Accident Analysis and Prevention 37 (2005) 962–971

Crossing roads safely: An experimental study of age differences

in gap selection by pedestrians
Jennifer A. Oxley a,∗ , Elfriede Ihsen b , Brian N. Fildes a , Judith L. Charlton a , Ross H. Day c
a Accident Research Centre, Monash University, Clayton, Vic., Australia
b Department of Psychology, Swinburne University of Technology, Lilydale, Vic., Australia
c Department of Psychology, LaTrobe University, Bundoora, Vic., Australia

Received 21 March 2005; accepted 30 April 2005

Abstract

The over-representation of older pedestrians in serious injury and fatal crashes compared to younger adults may be due, in part, to age-related
diminished ability to select gaps in oncoming traffic for safe road-crossing. Two experiments are described that examine age differences in gap
selection decisions in a simulated road-crossing environment. Three groups of participants were tested, younger (30–45 years), young–old
(60–69 years) and old–old (>75 years). The results showed that, for all age groups, gap selection was primarily based on vehicle distance and
less so on time-of-arrival. Despite the apparent ability to process the distance and speed of oncoming traffic when given enough time to do so,
many of the old–old adults appeared to select insufficiently large gaps. These results are discussed in terms of age-related physical, perceptual
and cognitive limitations and the ability to compensate for these limitations. Practical implications for road safety countermeasures are also
highlighted, particularly the provision of safe road environments and development of behavioural and training packages.
© 2005 Elsevier Ltd. All rights reserved.

Keywords: Ageing; Pedestrian; Crash; Behaviour; Crash risk

1. Introduction Some studies point to a link between age-related declines

Making decisions about when it is safe to cross roads
in relation to available traffic gaps is a complex everyday
task, implicating sensory, perceptual, cognitive and executive
functions. This may be especially difficult for older pedes-
trians, given the evidence demonstrating age-related changes
in performance in these areas (Corso, 1981; Korteling, 1994;
McDowd and Shaw, 2000; Roeneker et al., 2003; Salthouse,
1991). Non-optimal road-crossing behaviour may then, in
part, explain the high rate of fatal and serious injury crashes
found for older pedestrians (Australian Transport Safety
Bureau, 2002; Commission of the European Communities
(CEC), 2000; Mitchell, 2000; National Highway Traffic
Safety Administration (NHTSA), 2001; Organisation for
Economic Co-operation and Development (OECD), 2001).
∗ Corresponding author. Tel.: +61 3 9905 4374; fax: +61 3 9905 4363.
E-mail address: Jennie.Oxley@general.monash.edu.au (J.A. Oxley).
in driving and road-crossing skill and increased crash risk
(Carthy et al., 1995; Helmers et al., 2004; Mathey, 1983;
Oxley et al., 1997a; Oxley, 2000), and suggest that this may,
in part, be caused by increasing difficulties in judging safe
gaps in the traffic especially under complex conditions, such
as two-directional traffic or when a decision needs to be made
quickly. Gap selection difficulties in older people may in turn
be due to age-related declines in perceiving objects in motion
(Carthy et al., 1995; Cavallo and Laurent, 1988; De Lucia et
al., 2003; Staplin and Lyles, 1991), estimating the time-of-
arrival of an oncoming vehicle at the crossing point (Schiff
et al., 1992; Scialfa et al., 1991) and/or compensating ade-
quately for age-related declines in walking speed (Lee et
al., 1984; Yanik and Monforton, 1991). Clearly, the abili-
ties to perceive and integrate speed and distance information
of approaching vehicles accurately and to select a gap large
enough, taking account of any physical limitations, are cru-
cial components of safe road-crossing. The two experiments
reported here investigated age differences in the ability to

