Driver Distraction Is the Focus in the Right Place? · CVO Distracted Driving Study ... Hickman et...
Transcript of Driver Distraction Is the Focus in the Right Place? · CVO Distracted Driving Study ... Hickman et...
Driver Distraction – Is the Focus in the Right Place?
Richard Hanowski, Ph.D. Director, Center for Truck & Bus Safety
Virginia Tech Transportation Institute
Environmental, Health and Safety Communication Panel Webinar May 16th, 201200
Presentation Overview
• What is “driver distraction”?
• Overview of research approaches used to investigate driver safety/distraction
• What is a naturalistic driving study (NDS)?
• NDS examples investigating driver distraction
• Understanding the controversial findings
• Research to Practice implications (education, policy and design)
• Links to data, naturalistic driving videos, and safety-tip information
5/14/2012 VTTI | Driving Transportation with Technology 2
What is Driver Distraction?
• US-EU Bilateral ITS Technical Task Force (2010):
– Driver distraction is the diversion of attention from activities critical for safe driving to a competing activity.
• Competing activities and the types of information presented to drivers are wide and varied
5/14/2012 VTTI | Driving Transportation with Technology 3
Types of Driver Information
• External
– Information that is outside the vehicle
– Examples: billboards
5/14/2012 VTTI | Driving Transportation with Technology 4
Types of Driver Information
• External
– Information that is outside the vehicle
– Examples: traffic lights
5/14/2012 VTTI | Driving Transportation with Technology 5
Types of Driver Information
• External
– Information that is outside the vehicle
– Examples: street signs, etc…
5/14/2012 VTTI | Driving Transportation with Technology 6
Internal Information
• Information that is inside the vehicle
• Examples: integrated systems
5/14/2012 VTTI | Driving Transportation with Technology 7
Internal Information
• Information that is inside the vehicle
• Examples: nomadic devices, passengers, etc…
5/14/2012 VTTI | Driving Transportation with Technology 8
Potential Impact of Information
• Impact of information on driver behavior can be positive (support) or…
5/14/2012 VTTI | Driving Transportation with Technology 9
Potential Impact of Information
• Impact of information on driver behavior can be negative (distract)
5/14/2012 VTTI | Driving Transportation with Technology 10
5/14/2012 VTTI | Driving Transportation with Technology 11
Driver Safety Research Approaches
• Epidemiological Studies
• Police Accident Reports (PARs)
• Simulation
• Test Track
• Lab Experiment
• Missing Piece? NDS
Empirical Data Collection
Large-Scale Pseudo Naturalistic Data
Collection
• Proactive • Provides
important ordinal crash risk info
• Precise knowledge about crash risk
• Information about important circumstances and scenarios that lead to crashes
• Imprecise, relies on unproven safety surrogate
• Experimental situations modify driver behavior
• Reactive • Very limited
pre-crash information
• “Natural” driver behavior • Detailed pre-crash/crash info
•Distraction •Fatigue •Aggressive driving •Driver errors •Vehicle dynamics
• Potential validation of surrogate measures
Epidemiological Analysis
5/14/2012 VTTI | Driving Transportation with Technology 13
NDS
4X Crash Risk with Cell Phone Use (1 of 2)
• Redelmeir & Tibshirani (1997)
– Overlaid cell phone records with crashes
– Consent to review cell phone records
• Key finding: anytime within 10-minutes before the crash, the relative risk of a call was 4.3
– 76% of the cell phone calls lasted 2 mins or less (average call length was 2.3 mins)
5/14/2012 VTTI | Driving Transportation with Technology 14
NDS
4X Crash Risk with Cell Phone Use (2 of 2)
• Authors binned time intervals in relation to the crash
– 1-5 mins, 6-10 mins, 11-15 mins, and 16-20 mins
• Given average call duration, likely, in most of the crashes, the driver was not on the phone at the time of the crash.
• Even if driver was on the phone, what were they doing?
