Spoken word production: a theory of lexical access

Abstract

A core operation in speech production is the preparation of words from a semantic base. The theory of lexical access reviewed in this article covers a sequence of processing stages beginning with the speaker's focusing on a target concept and ending with the initiation of articulation. The initial stages of preparation are concerned with lexical selection, which is zooming in on the appropriate lexical item in the mental lexicon. The following stages concern form encoding, i.e., retrieving a word's morphemic phonological codes, syllabifying the word, and accessing the corresponding articulatory gestures. The theory is based on chronometric measurements of spoken word production, obtained, for instance, in picture-naming tasks. The theory is largely computationally implemented. It provides a handle on the analysis of multiword utterance production as well as a guide to the analysis and design of neuroimaging studies of spoken utterance production.

Figure 1

Serial two-system architecture of the theory: two stages of lexical selection followed by three stages of form encoding.

Figure 2

Fragment of the WEAVER++ lexical selection network. (Upper stratum) Lexical concept nodes. (Lower stratum) Lemma and gender nodes.

Figure 3

Fragment of the WEAVER++ form encoding network (Left) with corresponding form-processing stages (Right). (Upper stratum) Nodes representing morphemic phonological codes and their phonemic “spellouts.” (Lower stratum) Nodes representing syllabic articulatory scores.

Figure 4

(A) Experimental visual pattern to be described. Pattern is flat on the table in front of the subject. Description starts from the arrow. On the outside, directional terms used in “intrinsic” perspective. On the inside, terms used in ”relative” perspective. (B) A typical subject's drawing of the same pattern from listening to an intrinsic description.

Figure 5

Effects of a semantic distracter on picture-naming (A) and picture categorization latencies (B) for different SOAs of picture and visual distracter words. The semantic effect is the naming latency when the distracter is semantically related to the picture name minus the latency when the distracter word is unrelated to the picture name. Black squares represent data from Glaser and Düngelhoff (21). Open squares represent WEAVER++ simulations (6).

Figure 6

Additive effects of semantic competition and gender facilitation in lexical selection. The target utterance, a Dutch adjective-noun phrase, is produced in response to an English probe word. The target nouns in an experimental block are either semantically homogeneous or heterogeneous. The syntactic gender of the target nouns in a block is either homogeneous (all gender 1 or all gender 2) or heterogeneous (both genders mixed). Data is from Vigliocco et al. (28).

Figure 7

Effects of a phonological distracter on picture-naming latencies, for different SOAs between picture onset and onset of spoken distracter. All picture names and distracter words were disyllabic. (A) Distracter and target words share their first syllables. (B) Distracter and target words share their last syllables. The phonological effect is the naming latency when the distracter is phonologically related to the target name minus the naming latency when the distracter is unrelated. Black squares represent data from Meyer and Schriefers (36). Open squares represent WEAVER++ simulations (7).

Figure 8

Visual scan during a scene description. As soon as the two-object picture appears, the subject makes a saccade from the fixation point to the left object. After some scanning of that object, the gaze shifts to the right one. VT is the total duration of fixating on an object. The utterance produced here is “baby and dog.”

Figure 9

A metaanalysis of 58 neuroimaging studies of word production. Colors denote regions whose activations correspond to theoretical processing stages as indicated. Contributions of insula, subcortical regions, and cerebellum are not shown. [Reproduced with permission from ref. 64 (Copyright 2000, MIT Press, Cambridge, MA).]