Figure 132

The relationship between overall error rate and the anticipatory proportion. The filled circles come from studies reported by Dell et al. (1997) and unfilled circles come from other studies. Adapted from Dell et al. (1997).

spreading activation provides links between speech production and other cognitive activities (e.g., word recognition: McClelland & Rumelhart, 1981).

On the negative side, the focus of spreading-activation theory is mainly on individual words or concepts, with broader issues relating to the construction of a message with its intended meaning being de-emphasised. In addition, the theory predicts the nature and number of errors produced in speech, but cannot account for the time taken to produce spoken words. Thus, it is somewhat limited in focus.

Anticipation and perseveration errors

Dell et al. (1997) developed and extended the spreading-activation theory. The starting point was the notion that most speech errors belong to the following two categories:

1. Anticipatory: sounds or words are spoken ahead of their time (e.g., "cuff of coffee" instead of "cup of coffee").

2. Perseverated: sounds or words are spoken later than they should have been (e.g., "beef needle" instead of "beef noodle").

Note that Dell et al.'s theory was concerned only with these speech errors.

The key assumption was that expert speakers are better able than non-expert speakers to plan ahead when speaking, and so a higher proportion of their speech errors will be anticipatory. Dell et al. (1997, p. 140) expressed this point as follows: "Practice enhances the activation of the present and future at the expense of the past. So, as performance gets better, perseverations become relatively less common." Thus, the activation levels of sounds and words that have already been spoken are little affected by practice. However, the increasing activation levels of present and future sounds and words increasingly prevent the past from intruding into present speech.

Dell et al. (1997) assessed the effects of practice on the anticipatory proportion. This is the proportion of total errors (anticipation + perseveration errors) that is anticipatory. In one study, the participants were given extensive practice at saying several tongue twisters (e.g., five frantic fat frogs; thirty-three throbbing thumbs). As expected, the number of errors decreased as a function of practice. However, the anticipatory proportion increased from .37 early in practice to .59 at the end of practice, which was in line with prediction.

Dell et al. (1997) argued that speech errors of all kinds are most likely when the individual speaker has not formed a coherent speech plan. In such circumstances, there will be relatively few anticipatory errors, and so the anticipatory proportion will be low. Thus, the overall error rate (anticipatory+perseverative) should correlate negatively with the anticipatory proportion. Dell et al. (1997) tested this prediction by working out the overall error rate and the anticipatory proportion for several sets of published data. The anticipatory proportion decreased from about .75 with low overall error rates to about .4 with high overall error rates (see Figure 13.2).

This theory has implications for clinical patients having problems with speech production. Patients whose speech is error prone should tend to make relatively more perseverative errors than normals. There is some supporting evidence. Schwartz et al. (1994) tested a patient, FL, who was classified as suffering from jargon aphasia (discussed later). This patient had an anticipatory proportion of .32 (compared to the normal adult figure of about .75), indicating that FL was especially prone to perseverative errors. In addition, Helm-Estabrooks, Bayles, and Bryant (1994) studied patients with aphasia (an impairment of speech understanding or production). Aphasic patients whose speech was moderately impaired made many more perseverative errors than did those with mild impairment.

Levelt's theoretical approach and WEAVER++

The main focus in this section is on the computational model WEAVER++, which was proposed by Levelt, Roelofs, and Meyer (1999a). However, many of the ideas contained within this computational model derive from Levelt (1989). The model is called WEAVER to stand for Word-form Encoding by Activation and VERification. It is based on the following major assumptions:

• There is a feed-forward activation-spreading network, meaning that activation proceeds forwards through the network but not backwards.

• There are three main levels within the network: at the highest level are nodes representing lexical concepts; at the second level are nodes representing lemmas or abstract words from the mental lexicon; at the lowest level are nodes representing word forms in terms of morphemes (basic units of meaning) and their phonemic segments.

• The network does not contain any inhibitory links.

• Speech production involves a series of processing stages which follow each other in a strictly serial way.

• Speech errors are avoided by means of a checking mechanism: "Each node has a procedure attached to it that checks whether the node, when active, links up to the appropriate active node one level up" (Levelt et al., 1999a, p. 7).

A more detailed account of the theory is given in Figure 13.3. There are basically six stages of processing:

1. Conceptual preparation: potential lexical concepts are activated on the basis of meaning.

2. Lexical selection: an abstract word or lemma is selected, together with its syntactic features; a lemma is generally selected because it is more activated than any other lemma.

3. Morphological encoding: the basic word form of the selected lemma is activated.

4. Phonological encoding: during this stage, the syllables of the word are computed: this is syllabilification.

The WEAVER++ computational model. Adapted from Levelt et al. (1999a).

Stop Anxiety Attacks

Stop Anxiety Attacks

Here's How You Could End Anxiety and Panic Attacks For Good Prevent Anxiety in Your Golden Years Without Harmful Prescription Drugs. If You Give Me 15 minutes, I Will Show You a Breakthrough That Will Change The Way You Think About Anxiety and Panic Attacks Forever! If you are still suffering because your doctor can't help you, here's some great news...!

Get My Free Ebook


Post a comment