– Dialing, talking/listening, texting
5/14/2012 VTTI | Driving Transportation with Technology 15
NDS
Crashes vs Wireless Growth
Police Reported Crash Rates and Wireless Subscription Growth
1988-2008
0
50,000,000
100,000,000
150,000,000
200,000,000
250,000,000
300,000,000
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Year
Nu
mb
er o
f S
ub
scri
ber
s
0
50
100
150
200
250
300
350
Cra
shes
per
100
Mil
lio
n V
ehic
le M
iles
Tra
vele
d
Number of Wireless Subscribers Crashes per 100 Million Vehicle Miles Traveled
5/14/2012 VTTI | Driving Transportation with Technology 16
NDS
0
50
100
150
200
250
300
0
50000000
100000000
150000000
200000000
250000000
300000000
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Cra
she
s p
er
10
0 M
illi
on
Ve
hic
le M
ile
s T
rav
ele
d
Nu
mb
er
of
Su
bsc
rib
ers
Year
Number of Wireless Subscribers Predicted CR (Using 4x Odds) Crashes per 100 Million Vehicle Miles Traveled
Source: Hanowski, 2011
Study Conclusions vs. Reality
5/14/2012 VTTI | Driving Transportation with Technology 17
NDS
Cell Phone Use Similar to Driving Drunk (1 of 2)
• Strayer, Drews, & Crouch (2006)
– High-fidelity driving simulator
– 15-min car-following paradigm
– 3 counter-balanced driving conditions
• Baseline, talking on a cell phone, blood alcohol of 0.08
• Conversation was “naturalistic”
• Topics that were identified on the first day as being of interest to the participant
• No manual component to cell phone use
5/14/2012 VTTI | Driving Transportation with Technology 18
NDS
Cell Phone Use Similar to Driving Drunk (2 of 2)
• Results
– Drove slower in talking and alcohol conditions
– Followed closer in alcohol, further in talking conditions
– Reaction time 100 ms slower in cell phone, no difference between baseline and alcohol
– 3 crashes in cell phone, 0 in baseline and alcohol
• No difference in crashes and reaction time between baseline and alcohol…?
• Expectancy effects of alcohol…?
– Did the control include a beverage swiped with alcohol on rim
5/14/2012 VTTI | Driving Transportation with Technology 19
NDS
Brain Functioning
• Just et al. (2008) impact of an auditory task during simulated drive
• Functional magnetic resonance imaging
• Parietal lobe activation decreased by 37%
– No significant change in visual, motor, or executive functions
• Relationship to crash risk?
• More research needed…
5/14/2012 VTTI | Driving Transportation with Technology 20
• Study participants use an instrumented vehicle for an extended period (weeks to 1 yr or more)
• Able to get detailed pre-crash/crash information and routine driving behaviors
• Over 100 data measures collected on driving performance (e.g., lane position), actigraphy, questionnaires
• Examples include: 100-Car, Sleeper Berth, Local/Short Haul, Teen, Older, Motorcycle, Bicycle…
• High validity and Low control
Behind
Vehicle
Front
Vehicle
Camera 4
Camera 3
Camera 2
Camera 1
Naturalistic Data Collection
5/14/2012 VTTI | Driving Transportation with Technology 21
Data Acquisition System (DAS) box
DAS box behind driver seat
Front Radar Rearward Camera
Face & Forward Cameras
Instrumentation Layout Example
5/14/2012 VTTI | Driving Transportation with Technology 22
5/14/2012 23 VTTI | Driving Transportation with Technology
CVO Distracted Driving Study – Olson et al. (2009)
• Use VTTI’s naturalistic truck study data • 203 drivers, 7 fleets, 55 trucks, 3 million miles • Video review of all safety-critical events (n=4452) and
baselines/normal driving (n=19,888)
• Identify non-driving tasks/behaviors engaged in immediately prior to involvement in safety events
• What tasks do drivers engage in and do they increase risk?
• What is the impact of tasks on drawing the driver’s eyes away from the forward roadway?
