Emotions: a computational semiotics perspective

1
Emotions a computational semiotics perspective
Rodrigo Gonçalves, Ricardo Gudwin, Fernando
Gomide Electrical and Computer Engineering
School (FEEC) State University of Campinas
(UNICAMP) CP.6101 - CEP 13083-970 Capinas, SP,
Brasil
2
MEQ (Machine Emotion Quotient) Evolution
MEQ
?
Semiotics in AI
AI/Cybernetics Soft computing
Damasio's book
80's
90's
The beginning - big dreams leaded to big
disappointments
Cognitive representations of emotions instead
emotions itself
Emotions not as a psychological state of the soul
- a physical phenomena
3
Mind Body

• Our mind is physically associated with our body
  through our brain

Mind model
Mindbody model

¹
Body model
Intelligent agent
4
Imagetic and distributed nature of thinking

• There is no centralized big screen where our
  thoughts are projected.
• We process each kind of sense in physically
  different and distributed location.
• Two different types of mental images
• Perceptual - generated by our sensors or by
  dispositive images
• Dispositive - prototypical image that hold rules
  to reconstruct perceptual images.

Perception
Perceptual images
Learning
Backfire
Dispositive images
5
Emotions (I)

• Feelings ¹ Emotions
• Emotions - dispositive image that affects the
  body internal state in response to perceptive
  images
• Feeling - the perception that the body state has
  changed

• Two basic categories of emotions
• Primary
• innate
• often related to self-preservation and
  reproduction.
• Unconsciousness level
• Secondary
• not innate (learning)
• (Un)consciousness level

6
Emotions (II)

• Emotion is able to change the body internal state
  and consequently affects how the brain process
  others mental images
• changing the performance of the cognitive
  mechanism
• attributing an intuition to another mental image
  (somatic mark)

• Intuition Somatic Mark
• a value of desirability (an apraisive knowledge)
  attributed to any mental image in an
  unconsciousness level mechanism called somatic
  marker
• Somatic marker
• continuously analyses the body state and mental
  images calculating and attributing a desirability
  value based on basic instincts.

7
Emotions x Reason

• Instincts (reactive responses) are dispositive
  images that generate behavior and/or emotions
• characteristics of the species
• related to auto-preservation
• Reasoning might be considered as an instinctive
  process.

• Reason uses the intuition and consequently the
  emotion mechanism
• drastically decrease the search space for complex
  problems

8
Somatic agent

• Six independent and unsynchronized modules
  communicating through a blackboard-like memory
• sensing
• actuator
• body modeler
• somatic marker
• rational processor
• dispositive memory
• Communication through a blackboard-like memory
• mental images as elemental communication data
  unit
• modules are both image producers and consumers
• all messages (mental images) are posted in the
  working memory and all modules may access it

Work memory
Image producers consumers
9
Mental image

• Elementary communication data unit in somatic
  agents
• Composite knowledge
• set of knowledge units
• the meaning of the set is different of the sum of
  meaning of each part
• can be classified into the elementary taxonomy
  (based on the semantics of the hole set of
  knowledge)

10
Somatic agent hierarchy

• Object oriented structure based on the knowledge
  taxonomy
• allows a blackboard-like implementation
• Attributes
• type
• desirability
• time stamp
• mean life
• data
• Relations
• created_by
• created_from
• consequence pointer
• Methods
• compareTo

11
Mental image hierarchy

• Follows the knowledge taxonomy
• Object
• Sensorial
• Occurrence

12
Somatic agent image generators (and consumers)

• Sensor
• Body modeler
• Somatic marker
• Dispositive memory
• Actuator
• Rational processor

13
Sensor
Body modeler

• Creates perceptive sensorial images
• based on data obtained in the external world and
  posts them into the working memory
• Read concatenate perceptive sensorial images
  from working memory
• concatenate and create another sensorial image
  with higher level sensorial data
• Monitors the system body

• Body model
• perceptive image that holds a rhematic object
  specific knowledge
• Body modeler
• reads sensorial images in the working memory,
  processes it, and actualizes the body model image

Actuator

• Capture mental images with prescriptive content
  in the working memory and use them to act in the
  external world

14
Somatic marker

• Calculate a degree of desirability to every
  mental image in the system
• Used by the rational processor as some kind of
  intuition about an image and is calculated using
  innate rules or by image similarity (using
  compareTo method)
• It is not an emotions. It is only a judge value
  given to an image based on the somatic state and
  innate knowledge
• Unconscious level
• the rational processor of the somatic agent does
  not have any control over it and it accesses any
  mental image produced in the system

15
Dispositive imageDispositive memory

• Dispositive image
• similar to perceptual image. It holds generic
  knowledge instead of a specific one
• it might be triggered generating perceptual
  images (associative memory)
• Two types
• ordinary
• emotional ? emotions!
• Dispositive memory
• Module that holds a collection of dispositive
  images

16
Rational processor

• Conscious behavior generation
• Higher levels behavior. Most of lower level
  behaviors, like reactive ones, are responsibility
  of the dispositive memory module
• Semiotic cycle
• Consumes images performs abduction, induction and
  deduction over them, creating new images
• Integrated with emotions mechanism
• the implementation of the rational processor
  should consider the desirability value calculated
  by the somatic marker mechanism

17
Somatic Agent

• Might be integrated using Object Networks

18
Conclusions

• The emotion must be seen not as a heuristic that
  leads to an optimal solution to any problem but
  as a process, that turns complex tasks possible.
  In this paper we show how the real concept of
  emotions can be captured and implemented. For
  that, we used some concepts of computational
  semiotics and Damasios theory of emotions.
• Currently we are working in an application
  example that will be a subject for future
  publications.