Robotics Reference
In-Depth Information
Given that it is highly desirable for robots to be able to express their
emotions, an important question is, how should a robot decide which
emotion or emotions to display and when? Joseph Bates suggests as
a guideline that “the thought process reveals the feeling”, that robots
should emotionalise and display their thoughts. This in turn enables us
to see that emotionally aware and demonstrative robots really do “care”
about what happens in the world and that they appear to have genuine
desires. Bates also proposes that a robot's emotions should be “clearly
expressed” in order for us to relate to the robot and understand its ac-
tions. Furthermore, he recommends that a robot's emotions should be
accentuated, to convey them better to those around it, and appropriately
timed—the time relationship between an emotion and the situation or
action giving rise to it is what helps people and other robots to under-
stand which emotion is being expressed. If a robot were to exhibit happi-
ness during a fight or if it were to shout during a peaceful moment when
its owner was listening to quiet music, that robot would create confusion
in our minds and reduce our empathy for it.
Lynellen Perry at Science Applications International Corporation has
raised some interesting questions about robot displays of emotion. She
asks whether some robots will become unstable just like humans. Will
we need to send some robots to a professional psychiatrist or will we just
destroy those robots and try again? On the other hand, as she points out,
it might actually be useful for society to have some robots that have split
personalities or psychoses, so that psychologists can use them as research
tools to help understand and deal with these disorders. Another question
asked by Perry is “What happens when a robot behaves irrationally before
a correct diagnosis is made of its emotional disorder?” It could
. . . make a huge mess of things. We are quite comfortable blaming
a human for the acts they commit, but who will be blamed when
a psychotic robot commits a crime? The robot? The programmers
who wrote the software in the robot? [7]
This question is discussed in a later chapter.
11
Robot Personality
A believable robot must have personality. It is widely accepted by psy-
chologists that people prefer to interact with others who are similar in
11
See the section “The Legal Rights of Robots” in Chapter 13.