• “Instant replay”
5/14/2012 24 VTTI | Driving Transportation with Technology
Sample of Non-Driving Tasks
Task Odds Ratio
Lower Confidence
Limit
Upper Confidence
Limit
Safety-Critical
Events (n)
Baselines (n)
Text message on cell phone 23.24 9.69 55.73 31 6
Interact with/look at dispatching device
9.93 7.49 13.16 155 72
Write on pad, notebook, etc. 8.98 4.73 17.08 28 14
Use calculator 8.21 3.03 22.21 11 6
Look at map 7.02 4.62 10.69 56 36
Dial cell phone 5.93 4.57 7.69 132 102
Talk or listen to hand-held phone 1.04 0.89 1.22 195 837
Talk or listen to hands-free phone 0.44 0.35 0.55 91 901
Talk or listen to CB radio 0.55 0.41 0.75 50 399
• 60% of the safety-critical events had some type of non-driving task being performed
5/14/2012 25 VTTI | Driving Transportation with Technology
“Vision is King”
0
1
2
3
4
5
0
5
10
15
20
25
Text
mes
sage
on
cell
phon
e
Oth
er –
Com
plex
Inte
ract
with
/loo
k at
dis
patc
hing
dev
ice
Wri
te o
n pa
d, n
oteb
ook,
etc
.
Use
cal
cula
tor
Look
at
map
Use
/rea
ch fo
r ot
her e
lect
roni
c de
vice
Dia
l cel
l pho
ne
Oth
er -
Mod
erat
e
Pers
onal
gro
omin
g
Read
boo
k, n
ewsp
aper
, pap
erw
ork,
etc
.
Put o
n/re
mov
e/ad
just
sun
glas
ses
Reac
h fo
r ob
ject
in v
ehic
le
Look
bac
k in
Sle
eper
Ber
th
Oth
er –
sim
ple
Rem
ove/
adju
st je
wel
ry
Put o
n/re
mov
e/ad
just
hat
Put o
n/re
mov
e/ad
just
sea
t bel
t
Adj
ust i
nstr
umen
t pa
nel
Look
at
left
-sid
e m
irro
r/ou
t lef
t win
dow
Talk
/sin
g/da
nce
with
no
indi
catio
n of
pas
seng
er
Talk
or
liste
n to
han
d-he
ld p
hone
Use
che
win
g to
bacc
o
Eatin
g
Smok
ing-
rela
ted
beha
vior
–ci
gare
tte
in h
and/
mou
th
Dri
nk fr
om a
con
tain
er
Look
at
righ
t-si
de m
irro
r/ou
t lef
t w
indo
w
Oth
er p
erso
nal h
ygie
ne
Smok
ing-
rela
ted
beha
vior
–lig
htin
g
Talk
or
liste
n to
CB
mic
roph
one
Look
at
outs
ide
vehi
cle
Bite
nai
ls/c
utic
les
Talk
or
liste
n to
han
ds-f
ree
phon
e
Inte
ract
with
or l
ook
at o
ther
occ
upan
t(s)
Chec
k sp
eedo
met
er
Mea
n D
urai
ton
of E
yes
off
Forw
ard
Road
way
(se
c)
Poin
t Est
imat
e O
dds
Rati
o
Point Estimate Odds Ratio Mean Duration of Eyes off Forward Roadway (sec)
5/14/2012 26 VTTI | Driving Transportation with Technology
Inattention-Related Relative Crash/Near Crash Risk Estimates (Odds Ratio 0-2)
5/14/2012 27 VTTI | Driving Transportation with Technology
Inattention-Related Relative Crash/Near Crash Risk Estimates (Odds Ratio 2-4)
5/14/2012 28 VTTI | Driving Transportation with Technology
Inattention-Related Relative Crash/Near Crash Risk Estimates (Odds Ratio 4-6)
5/14/2012 29 VTTI | Driving Transportation with Technology
Inattention-Related Relative Crash/Near Crash Risk Estimates (Odds Ratio >6)
5/14/2012 30 VTTI | Driving Transportation with Technology
Noted Study Limitations
• Relatively few crashes
• Relatively few drivers/trucks/miles
• Volunteer participants
• Are SCEs valid measures?
