Academic journal article Memory & Cognition

When You Name the Pizza You Look at the Coin and the Bread: Eye Movements Reveal Semantic Activation during Word Production

Academic journal article Memory & Cognition

When You Name the Pizza You Look at the Coin and the Bread: Eye Movements Reveal Semantic Activation during Word Production

Article excerpt

Two eyetracking experiments tested for activation of category coordinate and perceptually related concepts when speakers prepare the name of an object. Speakers saw four visual objects in a 2 × 2 array and identified and named a target picture on the basis of either category (e.g., "What is the name of the musical instrument?") or visual-form (e.g., "What is the name of the circular object?") instructions. There were more fixations on visual-form competitors and category coordinate competitors than on unrelated objects during name preparation, but the increased overt attention did not affect naming latencies. The data demonstrate that eye movements are a sensitive measure of the overlap between the conceptual (including visual-form) information that is accessed in preparation for word production and the conceptual knowledge associated with visual objects. Furthermore, these results suggest that semantic activation of competitor concepts does not necessarily affect lexical selection, contrary to the predictions of lexical-selection-by-competition accounts (e.g., Levelt, Roelofs, & Meyer, 1999).

When referring to objects, speakers are influenced by the activation of concepts related to the concept they are about to put into words. If someone intends to name, for instance, a pizza, the conceptual representations of other members of the food category (e.g., BREAD) also become active, and accumulated evidence has indicated that this spurious activation has repercussions for both the accuracy of language production (e.g., Fay & Cutler, 1977) and the efficiency of this process (e.g., Schriefers, Meyer, & Levelt, 1990).

Most evidence for category coordinate effects comes from experimental studies using the picture naming task (e.g., Cattell, 1886; Glaser & Glaser, 1989; Schriefers et al., 1990; Starreveld & La Heij, 1996). For instance, in pictureword interference tasks (e.g., Schriefers et al., 1990), participants are presented with a Une drawing of an object that contains a printed distractor word to be ignored. These studies have demonstrated that picture naming is slower when the distractor word is from the same conceptual category as the picture than when the distractor is unrelated.

According to many accounts of word production, the semantic interference effect reflects competition for selection at the lexical level (e.g., Belke, Meyer, & Damian, 2005; Bloem, van den Boogaard, & La Heij, 2004; Caramazza & Costa, 2000; Damian & Bowers, 2003; Hantsch, Jescheniak, & Schriefers, 2005; Levelt, Roelofs, & Meyer, 1999; Roelofs, 1992, 1993, 2001, 2003; Starreveld & La Heij, 1995, 1996; Vigliocco, Vinson, Lewis, & Garrett, 2004). For example, the WEAVER++ model (Levelt et al., 1999; Roelofs, 1997) assumes a unitary conceptual representation for each word in the lexicon (a lexical concept), and these representations are thought to be arranged in a semantic network (Collins & Loftus, 1975; Collins & Quillian, 1969). According to this model, if you decide to name the picture of a pizza with its basic-level term, the lexical concept PIZZA will be activated and in turn will activate its corresponding lexical representation (the lemma). Activation in the semantic network will spread to the concepts of other foodstuffs (e.g., BREAD), which in turn activate their corresponding lemmas. The probability of selecting the correct lemma (pizza) is assumed to be the ratio of the degree of activation ofthat lemma to the total activation of all the other lemmas (such as bread). Active alternative lemmas therefore result in competition for selection and slow down the selection process. Semantic interference results from the lemmas of semantically related distractor words (e.g., "bread") being activated by both the distractor word and the conceptual representation, whereas the lemmas of unrelated words are only activated by the distractor word itself.

Other theories of word production (e. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.