Work Experience 2025 - What Does It Mean To Prove An Automated Vehicle Is Safe?
In the last week of May 2025, Reed Mobility was fortunate to be joined by a 17-year old work experience student for four days. Studying for her A-level in psychology and looking to continue her studies in this field at university, I was keen for her to learn about the work of Reed Mobility and how we apply knowledge of psychology in the field of road safety. The student was set several tasks and I was absolutely delighted with her work. The piece below is an such example, describing how we might go about proving that automated vehicles are safe for use on public roads. I particularly liked her analogy between feelings about AV safety and public confidence in food standards. For someone with no particular background in transportation, I thought this was a really insightful (and hopeful!) treatment of the subject - see what you think!
What does it mean to prove an Automated Vehicle is safe?
Automated Vehicles (AVs) won’t be alone on the road for decades, and subsequently will never be entirely and undoubtedly ‘safe’ from a collision (1). Proving the safety of an AV means convincing both the government and public of the reliance, and trustworthiness of the vehicles, as well as the standardised and regular testing and maintenance that each vehicle should undergo.
What is safe and good driving?
No strict definition of safe driving has been set in place, aside from what is presented by the Highway Code. Research by Fraade-Blanar et al (2018) defines safety as “the elimination, minimization, or management of harm to the public (with an emphasis on people, although it can include animals and property)”, while De Freitas et al (2021) introduced the SPRUCE model (safe, predictable, reasonable, uniform, comfortable, and explainable). However, a fundamental aspect of human driving behaviour is ‘common sense’ (De Freitas et al, 2021); the sense to make ethical and safe choices, and the ability to make split-second judgements in new situations - a sense we can’t, and many won’t, expect an AV to possess. In a possibly more obvious definition of good driving, it could be stated as the ability to make a passenger as well surrounding drivers feel safe. However, ‘good driving’ differs vastly with each driver; a person’s tendency for recklessness stems both from individual differences (personality, emotions, external attribution) and the influence of others in their life. Following social learning theory, a driver’s view of good driving is a result of how their parents, friends, and others around them all drive. Furthermore, the vast majority of drivers, somewhat regardless of gender, rate themselves as above average across multiple aspects of driving (Finn & Bragg, 1986; McCormick, Walkey & Green, 1986; Marottoli & Richardson, 1998; McKenna, Stanier & Lewis, 1991). As a result, the capabilities and driving behaviour of a ‘good’ AV will be expected not only to conform to a widely varying perception of good driving but also an extremely high standard in order to be accepted as an alternative to human drivers. Entering onto the road introduces a sense of vulnerability and this is furthered upon ‘giving up control’ to an AV. Not only does the driving behaviour become a factor in safety, but the general functions of the vehicle are critical as well - knowing the vehicle consistently performs to the standards of the Highway Code, as well as individual performance expectations is incredibly important to ‘prove’ the vehicle as safe.
How to test for safety
Framework for the testing of AVs (Fraade-Blanar et al, 2018) suggests that the vehicles should be tested in multiple environments (simulation, closed courses, and public roads both with and without a safety driver) as this increases the ecological validity of each test, making sure the vehicle works and behaves as expected in multiple realistic scenarios. The framework additionally states that the methods used to test the safety of AVs must be valid, feasible, reliable, and non-manipulative.
What makes people feel safe?
A notable demonstration of acceptance towards new products is attitudes towards food safety, and the way that the general population responds to new food brands and products. According to the Food Standards Agency (2), the factors influencing the trust and perceived safety of new foods are familiarity, regulation, reputation, and endorsement. Multiple studies lend themselves to the idea that familiarity breeds not contempt, but an increased appreciation; Winkielman, Schwarz, Fazendeiro and Reber (2003) understand that familiarity is rewarding as it leads to efficient processing - no new ideas are required to be understood. Similarly, Zajonc (2001) states that repeated exposure to a stimulus is a form of classical conditioning, wherein the absence of negative effects supports the safety of the stimulus - if nothing wrong occurs each time, there is reason to believe no negative effects will result in the next exposure.
Considering food safety, the laws and regulations in place are widely known across the UK, and as such many people feel confident that their food is of a high standard. A 2021 survey revealed that out of 2,182 UK adults, over half had become more conscious of food safety since the Covid-19 pandemic (3). As a result, consumers held more trust in the quality of new products - a result which could easily be applied to AVs. Consumers were also reporting more trust and safety when the food brand holds a strong reputation, for example McDonalds against a local, single-store butchers. The reputation of a product derives from the public opinion on it, and as such, endorsement and positive reviews are incredibly useful to establish trust and safety. Hearing from a friend or coworker about their new AV, or even seeing them using it, will strongly influence a driver to make a similar purchase. In a 2021 study by Nielsen, 88% of consumers trusted reviews from those they know in real life (4), as they empathise and relate to the people similar to them, and therefore trust them more than they would an advert. This is supported by group conformity, and the research by Asch (1951). Asch’s experiment looked into the behaviour of participants in a group, with the conclusion that an individual’s beliefs and opinions can be significantly swayed and altered by those around them. This lends to the idea that, since others accept a product, it is reasonable for an individual to follow this acceptance.
