Lewis S. Notestine & Melissa A. Redford, University of Oregon
Interactions of Verbal Ability and Morphological Processing in the English Mental Lexicon
A psycholinguistic study was conducted to explore possible effects of individual differences in verbal ability and experience in the processing of morphologically complex English words. The Lexical Decomposition Hypothesis posits that the storage of and access to morphemes in the lexicon and that of whole words are dealt with by different sets of cognitive processes (Taft & Forster, 1975; Stanners, Neiser, Hernon & Hall, 1979; Marslen-Wilson, Tyler, Waksler & Older, 1994). Thus, in a simple version of the Lexical Decomposition Hypothesis, monomorphemic words such as kangaroo are processed holistically, while morphologically complex words such as heliocentricity are processed analytically via their constituent morphemes.
An unanswered question from these studies is how and whether morphological effects in processing may differ between individuals and populations, and whether they may vary with verbal ability and experience. Although not a specific aim of the earlier studies mentioned above, such studies nevertheless discussed how variability in verbal ability and literacy might be found to interact with morphological processing effects. More recent research has addressed the possibility of such an interaction more directly. Dynamic models of the mental lexicon such as those developed by Bybee (1985, 1995, 2001) treat morphological regularities as emergent from language use and frequency of exposure. Such models have been adopted by Gonnerman (Gonnerman, 1999; Seidenberg & Gonnerman, 2000), whose main finding is that morphological effects on lexical decision and word recognition may depend on the semantic and phonological similarity of morphologically related words. This latter finding highlights the role of variability in morphological processing, and suggests that differences between people may explain some of this variability.
Because such models predict differences in the structure of the mental lexicon based on individual aptitude and linguistic experience, it was predicted that reaction time and error measures for a recognition memory task would show an interaction of the variables of morphological complexity in stimulus sets and individual language ability.
The study included two experiments comparing groups of participants with high verbal ability (high-verbals) and low verbal ability (low-verbals) in a recognition memory task, containing English-like nonsense words. Participants were grouped based on verbal ability by a composite measure of subjective familiarity with a set of 160 English words, and a short questionnaire designed to measure linguistic experience. These measures were similar to those used in some previous research (Lewellen, Goldinger, Pisoni & Greene, 1993). Participants learned two sets of 60 test items in a computer-based learning task. One set of nonsense word stimuli were comprised of recurrent phonological and orthographic strings, simulating the phonological and orthographic regularities found in morphologically related words (e.g., bis-ruck, bis-mar). These were termed complex stimuli. A second set of 60 control stimuli was composed of matched nonsense words, with the difference that each token in this set was phonologically and orthographically unique (e.g., vastem, tiekner), and none of these tokens had any phonological or orthographic regularity. These were termed simple stimuli. A recognition memory task included 60 familiar items (previously learned) and 60 unfamiliar items (as distractors). The conditions in the second experiment were the same, with the difference being that boundaries between the recurring phonological and orthographic units were explicitly marked with a period during the computer-based learning task.
The predicted interaction was found, and the most important findings were that (1): in Experiment 1, high-verbals showed better recognition memory than low-verbals for the simple stimuli but not for complex stimuli, (2): Explicit marking of unit boundaries in Experiment 2 facilitated recognition memory more in high-verbals than in low-verbals for complex stimuli, and (3): responses to unfamiliar stimuli showed greater ‘false positives’ for high-verbals when the unfamiliar stimuli were complex or simple, (e.g., bis-tem). These results suggest that high-verbals may be more sensitive to morphological cues in word recognition than low-verbals, and they imply that more research in individual differences may be needed to better understand the emergent, dynamic structure of the mental lexicon.
References:
Bybee, J. (1985). Morphology: A study of the relation between meaning and form. Philadelphia: J. Benjamins.
Bybee, J. (1995). Diachronic and typological properties of morphology and their implications for representation. In Feldman, L.B. [Ed.] Morphological aspects of language processing. (pp. 114-135) Hillsdale, NJ: Erlbaum.
Bybee, J. (2001). Phonology and language use. New York, NY: Cambridge University Press.
Gonnerman, L.M. (1999). Graded semantic and phonological effects in lexical processing: Implications for morphology. Psychonomic Society Abstracts, 4, p.44.
Lewellen, M.J., Goldinger, S.D., Pisoni, D.B. & Greene, B.G. (1993). Lexical familiarity and processing efficiency: Individual differences in naming, lexical decision, and semantic categorization. Journal of Experimental Psychology: General, 122(3), 316-330.
Marslen-Wilson, W., Tyler, L., Waksler, R. & Older, L. (1994). Morphology and meaning in the English mental lexicon. Psychological Review, 101, 3-33.
Seidenberg, M.S. & Gonnerman, L.M. (2000). Explaining derivational morphology as the convergence of codes. Trends in Cognitive Sciences, 4(9), 353-361.
Stanners, R.F., Neiser, J.J., Hernon, W.P. & Hall, R. (1979). Memory representation for morphologically related words. Journal of Verbal Learning and Verbal Behavior, 18, 399-412.
Taft, M. & Forster, K. (1975). Lexical storage and retrieval of prefixed words. Journal of Verbal Learning and Verbal Behavior, 14, 638-647.
