External Adaptors in Language

Can expression bridge between cultural processes and a neural world? Since realtime processes elude today's technology, I begin with how expressive dynamics affect behaviour. To separate neurally mediated evidence from reports, I trace nontrivial causal spread to interactional and, later, neural time-scales. Using examples, I show that sense making uses patterns to stabilise dynamics around semantic values. Then, turning to human symbol grounding, I argue that infants act in parallel ways. In this case, however, they use what Trevarthen (1998) calls IMF. Processes within the brain change infant motivations by attuning their activity to rewards that are likely in their social environment.

Kravchenko (2007) sketches a useful view of micro-scale language activity. In his biological theory, construals use wordings, together with reciprocal causation and self-organization ('intentionality'). Continuously changing dynamics mesh each person's initiatives with the other's responding. Cognition fuses with communication as events gain an experiential feel. In real-time, an actor's orientation affects the observer who, using experience-based sensitivity, may modify (or inhibit) expression. Interaction thus cascades as perceived orientations (and non-orientations) create a domain that is animated by vocal and visible patterns. As this environment changes, we hear how signs are meant. Thus, fused activity promotes coordination between embrained bodies. Orientations and reorientations mesh with both what we will come to call words and attitudes that echo past events. Over a history of interactions, a shared history accumulates. Using wordings and expression, connotations shape firstperson phenomenology.

To apply the organic process model to language, I begin with a classic study by John Gumperz (1982; see also, Eerdmans et al., 2003). In this work he reports on success in improving customer response to South Asian staff. Initially, customers were often negatively affected by simple offers of items like 'gravy'. While out-group members heard the utterances as unfriendly, South Asians regarded them as neutral. Miscommunication, Gumperz suggested, drew on contextualization cues (CCs). Thus, a falling cadence on gravy evoked 'presuppositions' that differed between groups. In short, different representational content was triggered by a single physical event. While used to change behaviour, the theory was not successful (Eerdmans et al., 2003).8 Offering an explanation Cowley (2006) argues that we report - not hearings of objective patterns - but our own felt responses. These are part of the feeling of what happens (Damasio, 1999) or core consciousness. To report talk as neutral or unfriendly is to use connotations to say how dynamics are for us.9 Like an adaptor, gravy manufactures a felt relation (semantic synthesis).

In Barbieri's terms, we hear a prosodic artefact. While nothing is known about the neural processes, this links the behavioural observations to the practical work. Staff used the CC label in a trick that works to their advantage. Having been given a reason to offer gravy (and other items) on a rising cadence, they changed their behaviour. Holding the idea of CC in mind, they spoke offers on a rise which, to English ears, sounded friendly. Using the language stance (or higher level analysis), they altered how they spoke. They changed the dynamics (Maturana's (1978) 'intentionality') that manufacture semantic syntheses or felt responses. Contextualization does not,

8 In Eerdemans et al. (2003) collection, it is rejected by Prevignano, Levinson, Thibault (see Cowley, 2006). While many views are possible, contextualization lacks any specifiable physical parameters.

9 This argument is influenced by Dennett's (1991a) discussion of the phi phenomenon (see Cowley and Love, 2006). In reporting that we see a green light turn red, he argues, we report a judgement.

therefore, depend on cues. Rather it is strategic coordination and, quite clearly, voice dynamics have cognitive powers.

Even if we report felt responses, this throws no light on neural processes. To consider whether these use organic coding, I use how R^czaszek et al. (1999) apply dynamical systems theory to meaning construal. In experimental work, they test the hypothesis that, in construing utterances, decision making tends to stabilise on single judgements. Thus, in one study, prosodic variables (the pause and vowel duration of rhythmic feet) are manipulated to test judgements of an ambiguous sentence. Specifically, the focus falls on construals of the wording:

(1) Pat or Kate and Bob will come.

Subjects were presented with utterances whose foot length varied such that, at one extreme they would be heard in meaning 1 and, at the other in meaning 2. In one experimental condition sentences were presented in either ascending/ descending fashion and, in the other, in random order. Using a reaction-time technique, they assigned one of two senses to what they hear (Either [Pat or Kate] and Bob will come OR Pat or [Kate and Bob] will come.) Overall, subjects tended to show hysteresis. All things being equal, construals of (phonetically) identical utterances will be influenced by recent events. Having just synthesised a percept, we are more likely to hear the same percept again. The figure shows conditional probabilities of assigning the same construals in a subject who shows this pattern (Fig. 1).

Construal is affected by history as well as the physical stimulus. It draws on wordings, the physical event and what are termed the dynamics of meaning. Judgements fall into attractor basins depending on, among other things, a previous event. In further work, reaction times to the same stimuli were found to be longer

Results for subject Q:

Hearing type (a) [Pat or Kate] and Bob will come after either (a) the same (♦) or type (b) the alternative (□) construal

100 90 80 70 60 50 40 30 20 10

♦ : probability of unchanged construal; □ = probability of switch in construal.

Duration in foot length (9 steps)

Fig. 1 Hysteresis in meaning construal when participants were about to switch their construals than in cases where this did not occur. Even in experimental conditions, percepts integrate verbal constraints, echoes in working memory and voice dynamics. Brains, it seems, seek out stability. Sentence construal integrates speech timing with what has been heard. If this happens in experimental conditions, internal conventions are likely to be even less powerful in the wild. Further, effects vary between people. While subjects show internal consistency, there are huge individual differences. Finally, R^czaszek et al. (1999) show that, when we use visual context, construal settles in a 350-400ms time-window. Strikingly, this is long enough to be conscious of what is seen.

As for the customers at the canteen, utterances set off events that stabilise on a pattern. In settling on a sense, participants exploit dual recognition. While hearing wordings, orientation (or reorientation), they also use prosody. As R^czaszek et al. (1999) emphasise, symbols constrain dynamical events (c.f. Pattee, 2000). Using dual recognition, dynamics are integrated with a heard pattern. Organic coding thus offers a model for how we use connotations. Neural 'openness' uses recent experience and timing to integrate recognition that manufactures cognitive-communicative fusion. While akin to Cowley's (2006) felt response, in the work of R^czaszek et al. (1999), the parallels with organic coding apply to neural functioning. The percepts that are identified become manufactured syntheses that arise as artefacts (utterances) prompt brains to fill out incomplete information.

Wordings are construed together with accompanying prosodic and other information. Far from using constructed codes, human sense making integrates wordings with what we hear, expectations, physical events and, indeed, contents of working memory. As described by the organic process model, established routines fill out incomplete information in real-time. Thus in both construing interactional attitudes and reacting to sentence meaning, we find a parallel with protein synthesis. Next, therefore, I extend the comparison to the developmental time-scale. I ask how coaction transforms neural systems which control infant motive formation. The focus falls, first, on how caregiver expression influences baby's percepts. Second, I ask how this change contributes to cultural learning.

Was this article helpful?

0 0

Post a comment