0001-4575/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aap.2005.04.017
 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971
choose safe time gaps in traffic in a simulated road-crossing
task as well as some of the factors involved in such judge-
ments.
2. Experiment one
The first experiment investigated age differences in gap
selection during road-crossing tasks in a simulated traffic
environment in which time gap and vehicle speed were sys-
tematically varied. On the basis of findings of on-road obser-
vations (Oxley, 2000; Oxley et al., 1997a,b) which showed
that some of the oldest pedestrians left larger distance gaps
than their younger counterparts, but not longer time gaps and
consequently made risky crossing decisions, it was expected
that the road-crossing responses of older adults would be
based on distance, not time gap, and that they would leave
smaller safety margins than younger adults.
2.1. Method
2.1.1. Participants
There were three groups of participants: 18 younger adults
aged between 30 and 45 years, 18 young–old adults aged
between 60 and 69 years, and 18 old–old adults aged 75 years
and over. All participants were volunteers, licensed drivers
and in good health. Assessment of functional performance
revealed group differences, with the oldest group performing
more poorly than younger adults on cognitive, perceptual,
visual and physical measures, but all performances were
within the normal range for the relevant age groups. In partic-
ular, no significant peripheral visual field losses were detected
in the older participants. Details of the outcomes of these
assessments and their relationship to road-crossing decisions
are reported elsewhere (Oxley et al., 2001).
2.1.2. Stimulus materials
A simulated road environment of moving traffic generated
from data files from a mid-range driving simulator was used
in these experiments (Fig. 1). It showed an undivided, straight
two-way residential road (with visual and audio features to
make the environment as realistic as possible) from the per-
spective of a pedestrian waiting at the kerb on a mid-block
Fig. 1. Stimulus traffic scenarios presented in the road-crossing simulation
(Note: the road appeared straight when projected onto a curved presentation
screen).
963
road section, with two approaching vehicles travelling from
the right-hand side (near-side lane), which is the lane clos-
est to the pedestrian in Australia. There was no traffic in the
far-side lane.
Time gap and speed of the vehicles were manipulated with
five levels of time gap (1, 4, 7, 10 and 13 s) and three levels
of vehicle speed (40, 60 and 80 k/h) resulting in 15 different
traffic scenarios. Distance co-varied as a function of these
two manipulations. Levels of time gap were chosen based
on group average walking speeds in observational studies
(Oxley, 2000; Oxley et al., 1997a), and represented theoret-
ically safe and unsafe time gaps. Simulated traffic scenes
were downloaded onto three video tapes, each containing the
scenes in a different random order. Images were projected
onto a large white curved screen measuring approximately
2.5 m in width and 1.5 m in height. The size of projection
was roughly equivalent to the visual angle of vehicles when
viewed from the side of the road. Participants were able to
scan the traffic scene freely, hence any undetected periph-
eral field loss would not have affected judgements. Each
participant viewed the traffic scenes on each video tape and
presentation was counterbalanced. Thus, they viewed a total
of 45 traffic scenes.
Responses were made on a computer keyboard on the desk
in front of participants. Most of the keys were blackened and
covered. Two keys (‘D’ and ‘J’) labelled ‘YES’ and ‘NO’
were available for participants to indicate whether they would
‘cross’ the road or not. The keys for numbers 1–9, with labels
‘very unsafe’ below the 1 key and ‘very safe’ below the 9 key,
provided a nominal rating scale on which participants were
asked to rate their safety of the road-crossing.
2.1.3. Procedure
Participants were seated at a desk in a darkened, quiet room
2 m in front of the projection screen with their right index fin-
ger resting on the ‘YES’ key and their left index finger on the
‘NO’ key and looking down towards the keyboard. A buzzer
sounded as the first ‘trigger’ vehicle passed the point of cross-
ing and at the same time activated a timer. Participants were
instructed to look at the traffic scene as soon as they heard the
buzzer and to decide whether or not they would ‘cross’ in front
of the second vehicle (walking normally across the street),
responding as quickly as possible by pressing the ‘YES’ or the
‘NO’ key. This deactivated the timer and the time interval was
recorded as decision time. After this, participants were asked
to rate how safe or unsafe they thought the ‘crossing’ would
have been by pressing the appropriate key (1–9). No time lim-
its were imposed for this response. Practice trials were given
until participants indicated that they fully understood the
task.
Walking time over a distance equivalent to the width of
an average road lane (5.6 m) at two walking paces (normal
walking pace and fast walking pace, i.e., walking as fast as
they could) was also measured.
Total testing time took approximately 60 min (with small
breaks for older participants, if required).
964 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971

Table 1
Mean walking times (normal and fast-paced) and decision times by age group (with standard deviation)
Normal-paced walking time (s) Fast-paced walking time (s) Decision time (s)
Mean (S.D.) Minimum Maximum Mean (S.D.) Minimum Maximum Mean (S.D.) Minimum Maximum
Young group (n = 18) 3.87 (0.31) 3.22 4.38 2.72 (0.32) 2.07 3.13 0.66 (0.20) 0.18 1.27
Young–old group (n = 18) 4.60 (0.61) 3.69 5.78 3.57 (0.55) 2.75 4.72 0.88 (0.15) 0.15 3.59
Old–old group (n = 18) 6.96 (1.57) 4.56 9.63 5.74 (1.72) 3.85 9.25 1.45 (0.42) 0.23 8.01

