Sentence context and meaning frequency effects in children's processing of ambiguous words

Abstract

Nine-and twelve-year-old children named target words which were preceded by sentences ending in words having more than one meaning. Sentences biased the ambiguous word toward its dominant (more frequent) or subordinate (less frequent) meaning. Targets were related to the same meaning as that biased by the sentence, the other meaning, or were unrelated. Targets were presented 0, 300, or 700 ms following the sentence. For both ages, dominant sentences facilitated responses only to the contextually appropriate target. However, subordinate sentences led to facilitation of the appropriate meaning only for the younger group. Older children showed greater facilitation for the inappropriate (but more common) meaning. These results indicate that younger children are more sensitive to the sentence context in which an ambiguous word appears, while the processing of the older children is determined more by the relative frequencies of the words meanings. RESUMEOn observe couramment qu'un lecteur a besoin de moins de temps pour identifier un mot prtsentt en contexte que le m&me mot apparaissant avec une information sans relation avec lui. On a tgalement montrt B plusieurs reprises que cet effet de facilitation contextuelle decroit systtmatiquement avec l'sge, c'est-&dire que le contexte est plus utilist par les jeunes lecteurs que par leurs aints. La recherche prtsentte ici concerne les effets de contexte pour des mots ayant plus d'une signification. Le theme de I'ambigui'tt lexicale convient t r b bien a l'ttude des caracttristiques de la facilitation contextuelle, puisqu'il est possible de stparer l'activation de chacune des significations et d'examiner ensuite 1'Ctendue de l'information activte par le mot. Notre prtctdente recherche a montrt qu'il y a des changements lits a l'ige dans le traitement des mots ambigus. Dans cette recherche, des enfants

Negative priming from the non‐selected meaning of the homograph

“…This explanation is very parsimonious since just a single mechanism, activation, is suf cient to account for a great variety of results on lexical ambiguity, particularly those involving changes in context-based facilitation. Such an account of ambiguity resolution has proved popular in a number of areas, for example, context effects in resolution of lexical ambiguity (Coolen, Jaarsveld, & Schreuder, 1993;Love & Swinney, 1996;Paul, Kellas, Martin, & Clark, 1992;Simpson & Krueger, 1991;Simpson, Krueger, Kang, & Elofson, 1994;Tabossi, 1988); syntactic ambiguity resolution (MacDonald, 1993); the neuropsychological basis of ambiguity (Burgess & Simpson, 1988); and aphasia (Deloche, Seron, Scius, & Segui, 1987).…”

Development of Memory Structures for Homographs Using Pathfinder Network Representations

“…For example, Simpson, Krueger, Kang, and Elofson (1994), using a sentence context with long ISI, did not find this effect. That is, children's semantic priming was not greater than that of adults.…”

“…That is, children's semantic priming was not greater than that of adults. However, whereas Simpson et al (1994) used homographic words, Schwantes et al (1980) used nonhomographic words. This might explain these discrepant results, because homographic words have specific properties (e.g., more than one meaning).…”

Lexical Processing of Ambiguous Words: Dominance or Associative Strength?

“…There is a great debate, however, about the role of the context in the activation of the meaning of ambiguous words. Some researchers (e.g., Coolen, Jaarsueld, & Schreuder, 1993;Glucksberg, Kreuz, & Rho, 1986;Paul, Kellas, Martin, & Clark, 1992;Schvaneveldt, Meyer, & Becker, 1976;Simpson, 1981;Simpson & Krueger, 1991, Simpson, Krueger, Kang, & Elofson, 1994Tabossi, 1988;Van Petten & Kutas, 1987) propose that the initial activation of the ambiguous word meaning is due to the context, as the context only primes the meaning that is consistent with it, making it unnecessary to access the other meanings. Other researchers (Conrad, 1974;Holley-Willcox & Blank, 1980;Kinoshita, 1985;Kintsch & Mross, 1985;Love & Swinney, 1996;Lucas, 1987;Onifer & Swinney, 1981;Seidenberg, Tanenhaus, Leiman, & Bienkowski, 1982;Simpson & Burgess, 1985;Simpson & Foster, 1986;Swinney, 1979;Tanenhaus, Leiman, & Seidenberg, 1979) have found that when an ambiguous word is presented, all its meanings are activated, and the context selects the meaning consistent with it.…”