• FMCSA-funded study using DriveCam® data was conducted to address these limitations…
5/14/2012 31 VTTI | Driving Transportation with Technology
• 13,305 vehicles (trucks and buses)
• 1,085 crashes; 39,036 near-crashes and events
• 211,171 baseline (non-events)
• Distraction analysis data- 92 consecutive days from June 6, 2009 through September 5, 2009
VTTI-DriveCam® Distracted Driving Study- Hickman et al. (2010)
5/14/2012 32 VTTI | Driving Transportation with Technology
DriveCam® Cameras, Views, and Kinematic Data
VTTI – DriveCam® Distracted Driving Study Results
5/14/2012 33 VTTI | Driving Transportation with Technology
Tertiary Task Odd
Ratio
Lower
Conf Limit
Upper
Conf
Limit
Freq of
Safety
Critical
Events
Freq of
Baselines
Cell Phone Usage 1.14* 1.06 1.23 895 4,262
Dialing Cell Phone 3.51* 2.89 4.24 165 256
Talk/Listen Hands Free
Cell Phone 0.65* 0.56 0.76 194 1,626
Talk/Listen Hand Held
Cell Phone 0.89 0.80 1.00 372 2,266
Reaching for Bluetooth
Device 3.38* 2.64 4.31 104 168
Reaching for Cell Phone 3.74* 2.97 4.71 122 178
5/14/2012 34 VTTI | Driving Transportation with Technology
Are SCE’s Valid Safety Measures?
Tertiary Task
Odds Ratios for Tractor
Trailers/Tankers Only in
Hickman et al. (2010)
Odds Ratios in
Olson et al. (2009)
Any Cell Phone Use 1.08 1.04
Dialing Cell Phone 5.44* 5.93*
Talk/Listen Hands-Free Cell Phone 0.58* 0.44*
Talk/Listen Hand-Held Cell Phone 1.01 1.04
Reaching for Bluetooth Device 4.43* 6.72*
Reaching for Cell Phone 7.60* Included in dial cell
phone
Text/Email/Web + 23.24*
Food/Drink 1.53* 1.01
* Asterisk indicates a significant odds ratio. These ratios are also shown in bold.
+ odds ratio calculation was only performed across all vehicle types
5/14/2012 35 VTTI | Driving Transportation with Technology
Discussion
• What does “Use a Cell Phone” mean?
5/14/2012 36 VTTI | Driving Transportation with Technology
“Use a Cell Phone”
• “Use a cell phone” is a higher-order task that is comprised of several sub-tasks:
– Reaching
– Dialing
– Talking/listening
– Texting
– Etc, etc, etc
• By not assessing the risk of each sub-task, or using terms interchangeably, don’t have a clear picture of the associated risk
• Effect of poorly defined terms is that it can mislead and misinform policy
5/14/2012 37 VTTI | Driving Transportation with Technology
• Main conclusion of this research is that information presented to a driver can have profound safety implication
• Findings have provided insight to:
– Educators/Trainers/Outreach
Commercial Motor Vehicle Driving Tips, FMCSA website: http://www.fmcsa.dot.gov/about/outreach/education/drivertips/index.htm
National Safety Council’s Motor Fleet Safety Manual, 5th Edition
– Technology designers
Human factors design of driver-vehicle interfaces
– Policy makers
Trucker texting ban; hand-held cell phone ban; model Fleet Distraction Policy (FMCSA website)
Research-to-Practice
5/14/2012 38 VTTI | Driving Transportation with Technology
TomTom® Study (Antin et al., 2009)
• Navigation system produced the lowest mental effort ratings, fastest reaction time on peripheral detection tasks, fewest driving errors, fewest occurrence of getting lost
• Study results support that notion that an onboard, moving-map navigation system can play a meaningful role in enhancing navigational efficiency and safety
• 38 participants drove instrumented cars on public roads • Navigation tasks (“wayfinding”)- participants unfamiliar
with the routes selected • Navigation tools:
• Navigation system • Paper maps • Yahoo printouts
5/14/2012 39 VTTI | Driving Transportation with Technology
Ford SYNC® Study (Owens et al., 2010)
• Driving performance (e.g., steering) degraded when dialing and selecting tracks with handheld devices, but not degraded with Ford SYNC®
• No difference when engaged in conversation between handheld phone and the Ford SYNC®- also, no difference from baseline
• 21 participants drove instrumented cars on public roads and on the Virginia Smart Road test track
• Visual distraction and driving performance was measured as drivers used handheld phones, mp3 players, and the Ford SYNC® system
• Tasks included: • Dialing • Phone conversations • Selecting music tracks
• Drivers able to maintain eyes forward when dialing and selecting tracks with Ford SYNC® but not with handheld devices
5/14/2012 40 VTTI | Driving Transportation with Technology
Policy Impacts of Naturalistic Research
• Within a few weeks of release of Olson et al. (2009), there was an Executive Order banning texting-while-driving by government workers
• FMCSA issued texting bans for truckers, while PHMSA recently promulgated a similar rule for HazMat drivers
• USDOT distraction policy efforts cite Olson et al. (2009) and Hickman et al. (2010) as scientific support
5/14/2012 41 VTTI | Driving Transportation with Technology
Cell Phone Policy Implications
FMCSA promulgated regulation to restrict hand-held phone use, but not talking
Hands-free and talking is permitted
• Unintended consequences: not allowing truck drivers to utilize a drowsiness countermeasure they use may negatively impact safety
• Driving for 11 hours is a monotonous task!
