What do speakers prepare when they prepare words in advance?

Zenzi M. Griffin
Georgia Institute of Technology

Researchers have argued that when speakers prepare an utterance, they structure a clause and decide which words (lemmas) to use before speaking. Only the processes of organizing the phonological forms of words and motor
programming are performed while speaking (e.g., Butterworth, 1989; Goldman-Eisler, 1968; Meyer, 1996). Three experiments investigated which aspects of word preparation speakers could use to modulate their speech preparation.

To identify the processes initiated prior to a response, properties of to-be-named objects varied orthogonally. The objects differed in codability, which affects speed to select a lemma. For example, controlling for other factors, an object with one dominant name (e.g., table) is named faster than an object with multiple names (television/TV). Objects' dominant names differed in frequency, which affects speed to organize their phonological forms (e.g., Jescheniak & Levelt, 1994). According to the lemma-preparation view, speakers should only select lemmas for non-initial words before speaking. Thus, response latencies were predicted to show effects of codability but not frequency. Alternatively, if speakers prepared phonological forms before speaking, latencies would also show frequency effects.

In two experiments, speakers named objects early or late within simple, repeated sentence frames. Patterns of codability and frequency effects in latencies suggested that whenever time allowed, speakers delayed initiating speech until the phonological forms of object names were prepared. They never used lemma selection alone as a criterion. In an
eyetracking study, speakers were asked to prepare descriptions of displays containing three objects. When they were ready to speak, they pressed a button that caused the display to change and then began speaking. Gaze durations on objects during the preparation phase reflected both codability and frequency, suggesting that preparing a name meant
retrieving its form. The speed, fluency, and lack of item effects in the articulation of the descriptions were consistent with advance preparation.

Taken together the experiments suggest that selecting a lemma for a noun is immediately followed by retrieving a phonological form for it (except in tip-of-the-tongue states). Earlier studies that were interpreted as showing lemma selection in the absence of phonological form retrieval will be discussed. The conclusion, consistent with studies by Wheeldon and Lahiri (1997), Ferreira (1991), and others, is that preparing words in advance at least involves preparing their phonological forms.

References

Butterworth, B. (1989). Lexical access in speech production. In W. D.

Marslen-Wilson (Ed.), Lexical representation and process (pp. 108-135). Cambridge, MA: MIT Press.

Ferreira, F. (1991). Effects of length and syntactic complexity on initiation times to prepared utterances. Journal of Memory and Language, 30, 210-233.

Goldman-Eisler, F. (1968). Psycholinguistics: Experiments in spontaneous speech. London: Academic Press.

Jescheniak, J. -D., & Levelt, W. J. M. (1994). Word frequency effects in speech production: Retrieval of syntactic information and of phonological form. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 824-843.

Meyer, A. S. (1996). Lexical access in phrase and sentence production: Results from picture-word interference experiments. Journal of Memory and Language, 35, 477-496.

Wheeldon, L., & Lahiri, A. (1997). Prosodic units in speech production. Journal of Memory and Language, 37, 356-381.