2.2. Results (p = 0.07) but significantly faster than old–old participants,

Five performance measures were analyzed. These were
walking times, decision times, yes/no responses, safety mar-
gins and safety rating responses. Safety rating responses will
not be presented here, as the pattern of these results closely
resembled that of the yes/no responses.
2.2.1. Walking times and decision times
Walking times and decision times were analyzed by
ANOVA and the effects were explored by post hoc Tukey
tests. Table 1 shows the means, standard deviations and range
(minimum and maximum) of walking times and decision
times of the three groups. A significant effect of age group was
found for normal walking time, F(2,51) = 48.17, p < 0.001.
Post hoc Tukey tests revealed that the normal walking times
of young and young–old participants were similar (p = 0.08),
but the old–old group walked more slowly than both younger
groups, p’s < 0.001. A significant effect of age was also found
for fast walking pace, F(2,51) = 37.93, p < 0.001, with the
young and young–old participants walking at a similar pace
p’s < 0.001.
A significant effect of age group was found for deci-
sion time, F(2,51) = 38.04, p < 0.001. The old–old group took
more time to make their decisions than both younger groups,
p’s < 0.001, who did not differ from each other (p = 0.6).
2.2.2. Yes/no responses
Analyses of yes/no responses were undertaken by employ-
ing hierarchical logistic regression modelling of the data to
examine the impact on the crossing decision of the variables
age group, time gap, vehicle speed, distance gap and walking
time. Walking time was adjusted for in the analysis so that
the other variables were free of this factor. Because distance
gap co-varies in a systematic fashion when vehicle speed and
time gap are manipulated (refer to Fig. 2), it was necessary
to undertake two separate analyses. Model 1 included walk-
ing time, age group, time gap and vehicle speed as variables,
and Model 2 included walking time, age group and distance
gap. Model 1 revealed that time gap was a strong predictor of
crossing decisions, χ2 (4) = 191.33, p < 0.001, R = 0.21. Age

