Hands free, voice and touchscreen distraction: how to lower your risk
How voice commands and touchscreens load your mind and eyes, and how to cut the risk
Picture a driver who asks the car to call a friend. The system offers three similar names, mishears the choice, and pulls the driver’s eyes to the screen to sort it out. By the time the call connects, the car has rolled several blocks with only part of the driver’s attention on the road.
Nobody held a phone in that scene. That is why human factors engineers, who study how people use machines, measure what a task does to your thinking, not just your hands and eyes.
Distraction is hard to prove after a crash, so official counts likely run low [1]. Even so, NHTSA reports that 3,208 people died in distraction affected crashes in 2024 [1]. This guide explains what the research shows, ranks common tasks by risk and gives you practical rules for the road.
Why hands free still loads your brain
Engineers sort distraction into three channels: visual (eyes off the road), manual (hands off the wheel) and cognitive (mind off driving). Hands free and voice systems mainly target the first two and leave the third behind.
Cognitive workload is the mental effort a task demands. Your attention is limited. So anything that claims part of it leaves less for spotting a brake light or a child near the curb.
AAA Foundation researchers built a 5 point scale to measure this load, where 1 equals undistracted driving and 5 equals a hard math and memory test [2]. In their 2013 study, a hands free phone call scored 2.27, a handheld call scored 2.45 and talking with a passenger scored 2.33 [2]. As the load rose, drivers scanned the sides of the road less and braked later in a simulator [2]. Your eyes point forward, but you notice less of what is there.
What engineers measure when they test a dashboard
No single number captures distraction, so engineers combine several measures.
- Glance length: NHTSA’s voluntary guidelines aim for glances of 2 seconds or less, because longer glances are linked to more crashes and near crashes [3].
- Total eyes off road time: all glances in one task should add up to 12 seconds or less, under the same guidelines [3].
- Detection response task (DRT): drivers press a button each time a small light flashes. Slow or missed responses signal a heavy mental load [4].
- Task time and errors: in AAA Foundation testing, longer interactions and more system mistakes both tracked closely with higher workload [4].
At 55 mph, you cover about 161 feet in 2 seconds, more than half a football field (calculation). A 12 second task at that speed covers almost 1,000 feet.
Voice commands: short and accurate beats long and clever
IIHS notes that voice commands lead to shorter glances away from the road than manual controls, though their effect on crash risk is unknown [5].
In a 2014 AAA Foundation study, 36 drivers used voice systems in six 2012 and 2013 model vehicles to place calls and choose music [4]. Workload ranged from 1.7 for the easiest system to 3.7 for the hardest [4]. Picking music took 22 to 43 seconds on average, depending on the system [4].
The length of the exchange mattered most. Systems that misheard drivers more often scored higher [4]. The researchers advised limiting spoken menu lists to about four or five items, since working memory is limited [4].
In a related 2014 study, a reliable voice menu scored 2.83, while the same menu with recognition errors scored 3.67 [6]. A 2015 study of 10 vehicles found scores from about 2.4 to 4.6. Five days of practice did not erase the gaps [7].
Nor is voice eyes free: drivers using voice systems in the 2015 study looked more at the dashboard and less at the mirrors and road [7].
Speech to text: you are still writing
Dictating a text removes the keyboard but not the writing. You still pick words, listen to the readback and fix mistakes, all with attention you need for driving.
In the AAA Foundation’s 2013 study, a speech to text task scored 3.06, the highest of the everyday tasks tested [2]. The researchers secretly typed drivers’ words themselves so the system never made a mistake [2]. Real systems that mishear you add more work on top.
A 2014 study found that listening to messages scored about 2.2, while listening and composing replies scored about 3.1 [6]. Across 30 model year 2017 vehicles, texting demanded significantly more than calling or tuning the radio [8]. In a 2019 study, sending a text took younger drivers 27.7 seconds and older drivers 33.8 seconds on average [9].
So do not compose, review or correct messages while moving. Send an “on my way” note before you start, or let a passenger handle it.
