Today I was reading Pierre’s paper about grounding in Multi-modal Task-Oriented Collaboration and I was thinking about the usage of the word beliefs that is often used to refer to common ground. The noun “belief” has 2 senses in WordNet.
1. belief — (any cognitive content held as true)
2. impression, feeling, belief, notion, opinion — (a vague idea in which some confidence is placed; “his impression of her was favorable”; “what are your feelings about the crisis?”; “it strengthened my belief in his sincerity”; “I had a feeling that she was lying”)
Therefore my consideration is that a belief is evanescent, because it is retained in the person cognitive schemas and expressed only in connection with impressions or feelings that do not account for a more concrete validity [I still have to find a proper reference for this]. My opinion is then to look for something more concrete of belief to define grounding or the grounding criterion, like acceptance or commitment. These two last concepts hold a more concrete effect on the persons actions and therefore are more observable than belief. This consideration I am raising here seems to be supported by two studies of B. diEugenio  and M. Backer .
A possible outline for my contribution may be>
Title_ From Belief to Acceptance/Commitment in Computational Models of Grounding
1. Introduction: Importance of Computational Models. Some of the most recent models. Pros and Cons of these. Need for empirical evidence.
2. Comparison of different Models
3. Limitations and convergences
4. Empirical approaches