Fig. 2. Proportion of actual (connected data points) and predicted (unconnected data points) yes responses as a function of age group, distance gap, vehicle
speed and time gap.
 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971
group, vehicle speed and walking time also predicted crossing
decisions, χ2 (2) = 153.54, p < 0.001, R = 0.20, χ2 (2) = 90.76,
p < 0.001, R = 0.16 and χ2 (1) = 32.33, p < 0.001, R = 0.10,
respectively. A significant time gap by age group interaction
was also found, χ2 (8) = 152.53, p < 0.001, R = 0.21. No other
interaction was significant in this model. Model 2 further
revealed that distance gap was a strong predictor of cross-
ing decisions, χ2 (14) = 426.27, p < 0.001, R = 0.36. As in
Model 1, age group and walking time predicted crossing deci-
sions, χ2 (2) = 152.26, p < 0.001, R = 0.21 and χ2 (1) = 36.25,
p < 0.001, R = 0.10, respectively. The interaction between age
group and distance gap was not significant. These findings
indicate that responses in general were primarily based on
the distance of the vehicle and to a much lesser extent on
time gap, and that groups responded differently to increases
in time gaps.
Fig. 2 shows the proportion of positive crossing responses
by vehicle conditions for each age group. These data show
that all participants were less likely to indicate that they
would cross when time and distance gaps were small than
when these were larger. At distances of 22 m and below
(time gap of 1 s) the response rate was close to zero for all
three groups. Beyond that, the proportion of yes responses
increased rapidly for the young participants who reached
close to 100% asymptote when the vehicle was more than
100 m or 7 s away from them. Response rates for the other
two groups increased more gradually and they reached close
to 100% asymptote later (young–old group at about 150 m or
10 s; old–old group at about 200 m or 13 s). It is noteworthy
that between 10 and 40% of old–old participants decided to
cross when time gaps were only 4 s (depending on vehicle
distance) even though most took longer to walk the distance
of a carriageway, even at their fastest pace.
Fig. 2 also illustrates the findings of regression analyses
that distance, not time gap, was the main determinant of cross-
ing decisions for all groups. Vehicle speed was also taken into
account, but to a lesser extent. For instance, for the three time
gap conditions of 4 s, the proportion of positive responses
of all groups increased as distance gap increased. This also
occurred for the 7 and 10 s time gap conditions for the two
older groups (the young group was already at or close to 100%
asymptote). Time gap also seemed to contribute to crossing
decisions. In the 88.9 m/4 s and 77.7 m/7 s conditions, partic-
ipants in all groups were more likely to respond positively in
the 7 s condition even though the corresponding distance was
smaller than that in the 4 s condition. If decisions had been
made solely on the basis of distance, the response rate should
have decreased for the 77.7 m/7 s gap.
An additional analysis was carried out to examine whether
any adjustment in response by older participants was large
enough to compensate for increased walking times. Actual
proportions of yes responses of young–old and old–old par-
ticipants in critical time gaps (4 and 7 s) were compared
with those predicted from the mean of individual normal and
fast walking times (shown in Fig. 2 as unconnected/unfilled
data points). Chi-square analysis revealed significant differ-
965
ences between actual and predicted responses, p’s < 0.01. In
time gaps of 4 s, it was predicted that 44% of young–old
participants could have responded yes, but that no old–old
participant should have done so. In the time gap of 7 s, all
young–old participants, but only 61% of the old–old partic-
ipants could have responded yes. Actual response rates of
the young–old participants were lower than predicted for all
critical time gaps, indicating that their decisions were overly
conservative. In contrast, actual response rates of the old–old
participants were generally higher than they should have been
(Chi-square p’s < 0.05). There was one exception from this
general trend: In the 7 s time gap condition, where vehicle
distance was close, fewer old–old participants than expected
indicated that they would have crossed.
2.2.3. Safety margins
Safety margins were defined as the sum of the mean walk-
ing time (i.e., the average of normal and fast walking times)
plus decision time, being subtracted from the time gap of the
approaching vehicle for each trial in which a yes crossing
response was made. This measure, calculated for each indi-
vidual separately for every trial, therefore, represented the
time interval between the participant reaching the centre of
the road and the vehicle arriving at the crossing line. Individ-
ual safety margins were then plotted against individual mean
walking times for the critical time gaps of 4 and 7 s and sub-
jected to regression analysis. These distributions are shown
separately for age groups in Fig. 3. Also shown on each dis-
tribution is the zero-line that denotes where safety margins
were at the zero level. Any points below this line indicate an
unsafe response, while points above the line indicate a safe
response.
For all age groups, the safety margin decreased with
increasing walking plus decision times and this was partic-
ularly so for the oldest group. The distribution of the young
participants (Fig. 3a) shows little variability and relatively
safe crossing decisions. Only 19% of young participants
indicated that they would have crossed in an unsafe man-
ner. Moreover, the unsafe crossings for this group were only
marginally unsafe, generally falling between 0 and −1 s of
safety margin. The distributions of the older groups (Fig. 3b
and c) showed more variability, especially that of the oldest
group. While only 18% of the young–old participants indi-
cated unsafe crossings, a much larger proportion (70%) of
the old–old participants did so. The unsafe responses of the
young–old group fell between 0 and −2 s of safety margin. In
contrast, the unsafe responses of the old–old group extended
to −10 s. This confirms that many of the old–old participants
who decided to cross with a 4 and 7 s time gap should have
said ‘no’.
3. Experiment two
In the first experiment, participants were asked to make
decisions under time pressure and consequently mean deci-
966 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971
Fig. 3. Distribution of safety margins for all positive crossing decisions at 4
and 7 s time gaps plotted as a function of walking time.
sion times were much lower than in the on-road observational
studies (close to six times shorter for all age groups). It is
possible that under time constraints, decisions are made pri-
marily on the most immediately available, and hence, most
easily accessible information which in this case is the distance
of the vehicle. Judgements about speed require information
about vehicles over time and this may require longer time to
process with advancing age. It is, therefore, possible that, if
given more time to inspect the stimulus display and to make
a crossing decision, participants are more likely to base their
decisions on the time gap (integrating distance and speed)
rather than the distance alone. This may be especially impor-
tant for the oldest participants, leading to less risky crossing
decisions. These possibilities were explored in the second
experiment by examining the effect of different presentation
times of virtual traffic scenes on the ability to judge safe gaps
in traffic.
3.1. Method
3.1.1. Participants
Forty-five participants took part in this experiment con-
sisting of 15 young adults aged between 30 and 45 years,
15 young–old adults age between 60 and 69 years and 15
old–old adults aged 75 years and over. All participants had
taken part in the first experiment and were familiar with the
task.
3.1.2. Materials and procedure
The apparatus used in this experiment was identical to that
in Experiment 1. A series of 12 traffic scenes were selected
from the stimulus films shown in Experiment 1 with manip-
ulations of vehicle speed, time gap and presentation time.
Three levels of time gap (4, 7 and 10 s), two levels of vehicle
speed (40 and 80 k/h), and two levels of presentation time (1
and 5 s) were used. Presentation times were chosen based on
mean decision and traffic inspection times across all groups
in Experiment 1 and in observational studies (Oxley, 2000;
Oxley et al., 1997a).
All traffic scenes were downloaded from the mid-level
driving simulator onto VCR tapes and edited to produce views
of traffic beginning with the first ‘trigger’ vehicle passing
the pedestrian crossing point and ending after either 1 or 5 s
presentation time. Presentation of traffic sequences was in
random order, participants viewed each scene twice (a total
of 24 traffic scenes were viewed), and they were instructed
to make yes/no and safety rating responses as in the previous
experiment. Again, practice trials were presented until partic-
ipants felt comfortable with the task. The results for the safety
ratings will not be presented here as, like in Experiment 1,
they closely resembled those of the yes/no responses.
3.2. Results
3.2.1. Yes/no responses
Hierarchical modelling of the data was employed to exam-
ine the effect of independent variables on yes/no responses.
Walking time was held constant while the impact of age
group, time gap, vehicle speed, distance and presentation
time were examined in two separate analyses. Model 1
included age group, time gap, vehicle speed and presen-
tation time. This analysis revealed that age group, vehicle
speed and time gap were significant predictors of crossing
decisions: age group, χ2 (2) = 51.17, p < 0.001, R = 0.30, time
gap, χ2 (2) = 131.45, p < 0.001, R = 0.30, and vehicle speed,
χ2 (1) = 102.01, p < 0.001, R = 0.27. Model 2 included age
group, vehicle distance and presentation time and revealed
that distance gap was a significant predictor of crossing deci-
 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971 967

