Research Hit: Taxi Drivers Navigate Differently to AI

New research shows that black cab drivers prioritise information differently to AI navigation

Originally published on my substack

Well, taxi drivers aren’t AI so isn’t it obvious that they think differently

Well, yes, but the question is what is the most efficient way to navigate?

We currently use plenty of navigation tools, think google maps, and understanding how expert human navigators process information helps us understand human brain processes better but also design better AI or navigation tools.

And these weren’t your average run-of-the-mill taxi drivers but the legendary London black cab drivers who are required to know every street in London and have to pass a fiendishly hard test to qualify.

Hasn’t there already been brain research done into their brains?

Yes, precisely there have been numerous studies into the brains of London black cab drivers - famously showing increased size of the hippocampus which is responsible for memory and also navigation.

But here the focus is on how information is continually processed and in what order. AI and navigation tools use sequential processing basically taking you on a sequential route to your destination - so start off with the first street you turn onto and then the second and so on and so forth.

This is not how taxi drivers brains worked.

How did they work?

Well it appears that the have a more holistic approach: they go to a higher-order global structure of the route based on critical key junctions and hot spots. Basically using a much more comprehensive understanding off the physical structure of London and key junctions.

These key junctions are then prioritised and then the details added in after. They also take into account the length and complexity of roads.

So it is a global-key-junction-outwards principle rather than a sequential logical task.

One up for taxi drivers then

Probably and they are probably better updated with local traffic conditions, and knowledge such as roadworks, or other factors such as school closing times, and ongoing sports events.

Yup, AI can’t beat a black cab driver, yet (to be fair they haven’t really tried yet). But good to know that humans still have unique abilities.

For the moment.

Could this apply to other areas in life and business?

I would think so. I presume in many problem solving scenarios in business or general life we function with a more holistic approach such as those seen in taxi drivers: think of critical junctions, blocks, and then work out from there. That will be something for future research though.

Reference

P. Fernandez Velasco, E. Griesbauer, I.K. Brunec, J. Morley, E. Manley, D.C. McNamee, H.J. Spiers Expert navigators deploy rational complexity–based decision precaching for large-scale real-world planning Proc. Natl. Acad. Sci. U.S.A. 122 (4) e2407814122, https://doi.org/10.1073/pnas.2407814122 (2025).

Review of Previous Studies Into Cab Drivers

Griesbauer, E. M., Manley, E., Wiener, J. M., & Spiers, H. J. (2022). London taxi drivers: A review of neurocognitive studies and an exploration of how they build their cognitive map of London Hippocampus (Vol. 32, Issue 1) https://doi.org/10.1002/hipo.23395