WIRED DIFFERENTLY: How London’s cabbies' brains could help shape the future of AI navigation
London’s black cab drivers are renowned for their unparalleled knowledge of the city’s streets. Passing The Knowledge test demands memorising over 26,000 streets, an extraordinary feat that has long fascinated scientists. Now, new research suggests their route-planning methods could inspire advancements in AI navigation systems.
A study published in Proceedings of the National Academy of Sciences (PNAS) reveals that London taxi drivers plan routes in a way that is starkly different from the algorithms used by satnav systems. While satnavs calculate every possible path to a destination, cabbies take a more strategic approach. They identify and prioritise the most challenging parts of the route first, then build the rest of the journey around these key points.
To understand this, researchers asked 43 drivers to plan routes between locations across London, mirroring the skills required for The Knowledge examination. The drivers’ response times and strategies were studied, revealing a highly efficient method of cognitive planning that outperforms current AI systems in dealing with complex environments.
Dr Pablo Fernandez Velasco, a visiting research fellow at UCL, explained: “London is incredibly complex, so planning a journey in a car ‘off the top of your head’ and at speed is a remarkable achievement. Instead of sequential planning, drivers consider the entire network of streets and prioritise the most critical junctions first.”
The study highlights the potential benefits of integrating human-like planning strategies into AI navigation. Dr Dan McNamee, joint senior author from the Champalimaud Foundation, noted that AI algorithms could be designed to collaborate better with humans if they were informed by the way expert planners, like London’s cabbies, think and adapt to dynamic conditions.
The study builds on earlier work led by the late Professor Eleanor Maguire at UCL, which showed that London taxi drivers have a larger posterior hippocampus – the brain region linked to memory and navigation – compared to the average person. This structural change is believed to be a result of their rigorous training and daily navigation challenges.
The researchers also hope this knowledge could contribute to Alzheimer’s disease research. The hippocampus is known to be affected in Alzheimer’s, and the Taxi Brains project, led by Professor Hugo Spiers at UCL, is taking MRI scans of taxi drivers to study the relationship between navigational skills and brain health.
With support from the British Academy, EPSRC UK, and Ordnance Survey, the research underscores the significance of human expertise in shaping the next generation of intelligent systems. London’s cabbies, it seems, are not just steering the streets but driving innovation.