sions, χ2 (5) = 56.06, p < 0.001, R = 0.18. It is noteworthy that

the effect of vehicle distance was markedly reduced in this
experiment compared with the first one, and that, here, time
gap and vehicle speed exerted stronger effects on the crossing
decisions than vehicle distance.
Presentation time by itself did not significantly affect
crossing decision in either of these analyses, but it seemed
to moderate the effect of other variables. Significant inter-
actions were found in Model 1 between presentation time
and vehicle speed, χ2 (1) = 8.36, p < 0.01, R = 0.07, between
time gap and age group, χ2 (4) = 12.12, p < 0.05, R = 0.06,
and between presentation time, age group and vehicle speed,
χ2 (2) = 6.03, p < 0.05, R = 0.04. In Model 2, a significant
interaction was found between presentation time and dis-
tance gap, χ2 (5) = 13.96, p < 0.05, R = 0.05. The three-way
interaction between age group, vehicle distance and presen-
tation time was not significant in this analysis, however, the
observed group differences in the other interactions can be
attributed to age effects on road-crossing decisions based on
distance and speed of approaching vehicles. These interac-
tions imply a base effect of presentation time.
Fig. 4 shows the proportion of yes responses by vehicle
conditions separately for the two presentation times and the
three groups. As in Experiment 1, response rates increased
with increases in time gap. For the younger group, this
increase was steeper than for the older groups and they
reached 100% asymptote at about the same points as in Exper-
iment 1, namely at 100 m or 7 s (Fig. 4a). Presentation time
made little difference to the proportion of yes responses of the
young group. In both presentation time conditions, distance,
and to a lesser extent, time gap seemed to influence their deci-
sion to cross. That is, even at short presentation times, young
participants were more likely to cross when distances were
longer even though time gaps did not vary, but they also said
‘yes’ more often when the reverse was true (see increase in
response rate from the 88.9 m/4 s to the 77.7 m/7 s condition).
Thus, younger adults were able to process both distance and
speed of vehicles in very short periods of time, even though
they primarily based their crossing decisions on vehicle dis-
tance.
Young–old participants, too, were able to process both
distance and speed of vehicles but apparently only at longer Fig. 4. Proportion of actual (connected data points) and predicted (uncon-
presentation times. In this condition, the proportion of yes nected data points) yes responses as a function of age group, distance gap,
responses increased as a function of increases in both time vehicle speed, time gap and presentation time.
and distance gaps and were similar to those found in Experi-
ment 1. When presentation times were 1 s, however, vehicle speed of vehicles under time constraints. Vehicle distance
distance alone seemed to be the determinant of crossing deci- alone determined crossing decisions of this group in the 1 s
sions. The proportion of yes responses was similar when presentation time condition. At longer presentation times, the
distances were similar even though time gaps increased, e.g., responses of the oldest participants were clearly more influ-
at 88.9 m/4 s and 77.7 m/7 s (before reaching asymptote at enced by both the time gap and distance (i.e., vehicle speed
111 m or 7 s). Thus, young–old adults seemed to experience was also taken into account). For example, the response rate
some difficulty processing both distance and speed of vehi- increased when the distance increased from 44.4 to 88.9 m
cles under extreme time constraints. and from 77 to 155.6 m, and vehicle speed increased from
The trend in these data for the old–old group (Fig. 4c) 40 to 80 k/h, despite time gaps being constant at 4 and 7 s,
was similar to that of the young–old group, only more pro- respectively. However, response rates also increased when
nounced, indicating major difficulty processing distance and time gaps increased from 4 to 7 s and stayed at asymptote
968 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971

