To overcome this complexity, Dubberly, Pangaro, Pask, and others have developed models that serve as a "way of thinking... [that] involves concepts" (von Glasersfeld, 1995, p. 146) and their formation and the creation of relationships between them.
One particular framework, Gordon Pask's (1976) Conversation Theory, presents a "formalism for describing the architecture of interactions or conversations, no matter where they may arise or among what types of entities" (Pangaro, 2002). Dubberly and Pangaro (2009) have also worked to simplify Pask's (1976) theory into six main tasks that comprise the "Process of Conversation" (Dubberly & Pangaro, 2009): the opening of a channel, a commitment to engagement, the construction of meaning, evolution, a convergence on agreement, and an action or transaction. They have also worked to clarify these steps into five main "requirements for conversation," which include "[the] establish[ment] of [an] environment and mindset", "[the] use of shared language", "[an] engagement in mutually beneficial, peer-to-peer exchange", "[a] confirmation in shared mental models", and "[an] engagement in a transaction - [the] execution of cooperative actions" (Dubberly & Pangaro, 2009). Conversation Theory, Dubberly and Pangaro's "Process of Conversation" (Dubberly & Pangaro, 2009) and "requirements for conversation" (Dubberly & Pangaro, 2009) bring attention to aspects and considerations designers should consider and employ when working with artificial agents. While some of the artificial agents we see today do address a number of these requirements, one would be extremely hard pressed to present an artificial agent that addresses all of them.
Dubberly and Pangaro, "Process of Conversation"
Others researchers including Erika Hall, Paul Grice, and W. Ross Ashby have also created related models. Hall has looked at the ways interaction can be "truly conversational" (Hall, 2019, Error Tolerant, para. 3) and described the "elements of a conversation" as being the system or "a set of interconnected elements that influence one another", the interface or "a boundary across which two systems exchange information", and an interaction or "the means by which the systems influence each other" (Hall, 2019, Interactions Require Interfaces, para. 1). Grice has taken a slightly different approach and developed the Gricean Maxims which describe the characteristics of productive communication (e.g., quantity, quality, relation, manner; Grice, 1975). At the same time, Ashby has created a visual model differentiating between the "immaterial aspects" and the "physical world" to show that "actions take place in the physical world, while goals do not. Goals, the province of cybernetics, are the 'immaterial aspects' of interaction" (Dubberly & Pangaro, 2011).
Grice, "Gricean Maxims"
Together, these models have provided me with a "way of thinking" (von Glasersfeld, 1995, p. 146) about the conversations I motivate between intimate partners. This led me to ask, how could these models inform the design of artificial agents so they could negotiate the complexity of conversation?