"Helmet use may deter people from cycling", that doesn't affect whether or not you should wear one.
"Benefits of biking out way the risks", you don't lose the benefits if you cycle with a helmet.
"However, if potential cyclists see everyone else in their community wearing helmets while riding a bike, it communicates that biking is a dangerous activity that requires special protective gear."
One of the author's reasons for not wearing a helmet is the image that wearing a helmet projects. The personal decision to wear or not wear a helmet affects the perception of non-cyclists, which in turn affects the number of cyclists. Or at least so goes the theory.
As an aside, since a lot of people seem to not have noticed and assumed otherwise by default: The author is female.
The best use case for bicycle is not a 20-mile ride to work or on muddy trails, but a 1-mile ride to a local coffeeshop. And for a competent cyclist on flat roads in normal weather, a helmet is plain unnecessary.
In concluding that studies about the efficacy of helmet use are unreliable, the author claims "The people who wear helmets are likely more safety-conscious than those who don’t, which makes comparing the two groups very difficult and will make it appear that helmets are more protective than they actually are."
But then, only a few sentences later, the author's agenda shifts to arguing that wearing a helmet causes other drivers to incorrectly conclude you are a safer cyclist and give you less space: "Just because someone wears a helmet doesn’t mean they’re a safer cyclist. It seems like a lot of people use helmet use as a proxy for caring about safety, and that’s just not true."
I can't see how both of these things could be true. Either the studies are not confounded as the author claims, or drivers giving cyclists less space is not the problem he thinks it is because those cyclists are actually safer cyclists. This is not to say that one or both of these arguments are totally incorrect -- there is probably a grain of truth to both of them -- but I suspect at least one of these claims is exaggerated somewhat.
If the author's argument were that localities should not mandate helmet use, I think I would agree. But it wouldn't be because of his tendentious arguments about helmet-wearing signalling that cycling is dangerous, or that drivers take more risks around helmeted cyclists. It is very very hard to believe that either of these dynamics, if they exist at all, outweigh the undisputed benefit of wearing a helmet in preventing serious head injuries. Rather, I think localities shouldn't mandate helmet use simply because helmets discourage cycling and, on balance, cycling is a very safe activity with or without a helmet, and its social benefits outweigh the risks of cycling without a helmet. (An argument which, to be fair, the author also makes.)
I never wear spandex, almost never wear a helmet, but I always have lights and mirrors, and I never, never run a red light (unless it is sensor-driven and won't change for me). Plenty of spandex'ed and helmeted cyclists will run a red light after checking to make sure there's no oncoming traffic. Yet drivers are almost certain to assume that the spandex'ed and helmeted cyclist is more safety-conscious than me. Even if helmets are much lower down on the list of what keeps cyclists safe than traffic norms and behavior, road choice (arterial vs. nonarterial), lights, mirrors, &c.
Meanwhile passing too closely is very dangerous. =/
mm, I think you missed something here -- It is always a bad thing for a car to give a cyclist less space. Even if a cyclist wearing a helmet is more likely to be a safer cyclist, giving that cyclist less space is dangerous. It doesn't matter if the cyclist is a safer rider than most -- what matters is that the driver may _think_ the cyclist is a safe rider because of the helmet, and so may give the cyclist less space.
It does, indirectly (you may think that more people biking will make you safer in the long run for example).
Of course, there is not zero marginal impact on cycling. Helmet advocates never seem to try to estimate what cost there is to cycling rates.
I suspect that even among committed cyclists, there is a marginal cost in the number and frequency of trips by bicycle to wearing a helmet, and this cost is much greater for more marginal cyclists.