I think that what they were getting at is that people noticing and commenting on things is more noticeable than people not doing so.

Which would imply that there's a tendency to notice the noticing more than the not noticing, and, as a consequence, over-estimate how often people do it.

It's a good point.