Phone assistants, CarPlay and Android Auto
In 2015 testing, the phone assistants Google Now, Siri and Cortana produced workload scores of 3.3, 3.7 and 4.1, all higher than a hands free call [10]. The assistants that made more errors were harder on drivers [10].
Apple CarPlay and Android Auto show a simplified version of your phone on the car’s screen. In a 2018 study of five 2017 and 2018 vehicles, both produced moderate overall demand. The cars’ built in systems produced very high demand [11]. Programming navigation took 48 seconds on average with the built in systems and 33 seconds with CarPlay or Android Auto [11]. Lower is not low, though: the researchers stressed that both still placed moderately high demands on drivers [11].
Newer assistants built on artificial intelligence understand natural speech better, which may cut recognition errors. From a human factors view, they also invite longer, open ended chats. Length is what drives workload [4]. This review found no independent workload or crash studies of these newer assistants, so treat them like any other conversation.
Touchscreens: the eyes and hands problem
A flat screen gives your fingers no landmarks. You look to find the target, look again to confirm it, and often dig through menus.
A large U.S. study used in car cameras to record more than 3,500 drivers over about 3 years, capturing 905 real world crashes [12]. Using in vehicle devices other than the radio or climate controls, such as touchscreen menus, was linked to 4.6 times the crash risk of attentive driving [12].
In the AAA Foundation’s 2017 tests, 23 of 30 new vehicles placed high or very high demand on drivers. None scored low [8]. Entering a destination was the hardest task, at about 40 seconds [8], or roughly half a mile at 45 mph (calculation). Yet 83% of the very high demand vehicles let drivers enter a destination while moving [8].
Surprisingly, touchscreen tasks in that study were less demanding overall than voice tasks, because voice tasks took longest at about 30 seconds [8]. Console controllers, the knobs or touchpads that move a cursor, were the most demanding [8]. No input always wins. Any long task is a poor choice while moving.
Buttons vs. touchscreens: what safety tests now reward
Physical buttons are making a comeback in safety ratings. Euro NCAP, Europe’s new car rating program, began scoring how basic vehicle controls work in January 2026 [13].
To earn credit, a car needs true physical controls for the turn signals, hazard lights, horn, emergency call (eCall) button and gear selector [13]. Euro NCAP defines these as controls that move when pressed, can be found by touch with little or no glance, and give a felt response [13]. Front wipers, high beams and cruise control speed settings also need physical controls. Audio, calling and navigation can score through voice or simple touch controls [13]. The protocol also rewards cameras that detect phone use. It checks for a warning when a driver’s eyes leave the road for a total of 10 seconds within 30 seconds [13].
In the U.S., NHTSA’s distraction guidelines for built in systems are voluntary and do not choose between buttons and screens [3]. They set the glance limits above and recommend blocking manual text entry, video, and reading web pages, social media or texts while driving [3]. NHTSA said in 2013 that it planned separate guidelines for voice systems [3]. Euro NCAP scores cover European models, but they offer a useful lens when you shop.
Your attention lags after the task ends
Ending a call or command does not instantly restore your focus, a delay researchers call lingering or residual distraction.
In the AAA Foundation’s 2015 study of 10 vehicles, reaction times stayed measurably slower for up to 27 seconds after a voice task ended [7]. At the study’s 25 mph speed limit, that is about 990 feet, more than three football fields (calculation). The phone assistant study found lingering effects for up to 18 seconds [10].
A peer reviewed simulator study of 97 drivers ages 21 to 78 found a similar pattern [14]. Drivers drifted side to side more than normal for 15 seconds after a voice task and up to 25 seconds after a texting task [14]. Their visual scanning also narrowed [14].
These figures come from specific systems and test setups, not a fixed recovery time for every driver. Still, the lesson holds: finish any interaction well before a turn, merge, work zone or crosswalk.
Lower risk vs. higher risk in vehicle tasks
This table ranks common tasks using the studies above. Lower risk never means no risk, especially in heavy traffic, bad weather or when you are tired.