with a change from 7 to 10 s, despite a decrease in distance
and vehicle speed. Interestingly, in the 5 s presentation time
conditions, asymptote was reached at the same point as for
the other two groups and earlier than in the 1 s presentation
time condition and in Experiment 1.
A comparison of the actual and predicted response rates
for critical time periods of 4 and 7 s (predictions were based
on walking times and are shown as unconnected data points
in Fig. 4b and c) suggests that, as in Experiment 1, young–old
participants responded conservatively for all critical time and
distance gaps under both presentation conditions. Proportions
of yes responses by old–old participants were generally above
predicted levels, for larger distances and fast vehicle speed,
regardless of presentation time. It seems, therefore, that even
when old–old adults had sufficient time to process the time
gap of oncoming vehicles, many of them still made unsafe
crossing decisions based on vehicle distance.
3.2.2. Safety margins
The analysis of safety margins (Fig. 5) shows that, as
in Experiment 1, only a small proportion of young and
young–old participants decided to cross with insufficient
intervals between them and the vehicle arriving at the cross-
ing line. In the 1 s presentation time condition only 14.5% of

Fig. 5. Distribution of safety margins for all positive crossing decisions at 4 and 7 s time gaps plotted as a function of walking time.
 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971
the young group fell below the zero line, in the 5 s condition it
was 15.9%. The corresponding proportions for the young–old
group were 23.4 and 10.6%, respectively. However, just as in
Experiment 1, many of the oldest group who indicated that
they would have crossed did so when it was clearly not safe.
In the 1 s condition, 76.9% made unsafe decisions, compared
with 64.2% in the 5 s condition. These differences in propor-
tions as a function of presentation time were evaluated by
Chi-square and found not to be significant (p = 0.08 and 0.09
for young–old and old–old participants, respectively). This
confirms the earlier conclusion that the unsafe behaviour of a
large proportion of old–old participants was not affected by
the time they had to inspect the traffic scene.
4. Discussion
The current study confirmed the findings of the observa-
tional studies, showing that older adults made risky crossing
decisions (Oxley, 2000; Oxley et al., 1997a). A large propor-
tion of the oldest group (>75 years) decided that they could
cross the virtual road when in fact it was unsafe for them to do
so, given their walking times. Negative safety margins sug-
gest that in an on-road environment, drivers would have had
to take evasive action to avoid a collision. Increased presenta-
tion time of the traffic scene did not significantly change the
proportion of risky decision makers amongst old–old partic-
ipants. The young–old participants, on the other hand, made
relatively conservative decisions and very few of them left
unsafe safety margins. Indeed, many of them decided not to
cross, when their walking times indicated that they could have
done so safely.
The most surprising finding was that it was not only the
oldest participants, but also both younger groups who made
road-crossing decisions that were based primarily on the dis-
tance of oncoming vehicles and to a lesser extent on time
of arrival. Others, too, have found that younger drivers and
pedestrian use the same visual information when making gap
judgements (Connelly et al., 1998; De Lucia et al., 2003;
Simpson et al., 2003). Therefore, while in theory consider-
ation of both distance and speed of approaching vehicles is
highly pertinent to crossing roads, in practice this does not
seem to happen. There are several reasons why crossing deci-
sions could be primarily based on vehicle distance in these
experiments. It may be that the results are an artefact of the
impoverished two-dimensional viewing conditions employed
here. This is, however, an unlikely explanation. First, a val-
idation study (Oxley et al., 1997b) showed that crossing
decisions and perceptions of safety by younger and older
adults in real-world and filmed versions of traffic scenes were
highly correlated. Second, the results clearly show that, given
favourable exposure conditions, time gap was processed by
all groups, including the old–old group, however, this vari-
able did not seem to function as the primary determinant of
crossing decisions. It is possible that, on inspection of the
oncoming traffic, pedestrians would make an initial decision
969
such as ‘the further the car is away from me, the safer it is
to cross’. This finding suggests an immediacy effect where a
vehicle far away, irrespective of its travelling speed, is judged
to be less threatening than one close up.
While making crossing judgements primarily on vehicle
distance may be a strategy that works reasonably well for the
young and young–old participants, who can walk fast enough
to avoid even relatively close vehicles, it clearly leads to
many risky crossing decisions by slower old–old participants.
However, considering that all groups use the same informa-
tion on which to base their decisions, group differences in
