code and the oracular

A Model of Social Robotics

leave a comment »

Social robots interact with each other through enacting “transactions“.

A transaction consists of two parts:
– an initial gambit by one bot
– a response gambit by the other bot

A gambit consists of two components:
semantic component
emotional component

Robots interpret the impact of a gambit by assessing semantic content and or emotional content and them responding as their personality and mind dictate.

Minds dictate knowledge as apparent in interpreting words and answering or posing questions etc. A mind is in simplest form a set of mappings between semantic content of initial gambit and semantic content of response gambit. A mind might specify a list of question/answer pairs for example, or a more advanced mind would have natural language question answering algorithms built in. The semantics of a gambit are wrapped by its emotion. Humans allow for many varied combinations of the two components as can be seen from the astonishing diversity in how they communicate.

Personality dictates how a bot reacts to its partner’s emotions. A personality is in simplest form a list of mappings between emotional content of initial gambit and emotional content of response gambit. Emotions act as a separate, sometimes independent channel to what is being communicated semantically. Bots with advanced personalities will also consider semantics when deciding their emotional components.

In addition to mind and personality a robot has acuity which is a measure of the effectiveness with which true semantic or emotional content is discerned from a received initial gambit. Acuity modifies mind and personality. A bot with low acuity will not tend to perceive the semantic content of its partners gambit very accurately. A bot with low acuity will misinterpret emotion similarly. Acuity will be influenced by the efficiency of the implementation of the bot, or a preset level may be chosen as part of the bots overall settings.

A game is a set of transactions, each composed of two gambits. The game proceeds in pairwise structure so that the partner’s last response gambit can be the initial gambit for the next transaction. A robot with advanced mind will also allow the wider context of the whole game to influence its gambits, to mimic the pragmatics of real human exchanges, which can have both complex emotional tone and complex semantic subject domain that each develop in an emergent way through speech and body language.


Written by Luke Dunn

October 19, 2012 at 2:29 pm

Posted in robotics

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: