of Traffic" can help students improve their ability to determine their relative safety in crossing streets.
The Procedure to Develop Judgment of the Detection of
Traffic was used with seven subjects at four uncontrolled
crossings. One of the four crossings was
at a street with moderate-to-high levels of traffic, while others were at side
streets with low levels of traffic. Some
of the streets had hills which partially blocked traffic noise, and one had a
constant source of traffic noise from a nearby street.
Subjects ranged in age from 16 to 20 years
of age. Two of them had functional
residual vision and five did not. None
had additional disabilities.
Four of the subjects had the opportunity to
judge their ability to detect vehicles in only one situation. That situation was one in which the TMAD
revealed that they could not hear the approaching cars well enough to know when
it was clear. Before testing with the
TMAD, two of these subjects recognized that they couldn't hear the cars well
enough there, and two failed to recognize it (they thought it would be safe to
cross whenever they heard nothing coming).
The other three subjects had the
opportunity to judge their abilities in more than one situation. One of those subjects judged his ability to
detect vehicles in 8 different situations.
He had functional vision, and judged in each of 8 situations that he could
see the cars well enough to know that when he saw nothing coming, it would be
clear to cross. The TMAD showed that he
was correct each time. However, he had
no opportunity to see if he could recognize situations in which he could not
see the traffic well enough to know when it was clear to cross.
The other two subjects each had at least
one situation where they could and one where they could not detect the cars
sufficiently. Both of them correctly
recognized the situation where they could detect the traffic well enough, but
both failed to recognize the situations where they could not. Instead, they thought that in every situation
to which they were exposed, it would be safe to cross whenever they heard
nothing coming.
Only one of those two subjects was exposed
to more than one situation in which he could not hear the cars well
enough. However he failed to recognize
the second situation as being risky, even after having had feedback and an
opportunity to learn from the first situation.
Discussion:
No definite conclusions can be postulated
within this study regarding whether any subjects actually improved his or her
performance, because time constraints prevented enough conditions being run
with each subject to establish good judgment or a pattern of judgment. Good judgment can be characterized by five
consecutive correct assessments of different situations, including several in
which the subject is unable to detect the traffic with sufficient warning, as
well as several in which the vehicles can be detected well enough to know when
it is clear enough to cross. Only one
subject had an opportunity to learn from his experience and improve his
judgment, but he was exposed to only two situations in which he could not
detect traffic sufficiently.
We speculated why that subject did not
recognize that the second situation was unsafe when he had received feedback
from the first unsafe situation. One
factor which we think could have affected it is that he was not asked to judge
whether he could hear the cars well enough; instead, he was asked to judge whether
it would be "safe" to cross when he heard nothing coming. He may have been reporting whether he thought
it was likely that he'd get hit if he crossed when it was quiet, not whether he
could hear the cars well enough to know when it was clear to cross.
Thus we suggest that when using this
procedure, the instructor be very precise when asking students to judge the
situation. The students must realize
that they are being asked to judge their ability to hear or see the traffic,
and thus ambiguous words that are not well defined, such as "safe,"
should be avoided. For example, rather
than say "Would it be safe for you to cross here when it is quiet?"
the instructor should say something like, "Do you think that you can
hear/see the traffic here well enough that you would know whether any vehicle
is coming too close or too fast for you to start your crossing? Or do you think that vehicles here are
appearing without enough warning, so that when you step out to cross, a vehicle
might be coming that would have to slow down to avoid you?"
Future
Directions:
Given the constraints inherent in this
study, one can only speculate whether the TMAD accurately measures students'
ability to detect the vehicles, and whether the TMAD is effective in increasing
the ability of travelers with visual disabilities to determine the relative
safety in crossing streets without traffic controls. The intent of this study is to encourage both
researchers and O&M instructors to explore the efficacy of the TMAD, to
consider using the TMAD to help their students and clients analyze and
understand the risks involved in crossing these streets, to consider how one
could be certain of observing a given situation enough to know the risks of
crossing there, and to share their observations.
_______________________________________
Bennett,
J. (1991). The fallacy
of timing methods. RE:view. Vol. 23, no. 2, 75-79.
Sauerburger, D. (1989). To cross or not to cross: Objective
timing methods of assessing street crossings without traffic controls. RE:view. Vol. 21, no 3, pp. 153-161.
Sauerburger, D. (1995). Safety Awareness for Crossing Streets
with No Traffic Control. Journal of Visual Impairment and Blindness. Vol. 89, No. 5, pp. 423-431.
Subjects 1-4 judged their ability to
detect vehicles in only one situation, which the TMAD revealed was one in which they could
not hear the approaching cars well enough to know when it was clear. Before testing with the TMAD, subjects 1 and
2 recognized that they couldn't hear the cars well enough there, but subjects 3
and 4 thought it would be safe to cross whenever they heard nothing coming.
Subjects 5 and 6 each correctly recognized
the situations where they could detect the traffic well enough, but both failed
to recognize the situations where they could not (subject #6 failed to
recognize that the second situation was risky even after he had an opportunity
to learn from the first trial). Instead,
both subjects thought that in every situation to which they were exposed, it
would be safe to cross whenever they heard or saw nothing coming.
Subject #7 had functional vision, and
correctly judged in each of 8 situations that he could see the cars well enough
to know that when he saw nothing coming, it would be clear to cross. He had no opportunity, however, to judge
situations in which he could not see the traffic sufficiently.
|
Subject
Number / age
/ gender |
visual
acuity |
trial |
conditions:
can/cannot detect vehicles |
Recognized situation correctly? Yes
/ no |
|
#1: 20 yrs -- F |
LP |
1 |
Cannot |
yes |
|
#2: 16 yrs -- M |
NLP
OD HM/LP
OS |
1 |
Cannot |
yes |
|
#3: 19 yrs -- M |
NLP
OU |
1 |
Cannot |
no |
|
#4: 18 yrs -- M |
NLP
OU |
1 |
Cannot |
no |
|
#5: 19 yrs -- M |
HM
OD NLP
OS (20/200
till 15 years old) |
1 |
Can |
yes |
|
|
|
2 |
Can |
yes |
|
|
|
3 |
Cannot |
no |
|
#6: 18 yrs -- M |
20/200
OD NLP
OS (used
hearing for trials) |
1 |
Cannot |
no |
|
|
|
2 |
Cannot |
no |
|
|
|
3 |
Can |
yes |
|
|
|
4 |
Can |
yes |
|
#7: 18 yrs -- M |
CF
OD HM
OS (used
telescope for trials) |
1 |
Can |
yes |
|
|
|
2 |
Can |
yes |
|
|
|
3 |
Can |
yes |
|
|
|
4 |
Can |
yes |
|
|
|
5 |
Can |
yes |
|
|
|
6 |
Can |
yes |
|
|
|
7 |
Can |
yes |
|
|
|
8 |
Can |
yes |
CF:
counts fingers;
HM: sees hand
movement; LP: light perception;
NLP: no light perception; OU: both eyes; OS:
left eye; OD: right eye