the processing of traffic information cannot fully explain the
high proportion of risky behaviour by old–old participants
observed in previous studies (Carthy et al., 1995; Oxley et
al., 1997a) and in the experiments reported here.
There are some alternative explanations. One, it is possi-
ble that old–old participants resort more frequently than their
younger counterparts to making crossing decisions which
are based on vehicle distance, as such decisions often have
to be made quickly. The findings of Experiment 2 sug-
gest that under time-constrained conditions, old–old par-
ticipants seemed to have difficulties integrating speed and
distance information of approaching vehicles. When presen-
tation times were short, time gap had little effect on the
crossing decisions of this group. This may lead to danger-
ous road-crossing behaviour.
Interestingly, decision times of the oldest group were
longer than those of younger participants both in obser-
vational studies (Oxley, 2000; Oxley et al., 1997a) and in
Experiment 1. Notwithstanding, neither longer decision nor
presentation times seemed to lead to safer decisions, both
on the road and here under experimental conditions. This
indicates that opportunity to process speed and distance infor-
mation about oncoming vehicles may help, but does not
necessarily lead to a much higher proportion of safer deci-
sions by old–old adults. In contrast, the other two groups
made crossing decisions that were generally safe, despite all
indications that vehicle distance was the primary determinant
of these decisions and despite indications that the young–old
participants also had difficulties processing vehicle speed and
distance when under time constraints in Experiment 2.
This leads us to consider the second alternative expla-
nation. At all vehicle distances up to asymptote, old–old
participants were less likely to indicate that they would have
crossed than their young counterparts. This was appropri-
ate as their longer walking times demanded larger gaps and
suggests that they seem to take their slower walking speeds
into account when making road-crossing decisions. Despite
this ability, old–old participants were not able to compen-
sate adequately for their slower walking speed to make safe
decisions. This possibility was also considered by Lee et al.
(1984) who argued that, with diminished perceptual, cogni-
tive and motor abilities, some older adults may experience
difficulty in accommodating their judgments of safe gaps in
the traffic to slowed walking times. Sabey (1988) and Yanik
and Monforton (1991) also suggested that losses occur in
970 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971
the compensatory senses in the oldest old, and, as a result,
the potential for risk increases and adoption of compensatory
behaviours becomes more difficult. Furthermore, it is gener-
ally acknowledged that the impact of age-related losses on
performance may be under-estimated by the oldest old. For
example, Holland and Rabbitt (1992) argued that, because of
reduced information processing capacity, older adults may
be less efficient than younger adults at monitoring their own
performance, less aware of their mistakes, or less able to
remember making mistakes. We have also found evidence of
this in earlier work (Oxley et al., 2001).
More recent studies with older drivers rather than pedes-
trians (Daigneault et al., 2002) indicated that risky driving
behaviour and crash propensity may be related to age-related
declines in executive function as well as reduced processing
capacity. They suggested that, due to executive dysfunction,
at least some crash-involved older drivers do not seem to be
able to accurately estimate the risk of specific traffic situations
and/or compensate sufficiently for declines in sensory and
cognitive function. In the context of the present findings, it is
possible that losses in executive function in old–old pedestri-
ans leads to difficulties with choosing gaps sufficiently large
to compensate for their decline in walking speed, which in
turn leads to unsafe crossing decisions.
In contrast, the young–old participants made quite con-
servative and generally safe decisions despite their obvious
difficulties in processing both vehicle speed and distance
under short presentation time conditions. This suggests that
cognitive slowing has started to affect decision-making in
60–69 year olds but their behaviour also indicates that they
do not yet experience executive function deficits.
In conclusion, these findings show that pedestrians of all
ages seem to make road-crossing decisions based primar-
ily on the distance of the oncoming vehicle and less so
on time of arrival. Despite this apparently unsafe strategy,
young and young–old participants generally made safe cross-
ing decisions. However, a large proportion of the responses
of old–old participants indicated that they would have made
unsafe crossings in an on-road environment. These risky deci-
sions appeared to be not so much a function of cognitive
inability to integrate perceptual information about oncom-
ing traffic, as these participants were clearly able to process
both vehicle speed and distance given sufficient time, but
