Why humans find faulty robots more likeable

It has been argued that the ability of humans to recognize social signals is crucial to mastering social intelligence - but can robots learn to read human social cues and adapt or correct their own behavior accordingly?