Editing Our Speech

Because the process of speaking involves so many complex processes, we’re bound to mess up occasionally. But our brains actually have a sort of editing process in place to correct our mistakes. Psycholinguist Willem Levelt discusses how speech perception interacts with speech production; he states that the type of feedback our brain receives involves:

Am I saying what I meant to say?
Is this the way I meant to say it?
Is what I am saying socially appropriate?
Am I selecting the right words?
Am I using the right syntax and morphology?
Am I making any phonological errors?
Is my articulation at the right speed and pitch?

He postulates that, “The major feature of editor theories is that production results are fed back through a device that is external to the production system. This device can be distributed in the sense that it can check in-between results at different levels of processing. The editor may, for instance, monitor the construction of the preverbal message, the appropriateness of lexical access, the well-formedness of syntax, or the flawlessness of phonological-form access.”

While I agree that editing monitors all levels of speech—when I’m speaking, I may realize I’ve pronounced a word wrong, or chosen the wrong word, or phrased something the wrong way (all of which involve different levels of the speech production process)—I don’t really understand the idea of this device as something external. Does he mean it is physically separate in the brain from our language production center? Or chronologically separate (it gets involved only after we have begun to speak)? Or is the device simply a neurologically different process from that of the speech production process?

Researcher Claudy Oomen explains that Levelt saw speech production as comprised of three stages—conceptualization, formulation, and articulation (which is basically a less compartmentalized view than the speech production levels described in the previous blog)—and saw the monitoring mechanism as proceeding through perception, so speakers can detect their own speech errors “by parsing their inner and overt speech in the same way.” That is, the mechanism is active in correction both before and after articulation. Levelt presents the idea of a “flow-through monitor”—one that allows the production process to continue while the conceptual/inner loop monitoring is taking place (so errors detected before articulation can still become overt).

But another hypothesis presents the idea that monitoring is “an integral part of the production process, allowing direct access to the various processing components of speech production” (like lexical selection); this hypothesis suggests that if an error is detected during any of the stages of the speech production process, further processing is canceled while the error is repaired, all before articulation. This model includes a sensory register model for checking overt speech.

Other linguists have pursued Levelt’s idea of a feedback system; one researcher developed a technique of replaying a person's speech to that person's own ears, subject to a variable time delay. From his studies, he came to the conclusion of a multiple loop control hierarchy, with four levels of feedback:

The Thought Loop: The highest level feedback loop monitors the pragmatic appropriacy of the output.

The Word Loop: The second highest loop monitors speech production for word selection accuracy.

The Voice Loop: The third highest loop monitors speech production at whole-syllable level for morphological accuracy.

The Articulating Loop: Finally, the lowest loop monitors speech production checking that the right phonemes have been used within each syllable.

This "loop" system seems logical; it corresponds to the stages proposed in the speech production model. Although it's important to remember that the speech production model itself is just that--a model. As Lera noted in class, it is only one way of explaining our language system. And in terms of the editing mechanisms, it’s interesting to see the different models proposed to explain our system for correcting our speech; the ones I’ve presented are only two of many! The existence of all these different hypothesized models is further testimony to the difficulty inherent in giving definitive answers to questions about how our brain processes and uses language.

http://www.springerlink.com/content/m761760977386125/fulltext.pdf

http://www.smithsrisca.demon.co.uk/speech-errors.html