Comparison compiled by Safety Behind the Wheel Foundation
| Task while moving | Relative demand | Why | Lower risk approach |
|---|---|---|---|
| Hearing spoken directions for a route set before you left | Lower | Listening only, with no input or decisions | Enter and check the route while parked |
| Listening to audio already playing | Lower | Radio and audiobooks scored low on the workload scale [2] | Pick the station, playlist or podcast before you go |
| Using a familiar physical knob or button | Lower | Found by feel, with little or no glance [13] | Learn the controls in a parked car |
| One step voice command for a car setting | Lower to moderate | Car commands scored about 1.9 of 5 [6] | Use only commands that work the first time |
| Short hands free call you answer | Moderate | Scored about the same as a handheld call [2] | Keep it brief; end it if traffic gets busy |
| Calling a contact by voice | Moderate | Took about 18 to 29 seconds in testing [4, 9] | Use saved favorites; quit after one failed try |
| Having messages read aloud | Moderate | Scored about 2.2 of 5 [6] | Turn off message readouts you do not need |
| Searching for music or podcasts | Moderate to high | Voice music picks took 22 to 43 seconds, with frequent recognition errors [4] | Queue your audio before driving |
| Dictating or correcting a text | High | Composing pushed workload to about 3 of 5 [2, 6] | Send updates before you start, or pull over |
| Digging through touchscreen menus | High | Linked to 4.6 times the crash risk in real world data [12] | Change settings while parked |
| Entering or changing a destination | Highest | About 31 to 48 seconds, depending on system and driver [8, 9, 11] | Set it before you go, or let a passenger do it |
| Open ended chat with an AI assistant | Unknown, likely high | Long, unscripted exchanges with little research so far | Save it for a stop |
A quick decision rule: if a task needs more than one short command or one quick glance, it can wait.
A setup routine and rules for the road
An IIHS survey published in 2020 found only about 1 in 5 iPhone owners had Do Not Disturb While Driving set to turn on automatically [5].
Once you are moving, follow these rules:
- One command, one try: if voice fails once, drop the task until you stop.
- No writing: do not dictate, review or correct messages.
- Keep calls short: call back later for anything complex or emotional.
- Mind your timing: attention can lag for up to half a minute. So avoid tasks as you near turns, merges and crosswalks [7].
- Keep glances under 2 seconds, then bring your eyes back to the road [3].
- Use a copilot: let a passenger run the screen, messages and route changes.
Advice for teens, older drivers and employers
Teens and parents: new drivers have less spare attention. GHSA reports that 36 states and D.C. ban all phone use, including hands free, by novice drivers [15]. Parents can model the setup routine and make “phone set, then go” a family rule.
Older drivers: in AAA Foundation testing, drivers 55 to 75 needed about 40 seconds to enter a destination, versus 31 seconds for drivers 21 to 36 [9]. Simple physical controls and setup before driving help most.
Employers: a policy that bans only handheld phones misses much of the risk [2]. Cover hands free calls and in vehicle messaging. Never expect workers to take calls while driving.
Frequently asked questions
Is talking on a hands free phone safer than holding the phone?
It frees your hands, which helps, but the conversation still loads your mind. In AAA Foundation tests, hands free and handheld calls produced nearly the same mental workload [2]. Keep calls short, or wait until you stop.
Is voice to text safe while driving?
Research does not support it as a safe option. Composing a message by voice raised workload well above listening, even with error free recognition. Texting tasks took longer than calls or radio changes [2, 6, 8]. Send messages before you drive or after you park.
Are CarPlay and Android Auto safer than the car’s own system?
In a 2018 study, both were less demanding than the built in systems in the same cars, especially for navigation [11]. They still placed moderately high demands on drivers, so set up first and keep any use while moving brief [11].
Are touchscreens more distracting than buttons?
Physical buttons let you work by feel with little or no glance. Euro NCAP’s 2026 tests reward them for key functions [13]. Touchscreens are not always worse, but deep menus and small targets pull your eyes down longer.
The bottom line
Recommendation from Safety Behind the Wheel Foundation
Hands free features change where a task happens, not how much attention it takes. Set your route, audio and phone’s driving mode before you move. On the road, stick to short, one step commands. Give your mind a half minute buffer before any turn, merge or crosswalk.
