Following the recent entry here on bike bell usage there was discussion about the turn-and-look phenomenon as an undesirable outcome of ringing a warning bell.
What is the turn-and-look phenomenon? As pedaller put it:
when I have used a bell, people usually turn to look toward the sound and their feet follow, this means they actually cross across my path more than they were before
So just how prevalent is the turn-and-look? And just how dangerous is it? Over the last two days I did a small field experiment: I simply counted the number of turn-and-look "events" I encountered on my daily commute.
Method
- ring the bell at everyone
- count every pedestrian (or group of pedestrians) overtaken from behind on shared pathways only
- of these, count the times where the pedestrian (or one in a group) turns to look around (turning includes any rotational movement of the head in response to the sound of the bell)
- don't count people who are stationary or not on the same track (as their turn-and-look doesn't pose any danger)
Results
| Day | Trip | Turn-and-Look | Total overtaken |
|---|---|---|---|
| Tuesday | Morning | 2 (20%) | 10 |
| Evening | 6 (35%) | 17 | |
| Wednesday | Morning | 2 (18%) | 11 |
| Evening | 4 ( 9%) | 43 | |
TOTAL |
14 (17%) |
81 |
Discussion
Of the 81 pedestrians I overtook, 14 (or 17%) did some kind of turn-and-look movement. I didn't really state my hypothesis at the outset but in thinking about it, I would have expected it to be lower - more like 10%, so that's a bit of a surprise.
I also paid attention to the number of turn-and-look events that actually did interfere with my passage and over the two days and 14 turn-and-look events, only 3 (or 21%) of those people showed any inadvertent tendency to make it harder for me to get past. So on the basis of these figures, for any pedestrian I pass I can expect a 17% chance that they will turn-and-look but most of those will be benign. There is only a 3% (17% × 21%) chance that they will turn-and-look and interfere with my line.
On the face of it, time of day doesn't seem to be a factor as the lowest and highest rates of turn-and-look both happened on evening runs (9% on Wednesday evening, 35% on Tuesday evening). However the Tuesday run was over an hour later in the day than the Wednesday run so it may be possible that walkers who get out onto the trails earlier in the evening are more bike savvy than those who come out after dinner.
To get more subjective, the general impression I have is that a turn-and-look was more likely to lead to the pedestrian edging off to the left-hand side (thus providing more space) than to interfere with the line of traffic.
Conclusions
While turn-and-look events are common there is a very low likelihood (3%) that a pedestrian, on hearing a bicycle bell, will react in a manner that impedes the line of the bicycle and endanger themselves and/or the cyclist.
So cyclists should ring their bells with alacrity because the risk of tangling with someone who does a turn-and-look is relatively minor. In other words, the turn-and-look is a poor reason to not ring your bell.
OK, this ain't science and the numbers are pretty dodgy, so interpret these "findings" how ever you like. They may only apply to the route I ride in Melbourne. Hell, they probably only apply to the four rides I did in the last two days!
How's that for a pointless exercise?
Comments
Good for you Treadly, there's nothing to compare to the scientific method. I guess we now need to enlist the help of lots of cyclists on different shared path routes and at different times in order to reach some sort of consensus on this phenomenon. I guess I shall have to bite the bullet (bike handlebars) and even venture back onto shared paths in order to add my two-cents-worth to the discussion (and yes, I do try to avoid shared paths as much as possible especially in the mornings, evenings and weekends when they are filled with pedestrians, walkers, joggers, runners, dog-walkers and recreational cyclists especially those with small children).
Fantastic study...
Next factor to analyse, in my view the most important... Time given between anticipated overtaking and cyclists bell ringing.
My hypothesis: Many cyclists ding too late, and turn and look becomes a problem. one or two bad experiences early on in a bell dinging career teach cyclists not to ding, for fear of becoming just as bad as the angry horney driver that we all know and love.
Hypothesis part b: Bicycle bells available in Australia are tinny and crummy, and don't have the volume or traditional bicycle bell sound that people recognise (resulting in more turn and looks)
My daily solution, get an Indian bicycle bell from the Indian bicycle shop in StKilda (cnr Barkley and Grey St for 0) or some similar Asian bell from China or Indonesia and then ring your bell much further ahead of overtaking time.
People can turn and look as much as they want, and the figure everything out before you get there..
This new behavour has resulted in a massive increase in the number of thank-you's I receive from pedestrians (amazing, I am humbled).
Ring Early, Ring Often, Ring Again.
Fantastic analysis, this is great data to have, thanks a lot. I'm new to using the bell and I like to warn people before overtaking them and I always use it when I'm turning in a corner where I don't see too far ahead because of the vegetation, or next to fences.
Actually, like anything, people learn. I think the same study 5 years ago would have had a different result, 10 years ago, different again. I think pedestrians in general seem to be learning the 'right' behaviour upon hearing a bell. Reminds me of this lovely video: http://www.youtube.com/watch?v=O_MphtzCOEc
Dude, this was done 5 years ago! [Scary that the blog is still here after all this time, huh?] But I take your point.
Nowadays, I agree, pedestrians are probably even more used to cycle traffic, especially on the busier trails on which the above numbers were recorded.