Multimodal Context Models

The pragmatic and personal representation of the social and material environment coordinates and synchronizes situation models, as well as cultural models in communicative interactions, conceived as cooperative actions. In order to meaningfully proceed in social cooperation and interpersonal communication, speakers need to take for granted that, to some extent, their representations are shared with their addresses (Givón, 2005; Tomasello, 2008). In communicative interactions, the sensation that we share goal specific and relevant situation models with our addresses relies on our subjective and unique representation of the context in which the interaction unfolds. These unique representations of the communicative interaction are defined as context models (van Dijk, 2008, 2009). The speaker’s context models include a representation of the mind of the interlocutor that may shift constantly from one utterance to the next during live communication. This cognitive and discursive process allows us to make strategic hypotheses about what our addressee knows. A language user’s context models are not only about his/her interlocutor’s epistemic (knowledge) and deontic states (intentions). Rather, they are constituted by the interplay of the following schematic categories: setting (time and place), current action and participants with their social and cognitive proprieties such as identities, goals and knowledge (van Dijk, 2008). These elements determine how we communicate the reconstructed and updated experiences in situation models. Here is a brief description of the schematic categories which constitute context models of the situation described above:

The pragmatic and communicative relevance of context models relies on the fact that they control the way in which speakers accommodate their utterances to the communicative situation (van Dijk, 2008, 2009). Context models are like other situation models (van Dijk, 2008) reconstructed and updated in episodic memory, and employed to conceptualize experiences, but they are specifically for interaction. Therefore, working memory also plays an active role in processes of updating situation models). Nonetheless, in contrast to situation models which reconstruct and represent episodes in memory and indicate what discourses are about in terms of their meanings, namely, their semantic content, context models as models of communicative interaction have a marked pragmatic nature. Context models are personal, and represent for each participant the unique definition of the current communicative situation. They enable language users to accommodate situation models in the current communicate interaction. According to van Dijk (2008), context models determine what experiences, knowledge, worldviews, emotions may or should be appropriately shared and communicated in the ongoing communicative situation.

Each new communicative interaction does not lead language users to construct completely new context models from scratch (as occurred with situation models). To build a completely new representation for each communicative interaction from scratch would generate too much cognitive effort, and thereby, such cognitive processes would not be efficient. Context models must be partially planned in advance (van Dijk, 2008). This becomes evident if we take a look at the schematic categories that constitute context models. Every time we engage in a new communicative interaction we have representations of similar communicative situations grounded in cultural models driven by socially shared knowledge and episodic memories of personally experiences of situations alike. On the other hand, context models need to be dynamic and flexible as well. The language users need to construct and update context models automatically, implicitly and subconsciously. The updating of context models is driven by working memory and event segmentation processes. However, not every perceptual change provokes a shifting of context models. These changes need to be pragmatically relevant for each language user. Obviously, a shift in setting (time and place) and number of participants engaged in the communicative interaction will automatically cause an updating of context models. Perceived shifting in epistemic (knowledge) and deontic (intentions) states of the participants, as well as, their roles and emotions also affect the online updating and reconstruction of context models. The updating processes are fundamental to represent perceptual changes in multimodal communicative interactions.In this way, as van Dijk puts it, context models are able to:

“[..] represent the experience of time and time change, the embodied feeling of location (“being there”), of perspective and orientation, the visual aspects of the communicative environment, and especially of the other participants and their relevant appearance, movements and gestures, as well as of the visual aspects of written discourse (and its drawings, schemas and pictures), the sounds and intonation of spoken discourse and its paraverbal properties (laughter, etc.), mutual touching (e.g. shaking hands, etc.), and so on.” (van Dijk, forthcoming: XX-XX).

References

Givón, T. (2005). Context as Other Minds. The pragmatics of Sociality, Cognition and Communication. Amsterdam: Benjamins.

Tomasello, M. (2008). Origins of Human Communication. New York: MIT Press.

van Dijk, T.A. (2008). Discourse and Context. A Sociocognitive Approach. Cambridge: Cambridge University Press.

van Dijk, T.A. (2009). Society and Discourse. How Context Controls Text and Talk. Cambridge: Cambridge University.

van Dijk, T.A. (forthcoming). Discourse and Knowledge.