In a laboratory experiment, under tightly controlled conditions, treatment of the muscles of facial expression with BOTOX® [or other fomulations of BTX-A like Dysport® or XEOMIN®] can [to a limited extent] modulate either the expression or the perception of certain emotions.
This modulation under laboratory conditions can be a positive or negative thing, depending on the experimental subject.
In the real world, it is common for people being properly treated with BOTOX® [as opposed to being grossly over treated] to note that after BOTOX® they are able to express themselves more accurately because they are not making unwanted or excessive facial expressions which could give others the wrong impresson.
For example, a person who frowns when they are simply concentrating might give the wrong impression that they were unhappy or concerned. After softening of the frowning by treatment with BOTOX®, that individual's emotional state would be more accurately percieved by others.
In a similar manner, someone who excessively mirrors the emotional state of others may find that relaxation of the muscles of facial expression reduces their "mirroring" to a more normal level, and makes it easier for them to accurately assess the emotional state of those around them.
I have a book chapter in press on this subject.