That study
The effect of rear bicycle light configurations on drivers’ perception of cyclists’ presence and proximity was done by video projections in a driving simulator:
We recruited a sample of experienced drivers to take part in two laboratory-based experiments in which they sat in a simulated driving setup and viewed life-size real-world vi. deo simulations at daytime and at dusk.
In Experiment 1, we recorded the speed and accuracy with which the drivers detected the presence/absence of a cyclist.
In Experiment 2, the drivers’ task was to estimate the distance between their vehicle and a cyclist ahead in the road, and to rate their confidence in their decisions.
....
The participant responded by clicking the right paddle shifter on the steering wheel setup when they thought the cyclist was present and clicking the left paddle shifter when they thought the cyclist was absent. As soon as the participant responded, the trial ended, a 2-second gray screen appeared, and then the next trial began
~~~
In daytime bright sunlight, the sharp, fast flash of a taillight far down the road immediately catches my eye, even when I'm not looking for cyclists. This is way different than looking for cyclists on video. Drivers are distracted and don't pay careful attention during their typical driving situations.
I want that driver to see the flash a quarter mile ahead on a straight road, and start paying attention to their driving.
I want the driver coming around a bend to immediately think "cyclist just ahead!" from the flashing.
I ride with other cyclists that have 100, 150, 200 lumen taillights.The group sometimes gets very spread out. I instantly notice the flash when I can't even see that it's a bike on the road ahead, it's just too far away.
I aways run a two-per-second short flash tail light during the day. Flashing headlight too.