There isn't enough data to say with any certainty. If you want my opinion, which isn't worth much, I think helmets probably do offer some protection against bruising and lacerations in low-speed falls when the unhelmeted head would have made contact with the ground or whatever. It's one reason why I am in favour of helmets for young children learning to ride, as they are the most likley victims of such falls and I wouldn't want them put off cycling by unpleasant, if not life-threatening, experiences.

As far as adults are concerned, it is much more complex. In the first place I am quite unlikely to fall off my bike at low speed with no other vehicle involved, and If I did I'd be quite unlikely to hit my head. But leave that aside for a moment. I'm much more likely to come off when I, or the other vehicle, is moving reasonably quickly. My concern is more for my brain than for my scalp, and given that brain damage generally results from the shearing effect of rotational forces - the brain moving around within the skull and severing blood vessels etc - it may actually be better to suffer the superficial damage because that damage protects my brain. What I mean by that is that if my head glances off the ground at speed, what is likely to happen is that my scalp will tear and slide across my skull. Nasty, but much better than rapid rotation of the skull. A helmet as currently designed isn't going to do that, the friction between it and the ground is quite likely to cause my head to rotate and increase the risk of trauma to my brain. So in this instance yes, the helmet may offer protection against the superficial injury but at the expense of increasing the risk of something much worse.