5/14/2012 42 VTTI | Driving Transportation with Technology
New Naturalistic Studies • SHRP 2 Safety Program (TRB)
– ~ 2,000 cars • Cell phone data?
– Canadian and Australian cohorts in development
• 270 Truck Study (FMCSA) • Cell phone data?
• Both studies will have analysis opportunities for outside researchers
• Goal of both efforts is for the data to be open access
• Video would have protection (Institutional Review Board requirement)
5/14/2012 43 VTTI | Driving Transportation with Technology
Analysis Opportunities • Kinematic naturalistic data from light vehicle (100-
Car) and heavy vehicle (8-Truck) studies available on-line: http://forums.vtti.vt.edu/
• Open to all researchers to use the data
• Open forum to add algorithms, etc (e.g., SAFER)
5/14/2012 44 VTTI | Driving Transportation with Technology
Presentation Recap • Naturalistic driving- important approach that
supplements empirical and epidemiological methods
• Driver distraction- a complex issue that must be addressed via policy/enforcement, education and well-designed technologies (driver-vehicle interface)
• Research findings controversy- lab and simulator studies can give important information on expected driver behavior, but important to investigate actual driver behavior that is collected in the full-context of the driving environment
5/14/2012 45 VTTI | Driving Transportation with Technology
Presentation Recap • Research to Practice Implications- naturalistic driving
research has provided compelling insight and scientific support for safety programs and countermeasures directed at education, policy and design
• Analysis Opportunities- Naturalistic data are currently available to answer your research questions
5/14/2012 46 VTTI | Driving Transportation with Technology
Key References Used • Dingus, T. A., Hanowski, R. J. and Klauer, S. (October, 2011). Estimating crash
risk. Ergonomics in Design: The Quarterly of Human Factors Applications, 19 (4):8-12
• Hanowski, R.J. (2011). The naturalistic study of distracted driving: Moving from research to practice. SAE International L. Ray Buckendale Lecture, Paper No. 2011-01-2305; doi: 10.4271/2011-01-2305.
• Hickman, J.S., Hanowski, R.J., and Bocanegra, J. (September, 2010). Distraction in commercial trucks and buses: Assessing prevalence and risk in conjunction with crashes and near-crashes. Report No. FMCSA-RRR-10-049. Washington, DC: Federal Motor Carrier Safety Administration. http://www.fmcsa.dot.gov/facts-research/research-technology/report/Distraction-in-Commercial-Trucks-and-Buses-report.pdf
• Olson, R.L., Hanowski, R.J., Hickman, J.S., & Bocanegra, J. (September, 2009). Driver distraction in commercial vehicle operations, final report. Report No. FMCSA-RRR-09-042. Washington, DC: Federal Motor Carrier Safety Administration. http://www.fmcsa.dot.gov/facts-research/research-technology/report/FMCSA-RRR-09-042.pdf
• Redelmeier, D.A., and Tibshirani, R.J. (1997). Association between cellular-telephone calls and motor vehicle collisions. The New England Journal of Medicine, 336(7), 453-458.
• Strayer, D.L., Drews, F.A., and Crouch, D.J. (2004). A comparison of the cell phone driver and the drunk driver. Working Paper 04-13. AEI-Brookings Joint Center for Regulatory Studies, aei-brookings.org
Questions?