This is an interesting read.

It is well established that an individual's performance is littered with what different people say

The kind of energy around us totally affect the performance even on individual level. Motivation is essential. The uttered words sips to the head which is processed to direct how we perceive ourselves.

Future work would focal point on nonverbal expression between robot and people.

Maybe this would be a good thing, however I think taking away the conversational algorithm would kill the excitement and reduce the effectiveness of the technology. Imagine removing the verbal interactiveness in Siri and Alexa.