Page 203 - Socially Intelligent Agents Creating Relationships with Computers and Robots

186 Socially Intelligent Agents

agents (as computed by the plan-relevance criteria proposed by desJardins and
Wolverton, 1998). Plans may be interdependent in the sense that one depends
on effects produced by another.
Emotional State: The social layer incorporates a model of emotional rea-
soning, Emile, that derives an emotional state from syntactic properties of an
agent’s plans in memory [1]. Emile incorporates a view of emotions as a form
of plan evaluation, relating events to an agent’s current goals (c.f., [4]). Emile
computes an agent’s overall state, tracks emotions arising from a specific plan,
and makes inferences about the emotional state of other agents (given an un-
derstanding of their goals and plans). Emotional state is represented as a real-
valued vector representing the intensities of different emotional states (Fear,
Joy, etc.) and Emile dynamically modifies this state based on the current world
situation and the state of plans in memory.
Static State: Static social state components describe features of an agent that
are invariant in the course of a simulation. These components can be arbitrary
and act simply as conditions to be tested by the social control program. One
can manipulate an agent’s top level goals, its social status, its etiquette (its
sensitivity to certain social cues), its independence (is it willing to construct
plans that depend on the activities of other agents), and characteristics of its
relationship with other agents (friendly, adversarial, rude, deferential, etc.).

3.2 Control Primitives
Control primitives are social-level actions and consist of communicative and
plan-control primitives.
Communicative Primitives: The social layer defines a set of speech acts that
an agent may use to communicate with other agents. As they are defined at
the meta-level, they can operate on plans only as an atomic structure and can-
not make reference to components of a plan (although one has the option of
breaking a plan into explicit sub-plans). Some speech acts serve to communi-
cate plans (one can INFORM another agent of one plans, REQUEST that they
accept some plan of activity, etc.). Other speech acts serve to change the state
of some previously communicated plan (one can state that some plan is under
revision, that a plan is acceptable, that it should be forgotten, etc.).
Planning Primitives: Planning primitives alter base-level planning behavior.
Classical planning algorithms can be viewed as a sequential decision process:
critiquing routines identify problems with the current plan and propose a set
of changes that resolve at least one of these problems (e.g. add an action); a
change is applied and the process continues. Planning primitives act by con-
straining the set of viable changes. Recall that from the perspective of the plan-
ning algorithm, all activities are represented in a single task network (whether
they belong to the agent or represent the activities of other entities). One set