Further explorations of the consistency effect in word and nonword pronunciation

Abstract

Three experiments explored the consistency effect in word and nonword pronunciation. All three experiments involved a common second phase in which subjects were required to pronounce a sequence of consistent and inconsistent words and nonwords. The experiments were distinguished by the nature of Phase 1. In Experiment 1, subjects pronounced a sequence of words and nonwords in which all the words were exceptions. In Experiment 2, all the items in this phase were consistent, and in Experiment 3, they were inconsistent. Consistency effects for words, as measured by pronunciation latency, were confined to Experiment 1. Consistency effects for nonwords were found in Experiment 2. No consistency effects were found for either words or nonwords in Experiment 3. There was clear evidence of a consistency effect for nonwords in Experiment 1 in terms of pronunciation errors, but the latency data were not significant. This is attributed to the very slow response times in this condition. All these data are compatible with both revised analogy and dual-route theories.Two classes of theory seek to explain the processes whereby the pronunciation of words and nonwords is achieved: the dual-route theory and the analogy theory. The former, as originally proposed by Coltheart (1978) and subsequently developed by Morton and Patterson (1980) and Parkin (1984), postulates two functionally independent methods of word processing. One of these is a lexical processing route that operates by the direct mapping of the visual characteristics of a letter string onto its stored lexical representation in the mental lexicon. In addition, a nonlexical processing route is assumed to translate the visual code of a word into a phonological one on the basis of spelling-to-sound rules. Coltheart proposed that this phonemic transcription operated by means of grapheme-to-phoneme correspondence rules (GPCs) as specified by Venezky (1970). These rules were derived by considering a maximum of two letters at a time (Le., vowel, vowel digraph, consonant, and consonant clusters) to define their major and minor correspondences. For example, the major correspondence (Le., the most common pronunciation) of the vowel digraph EA is Ii! as in BEAN, and minor Oess common) ones are leil as in BREAK and lei as in BREAD. Thus the phonemic transcription of the word BEACH would yield fbi, Iii, and ItJI. ' Recent modifications to dual-route theory have amended Coltheart's (1978) assertion that the nonlexical route operates solely by the application of GPCs. Shallice, Warrington, and McCarthy (1983) and Parkin (1984) suggested that nonlexical assembly of phonology may also use larger patterns of letters operating beyond the GPC level. If this is the case, some words exhibiting a minor correspondence in terms of Venezky's (1970) rules have a major correWe are grateful to the editor and to two anonymous reviewers for their many helpful comments on earlier versions of this manuscript. Please direct all correspondence to: Alan J. Parkin, Laboratory of Experimental Psychology,...

Understanding normal and impaired word reading: Computational principles in quasi-regular domains.

“…Although MINT may be a regular word according to GPC rules, its spelling-sound relationship is inconsistent with that of its orthographic neighbor, PINT. To the extent that the process of computing phonology from orthography is sensitive to the characteristics of the neighborhood, performance on a regular but inconsistent word like MINT may also be adversely affected. Glushko (1979) did indeed demonstrate longer naming latencies for regular inconsistent words than for regular words from consistent body neighborhoods, though this result was not always obtained in subsequent experiments (e.g., Stanhope & Parkin, 1987).…”

“…To the extent that the process of computing phonology from orthography is sensitive to the characteristics of the neighborhood, performance on a regular but inconsistent word like MINT may also be adversely affected. Glushko (1979) did indeed demonstrate longer naming latencies for regular inconsistent words than for regular words from consistent body neighborhoods, though this result was not always obtained in subsequent experiments (e.g., Stanhope & Parkin, 1987).In 1990, Jared, McRae, and Seidenberg offered a more sophisticated hypothesis that captures aspects of results not handled by previous accounts referring solely to either regularity or consistency. According to Jared and colleagues, the magnitude of the consistency effect for a given word depends on the summed frequency of that word's friends (words with a similar spelling pattern and similar pronunciation) and of its enemies (words with a similar spelling pattern but a discrepant pronunciation).…”

The inconsistency of consistency effects in reading: The case of Japanese Kanji.

“…To the extent that the process of computing phonology from orthography is sensitive to the characteristics of the neighborhood, then performance on a regular but inconsistent word like five should also adversely be affected. Glushko did indeed demonstrate longer response times (RTs) for regular inconsistent words than for regular words from consistent neighborhoods, though this result was not always obtained in subsequent experiments (e.g., Stanhope & Parkin, 1987;Taraban & McClelland, 1987; see also Brown, 1987;Patterson & Coltheart, 1987). In all studies manipulating either regularity or consistency, the effect has been shown to be stronger for low-than for high-frequency target items and is often statistically reliable only for the lower frequency words.…”

“…Many studies have addressed this question in English (e.g., Andrews, 1982Andrews, , 1989Andrews, , 1992Glushko, 1979;Jared, McRae, & Seidenberg, 1990;Parkin, 1982;Seidenberg, Waters, Barnes, & Tanenhaus, 1984;Stanhope & Parkin, 1987;Taraban & McClelland, 1987;Waters & Seidenberg, 1985). On the basis of the frequent observation (e.g., Coltheart, 1978;Parkin, 1982;Waters & Seidenberg, 1985) that words with regular or typical spelling-sound correspondences (such as five) produce shorter naming latencies and lower error rates than words with exceptional correspondences (e.g., give), regularity was originally considered to be the critical variable.…”

Properties of attention during reading lessons.

“…These were the following: (a) scaled scores from the reading comprehension subtests from the MAT, Primary 1 and Primary 2, Forms L (in the standardization sample, K-R 20 = .93 for both levels; Prescott, Balow, Hogan, & Farr, 1986); (b) scaled scores from the reading comprehension subtest of the Illinois Goal Assessment Program (IGAP; in a standardization sample, coefficient a = .84; Valencia, Pearson, Reeve, & Shanahan, 1988); (c) teachers' ratings of students' comprehension measured on a 6-point Likert scale (factor loadings with other measures of comprehension from the study sample provided lower bound estimates of reliability, which were .86 for Grade 2 and .68 for Grade 3); (d) time, in hundredths of a second, to read two passages from the Gray Oral Reading Tests-Revised (in the study sample, coefficient a = .92 for Grade 2 and .89 for Grade 3; Weiderholt & Bryant, 1986); and (e) time, in hundredths of a second, to pronounce two lists of pseudowords adapted from Stanhope and Parkin (1987) and Stanovich, Cunningham, and Feeman (1984) (in a study sample, coefficient a = .90 for Grade 2 and .81 for Grade 3). We took the mean of reading times on the two passages and the mean of pronunciation times on the two lists of pseudowords.…”