How to apply feelings of safety to AVs
While the current MOT system could be improved, there is currently no specialised and standardised testing in place for AVs - an issue that needs to be addressed. As stated previously, many consumers feel safer knowing that their products are being produced and maintained at an extremely high standard, and these findings should be applied to the introduction of AVs. Publicly and ‘loudly’ implementing a thoroughly regulated system to maintain the safety and quality of automated vehicles should drastically improve the public perception, as the knowledge that there are no secrets kept will build trust, and drivers now have reason to believe they are safe - whether inside or alongside an AV. To combat familiarity, introducing as many AVs to the roads as is initially possible is likely to aid in society’s adjustment. While this has the potential to fail, if a number of successful AV journeys are completed along public roads, the absence of a negative consequence will be highlighted in the brains of drivers, and increasingly more people will grow open to not only driving next to an AV, but eventually to travelling inside of one. Furthermore, with as many people in AVs as would currently be permitted, there will be a relatively large number of people to provide first hand accounts, and to support and vouch for the vehicles - subsequently influencing their acquaintances and increasing the number of people willing to travel in an automated vehicle. Utilising the reputation of established car companies in the production and advertising of such vehicles will be critical; drivers are unlikely to feel safe in what is already a controversial mode of travel if it has been produced by an unknown company, or one lacking a solid reputation. Encouraging popular brands to work with AVs will enhance the publicity of the vehicles, and ensure people feel as though their cars are produced to a good quality, and primarily a safe standard.
Footnotes and references
1 While the proportional acceptance of AVs is increasing in the UK, and the primary demographic that is open to the idea is the younger generation, the percentage found of those who would trust a driverless car is still only 22%. (Hope, G. 2024)
2 Chapter 3: Factors influencing trust and confidence in food bought online. (2023, October 27). Food Standards Agency. https://www.food.gov.uk/research/chapter-3-factors-influencing-trust-and-confidence-in-food-bought-online
3 Survey Shows UK Public More Aware of Food Safety and Hygiene Since Beginning of Pandemic. (2021, April 19). Food-Safety https://www.food-safety.com/articles/7088-survey-shows-uk-public-more-aware-of-food-safety-and-hygiene-since-beginning-of-pandemic
4 Cunningham, P. (2025, March 21). 88% of Consumers Trust Word of Mouth Buyapowa https://www.buyapowa.com/blog/88-of-consumers-trust-word-of-mouth/
Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgments. In H. Guetzkow (Ed.), Groups, leadership and men; research in human relations (pp. 177–190). Carnegie Press
Finn, P., & Bragg, B. W. (1986). Perception of the risk of an accident by young and older drivers. Accident Analysis and Prevention, 18(4), 289–298. https://psycnet.apa.org/doi/10.1016/0001-4575(86)90043-6
Fraade-Blanar, L., Blumenthal, M. S., Anderson, J. M., & Kalra, N. (2018, October 11). Measuring Automated Vehicle Safety: Forging a Framework Rand https://www.rand.org/pubs/research_reports/RR2662.html
Freitas, J. D., Censi, A., Smith, B. W., Lillo, L. D., Anthony, S. E., & Frazzoli, E. (2021, March 1). From driverless dilemmas to more practical commonsense tests for automated vehicles Proceedings of the National Academy of Sciences https://doi.org/10.1073/pnas.2010202118
Hope, G. (2024, November 13) Survey Shows UK Drivers Are Warming Up to Driverless Cars IOT World Today https://www.iotworldtoday.com/transportation-logistics/survey-shows-uk-drivers-are-warming-up-to-driverless-cars#close-modal
McCormick, I. A., Walkey, F. H., & Green, D. E. (1986). Comparative perceptions of driver ability— A confirmation and expansion. Accident Analysis and Prevention, 18(3), 205–208. https://doi.org/10.1016/0001-4575(86)90004-7
McKenna, F. P., Stanier, R. A., & Lewis, C. (1991). Factors underlying illusory self-assessment of driving skill in males and females. Accident Analysis and Prevention, 23(1), 45–52. https://doi.org/10.1016/0001-4575(91)90034-3
Marottoli, R. A., & Richardson, E. D. (1998). Confidence in, and self-rating of, driving ability among older drivers. Accident Analysis and Prevention, 30(3), 331–336. https://doi.org/10.1016/S0001-4575(97)00100-0
Winkielman, P., Schwarz, N., Fazendeiro, T., Reber, R., Musch, J., & Klauer, K. C. (2003, January 30). The psychology of evaluation: Affective processes in cognition and emotion. In The hedonic marking of processing fluency: Implications for evaluative judgment (pp. 189-217). Mahwah, NJ: Lawrence Earlbaum.
Zajonc, R. B. (2001, December). Mere exposure: A gateway to the subliminal. Current directions in psychological science, 10(6), 224-228.