Ah, so you're saying this contrast we use has sociological influences, then. That makes more sense than calling this contrast an illusion. I would not be surprised, however, if it had biological roots, further exaggerated by social influences. I understand that whether we perceive a face to be male or female can be an illusion, but I see no evidence that can indicate men and women's faces are not visually different. In other words, I see no reason to believe gender role is the only thing at work, here.