Information Technology Reference
In-Depth Information
is. So, neuroticism is implemented as the variation factor of the emotional state. Thus,
the emotional state of a neurotic agent will vary more suddenly than a stable one when
actions are applied, as described later.
Conscientiousness includes elements such as self-discipline, carefulness, thorough-
ness, organization, deliberation and need for recognition. We implement conscientious-
ness as a factor in the decrements of the drives due to action executions, representing
how meticulous the agent is in carrying out the action. Thus, an agent with a high value
of conscientiousness gets a bigger effect when applying actions (a bigger decrease of
the involved drive). But, similarly, the other drives will also increase proportionately to
the conscientiousness value as time passes. The conscientiousness value also influences
the duration of the actions performed by agents. For instance, the actions performed by
a meticulous agent take more time than the ones performed by a careless agent.
The last two factors, extraversion and agreeableness, are related to social interaction.
Thus, they will be used in future versions of the system that include multiple agents
and interactions among them. Personality traits are represented in the domain through
functions.
3.4
Emotional State
The agents emotional state is determined by two components: valence and arousal.
Valence represents whether the emotional state of the individual is positive or negative
and to which degree. Arousal represents the bodily activation or agitation. We represent
them in the domain as PDDL functions. Since we want to obtain plans that maximize
the valence, we have to define the planning problems metric accordingly. Even if PDDL
allows generic functions to be defined as metrics, most current planners can only deal
with metrics that are defined over minimizing an increasingly monotonous function (no
action can have an effect that decreases its value), since metrics are considered in PDDL
as costs and each action has an associated cost.
In our model, objects used in the actions can cause valence both to increase (when
the agent likes the object) or decrease (when it does not like it). Therefore, it is not
possible to use the valence directly as the problem metric. Instead, we define an in-
creasingly monotonous function,
v-valence
, that the planner tries to minimize. Each
action increases
v-valence
, with positives values between 0 and 10 depending on the
preference for the object used, in the following amount:
n
n
max
)
×
(
p
max
−
(
p
a
+
p
o
)
2
Δv
=(
)
where
v
is the value of
v-valence
,
n
the agent neuroticism,
n
max
the maximum
possible value for neuroticism,
p
max
the maximum possible value for a preference,
p
a
the agent preference for the executed action and
p
o
the agent preference for the used
object. In case the object is new to the agent,
p
o
=-1 and we replace
p
o
for the value of
the agent openness. Thus, this value can be used as a metric alone or combined with
others such as the duration of the plan.