of declines in executive function, which may lead to mis-
perception of risk and/or difficulties with choosing gaps
sufficiently large to compensate for their decline in walking
speed.
These findings have practical implications for road safety
countermeasures aimed at improving the safety of older
pedestrians. An emphasis on designing safe road environ-
ments, such as vehicle speed restrictions, to facilitate older
pedestrians’ mobility and safety is clearly warranted. It
would also be worthwhile to consider the development of
behavioural and training packages aimed at older pedestrians
to assist their awareness of declining abilities and adoption
of safe road-crossing practices.
References
Australian Transport Safety Bureau (ATSB), 2002. Pedestrian Facts.
ATSB, Canberra.
Carthy, T., Packham, D., Salter, D., Silcock, D., 1995. Risk and Safety on
the Roads: The Older Pedestrian. University of Newcastle Upon Tyne:
Report Prepared for the AA Foundation for Road Safety Research.
Cavallo, V., Laurent, M., 1988. Visual information and skill level in time-
to-collision estimation. Perception 17, 623–632.
Commission of European Communities (CEC), 2000. Priorities in EU
Road Safety: Progress Report and Ranking of Actions. CEC, Brussels.
Connelly, M., Conaglan, H., Parsonson, B., Isler, R., 1998. Child pedes-
trians’ crossing gap thresholds. Accid. Anal. Prev. 30, 443–453.
Corso, J., 1981. Aging Sensory Systems and Perception. Praeger Publish-
ers, New York.
Daigneault, G., Joly, P., Frigon, J.-Y., 2002. Executive functions in the
evaluation of accident risk of older drivers. J. Clin. Exp. Neuropsy-
chol. 24, 221–238.
De Lucia, P., Bleckley, M., Meyer, L., Bush, J., 2003. Judgments about
collisions in younger and older drivers. Transport. Res. Part F 6,
63–80.
Helmers, G., Henriksson, P., Hakamies-Blomqvist, L., 2004. Trafikmiljö
för alder bilförare. Analys och rekommendationer utifrån en lit-
teraturstudie. VTI Rapport 493. Linköping, Sweden (Summary in
English).
Holland, C., Rabbitt, P., 1992. People’s awareness of their age-related
sensory and cognitive deficits and the implications for road safety.
Appl. Cogn. Psychol. 6, 217–231.
Korteling, J., 1994. Effects of aging, skill modification and demand
alternation on multiple-task performance. Hum. Factors 32 (5), 597–
608.
Lee, D., Young, D., McLaughlin, C., 1984. A roadside simulation of
road-crossing for children. Ergonomics 27 (12), 1271–1281.
McDowd, J., Shaw, R., 2000. Attention and aging: a functional perspec-
tive. In: Craik, F., Salthouse, T. (Eds.), The Handbook of Aging and
Cognition, second ed. Lawrence Erlbaum Associates, New Jersey, pp.
221–292.
Mathey, F., 1983. Attitudes and behaviour of elderly pedestrians. Int. J.
Aging Hum. Dev. 17, 25–28.
Mitchell, C., 2000. Some implications for road safety of an aging popula-
tion. In: Transport Trends. Department of the Environment, Transport
and the Regions, Stationary Office, London, pp. 26–34.
National Highway Traffic Safety Administration (NHTSA), 2001. Traffic
Safety Facts 2000: Pedestrians (Report No. DOT-HS-809311). US
Department of Transport, Washington DC.
Organisation for Economic Co-operation and Development (OECD),
2001. Aging and Transport. Mobility Needs and Safety Issues. OECD
Publications, Paris.
Oxley, J., 2000. Age differences in road crossing behaviour. Unpublished
Ph.D. Thesis. Monash University, Melbourne.
Oxley, J., Fildes, B., Ihsen, E., Charlton, J., Day, R., 1997a. Differences in
traffic judgments between young and older adults pedestrians. Accid.
Anal. Prev. 29 (6), 839–847.
Oxley, J., Fildes, B., Ihsen, E., Charlton, J., Day, R., 1997b. Simulation
of the road-crossing task for older and younger adult pedestrians: a
validation study. In: Paper presented at the 1997 National Road Safety
Research and Enforcement Conference, Hobart, Australia.
Oxley, J., Ihsen, E., Fildes, B., Charlton, J., 2001. Age-related functional
impairments and the impact on the ability to cross roads safely. In:
Proceedings of the Traffic Safety on Three Continents Conference,
Moscow, Russia.
Roeneker, D., Cissell, G., Ball, K., Wadley, V., Edwards, J., 2003. Speed-
of-processing and driving simulator training result in improved driving
performance. Hum. Factors 45 (2), 218–234.
Sabey, B., 1988. Pedestrian accidents: nature and scope of the problem.
In: Paper Presented to the Conference of Pedestrian Safety—Positive
Step Forward, London.
 J.A. Oxley et al. / Accident Analysis and Prevention 37 (2005) 962–971 971

Salthouse, T., 1991. Theoretical Perspectives on Cognitive Aging.
Lawrence Erlbaum, New Jersey.
Schiff, W., Oldak, R., Shah, V., 1992. Aging persons’ estimates of vehic-
ular motion. Psychol. Aging 7 (4), 518–525.
Scialfa, C., Guzy, L., Leibowitz, H., Garvey, P., Tyrell, R., 1991. Age
differences in estimating vehicle velocity. Psychol. Aging 6, 60–66.
Simpson, G., Johnston, L., Richardson, M., 2003. An investigation of road
crossing in a virtual environment. Accid. Anal. Prev. 35, 787–796.
Staplin, L., Lyles, R., 1991. Age differences in motion perception and spe-
cific traffic manoeuvre problems. Transport. Res. Rec. 1325, 558–561.
Yanik, A., Monforton, R., 1991. At fault accident patterns of older drivers.
J. Traffic Med. 10 (2), 87–92.