differences in stress responses.
When a given individual does experience stress, it involves both an
emotional experience and a specific stress response. These two processes
are not as easily separable as the model would suggest. The distinction is
made, however, to emphasize the differences between the subjective
experience of stress and its more objectively quantifiable physiological
and psychological effects. The next stages of the model involve cognitive
and behavioral attempts to cope with the stressor. If such responses are
successful, the event is reevaluated as non-stressful and further coping
behaviors are unnecessary. This benign reappraisal may be the result of
cognitive restructuring or an objective change in the environment caused by
an overt behavioral coping response. If reappraisal processes still
indicate an imbalance, the individual continues to experience stress and
continues to make attempts at coping.
The overall form of the model is a continuous loop, and this form is
important in that it emphasizes the interaction between individual and
environment. This process is even clearer in the simplified transactional
model provided by McGrath (1976).

This model is shown in Figure 2 below.

Outcome
Process
A. Situation

I

7

Process

\

lntraorganism

Figure 2. A simplified transactional model (McGrath, 1976).

9

ENVIRONMENT
INDICTOR
Traflk

Sins
INDICTOR
KEYB ACCX
ACCY E L
ACCZ

II

SYSTEMS
RELATED
FUNCTIONS

Brclk SlWA

lnolc-rom
EDR HR EMQ

Figure 4. A model of driver, vehicle, and road system interactionsin
the driving siluation (Helander, 1975).

Perceptual complexity, the driver's interpretation of environmental
complexity, describes the total influence of the environment on the driver.
The level of perceptual complexity determines the activation level of the
driver.

Activation level can be measured through physiological correlates

such as heart rate and galvanic skin response.

Activation level, in turn,

affects performance in terms of vehicle control operations, which affect
the external environmental situation.
Neither of these two models directly deals with the idea of driving
stress, but it is apparent at what point such a concept would enter each
scheme.

In Getalter's model, the perceived situation is compared with

perceived coping possibilities to arrive at a subjective confidence level.
If the perceived situational demand exceeded coping abilities at this
comparison point, stress would result.

In Helander's model, perceptions of

environmental complexity influence the driver's activation level.

A high

activation level is a necessary prerequisite for the experience of stress.

accident group also exhibited a higher rate of psychopathology, including
alcoholism, paranoia, depression, and suicidal thoughts, than the control
group.

If the alcoholics in the fatality group are excluded from the

analysis, rates
. of psychopathology do not differ between the two groups,
but those drivers in the fatality group still reported significantly more
life stresses than the controls ( 4 2 % v s . 18%).

A slightly different sort of investigation was conducted by McMurray
(1970).

Accident and violation rates of individuals involved in divorce

proceedings were compared to average population levels.

Subjects were

grouped into four categories (male plaintiff, female plaintiff, male
defendant, and female defendant) for analysis.

During the seven year

.

period covered by the driving record, the accident rates for those people
involved in divorce proceedings were from 43 to 8 2 % higher than the average
rates.

Violation rates were 7 8 % to 195% higher.

Similar results were

found for the one year period associated with filing the divorce petition
(from six months before to six months after the filing of the document).
For all four of the groups studied, the period of greatest accident and
violation activity was the first three months after filing the divorce
petition (see Figure 6, below).

v

I

I

I

I

I

68
38
Months before

1618 12-16 9-12

The types of violations for which the

I

0-3

I

'

0-3
Time of filing

I

3-6

I

I

I

6-8 9-12 12-16 16-18
Months after

Figure 6. Percentage of drivers involved in accidents and violations
before and after filing for divorce (McMurray, 1970).

divorce group were cited differed from the average types of violations in
the population.

Fewer serious violations (those requiring mandatory

license suspensions) were found in the divorce group, but minor violations
involving speeding, failure to yield, and failure to stop occurred more
frequently in the divorce group than in the population at large.
These studies do suggest that driving stress, as evidenced by accident
and violation rates, is greater in individuals who have experienced recent
significant life stresses than in those who have not.

One group of drivers

in which this relationship may prove to be particularly strong is the
elderly driving population.

DRIVING STRESS AND THE ELDERLY
The driving population of the United States is aging.
11.3% of the population was over 65 years of age.

In 1988, only

This group is expected

to increase to 16.9% by the year 2000, and by 2030, over 20% of the
population will be over 65 (U.S. DOC., 1981, 1984).

Figure 7, below, shows

these changes graphically. The number of licensed drivers within the

AGE

Figure 7. Age d i i t f h t h of the U.S. population, 1950-2030
(Transportation Research Board, 1988).

The number of driver fatalities per million licensed drivers evidences a
similar increasing trend (Figure lo),
Driver fatalities per million licensed driven

500

0

10

20

30

40
50
60 70
Driver age, yean

80

90

100

Figure 10. Driver fatalities per million million licensed drivers,
1983 (Evans, 1988).

as does the number of driver fatalities per unit distance of travel (Figure

11).

, famlitbsper unit distance d

travel, ~ m ' '
1

F i n 11. Driver fatalities per terameter (1 Tm = 621 million miles)
of travel, 1983 (Evans, 1988).

Evans suggests that the increases in driver fatalities may, in fact, be due
to the fact that a crash of given serverity is more likely to prove fatal

14

CN
-

Ri

MAGNITUDE PER MIN. h9 U.S. 1
MAGNITUDE PER MIN. I TURNPIKE

A

mu

B

C

D

E

F

G

H

I A V G .

TEST DRIVERS

Figure 15. Mean tension generated on the Maine Turnpike and
US 1 for each test triver (Michaels, 1965).

Taylor (1964) conducted a study that compared percentage change in GSR
as drivers traversed a 21 mile route that contained 40 homogenous sections
of different road conditions, including winding country roads, urban
streets, and an expressway.

The GSR rate, in nominal response units per

minute, was, for the most part, quite stable across the different road
sections.

Some significant relationships, such as higher GSR rates for

road sections containing road junctions than for immediately adjacent
sections, were identified, but a constant GSR rate was maintained
throughout most conditions.

The researcher explains this by postulating

that GSR may function as a pacing mechanism.

A driver may have a certain

level of tension that he or she is willing to tolerate.

If road conditions

impose a greater or lesser amount of tension, the driver will adjust speed
accordingly.
C

a

t

e

m Leveb.

Berger, Bliersbach, and Dellen (1976) report

that levels of catecholanines indicate that greater stress was experienced
on city roads than on highways.
Road D e s m Element&
Certain road elements have been shown to have a direct relationship to
physiological indicators of stress.

For example, Babkov (1975) reports