Related articles
- Distracted driving explained: how attention fails and how to protect it
- Phone setup before driving: set it once, then drive distraction free
- Using GPS navigation safely: plan the route, skip the sudden moves
- Distracted driving laws by state: 2026 guide to phone and texting rules
- Distracted driving policy at work: a guide for employers and drivers
Sources and further reading
- 1National Highway Traffic Safety Administration. Distracted Driving in 2024. DOT HS 813 790. April 2026. crashstats.nhtsa.dot.gov/…/813790
- 2AAA Foundation for Traffic Safety (Strayer DL, et al.). Measuring Cognitive Distraction in the Automobile. June 2013. aaafoundation.org/research/measuring-cognitive-distraction-automobile
- 3National Highway Traffic Safety Administration. Visual-Manual NHTSA Driver Distraction Guidelines for In-Vehicle Electronic Devices. Federal Register notice 2013-09883. April 26, 2013. federalregister.gov/…/2013-09883
- 4AAA Foundation for Traffic Safety (Cooper JM, Ingebretsen H, Strayer DL). Mental Workload of Common Voice-Based Vehicle Interactions across Six Different Vehicle Systems. October 2014. aaafoundation.org/research/mental-workload-common-voice-based-vehicle-interactions
- 5Insurance Institute for Highway Safety. Distracted driving (research area overview). Accessed October 2026. iihs.org/research-areas/distracted-driving
- 6AAA Foundation for Traffic Safety (Strayer DL, et al.). Measuring Cognitive Distraction in the Automobile II: Assessing In-Vehicle Voice-Based Interactive Technologies. October 2014. aaafoundation.org/research/measuring-cognitive-distraction-automobile-ii
- 7AAA Foundation for Traffic Safety (Strayer DL, et al.). Measuring Cognitive Distraction in the Automobile III: A Comparison of Ten 2015 In-Vehicle Information Systems. October 2015. aaafoundation.org/research/measuring-cognitive-distraction-automobile-iii
- 8AAA Foundation for Traffic Safety (Strayer DL, et al.). Visual and Cognitive Demands of Using In-Vehicle Information Systems. September 2017. aaafoundation.org/research/visual-cognitive-demands-using-vehicle-information-systems
- 9AAA Foundation for Traffic Safety (Cooper JM, et al.). Age-Related Differences in the Cognitive, Visual and Temporal Demands of In-Vehicle Information Systems. July 2019. aaafoundation.org/research/age-related-differences
- 10AAA Foundation for Traffic Safety (Strayer DL, et al.). The Smartphone and the Driver’s Cognitive Workload: A Comparison of Apple, Google, and Microsoft’s Intelligent Personal Assistants. October 2015. aaafoundation.org/research/smartphone-drivers-cognitive-workload
- 11AAA Foundation for Traffic Safety (Strayer DL, et al.). Visual and Cognitive Demands of Using Apple CarPlay, Google’s Android Auto and Five Different OEM Infotainment Systems. June 2018. aaafoundation.org/research/visual-cognitive-demands-apples-carplay-googles-android-auto
- 12Dingus TA, et al. Driver crash risk factors and prevalence evaluation using naturalistic driving data. Proceedings of the National Academy of Sciences. 2016;113(10):2636 to 2641. pmc.ncbi.nlm.nih.gov/articles/PMC4790996
- 13Euro NCAP. Safe Driving: Driver Engagement Protocol. Version 1.2. July 2026 (implementation January 2026). cdn.euroncap.com/…/Driver_Engagement_v1_2
- 14Snider J, et al. Distraction “hangover”: Characterization of the delayed return to baseline driving risk after distracting behaviors. Human Factors. 2023;65(2):306 to 320 (online 2021). pubmed.ncbi.nlm.nih.gov/33908806
- 15Governors Highway Safety Association. Distracted Driving (state laws). Updated October 2025. ghsa.org/state-laws/issues/distracted-driving
This article reflects the safety priorities of automotive human factors engineering. It is general education, not legal, medical or professional advice for any individual. Laws, data and vehicle features change, so verify details with official sources. Reviewed October 2026.
Keep learning
Safe Driving Resource Center
Guides, self-checks and articles for drivers, parents of new drivers, and families recovering after a crash.
