This book "supplies a vocabulary of English words and idiomatic phrases 'arranged … according to the ideas which they express'. The thesaurus, continually expanded and updated, has always remained in print, but this reissued first edition shows the impressive breadth of Roget's own knowledge and interests."
Date: Wed, 30 Apr 2003 13:23:42 -0500 From: Phaedra Royle <firstname.lastname@example.org> Subject: Words in the Mind: An Introduction to the Mental Lexicon, 3rd ed.
Aitchison, Jean (2003) Words in the Mind: An Introduction to the Mental Lexicon, 3rd ed., Blackwell Publishing.
Phaedra Royle, School of Communication Sciences and Disorders, McGill University
Words in the Mind presents a theoretical background to the mental lexicon. It discusses how words are learned, dissected, stored and produced. This book is an introductory text intended for the undergraduate scholar or layman, although a more knowledgeable public might find it an interesting introduction to the field of psycholinguistics, in particular as it relates to the mental lexicon.
The book is divided into four sections, "Aims and Evidence", on what the data are and what we want to do with it; Basic Ingredients, presents different elements from which words are 'made'; "Newcomers" spans word creation, semantic drift and lexical acquisition; while "The Overall Picture" discusses models of the mental lexicon. Each section is subdivided into chapters. The book also contains a Notes section, a Bibliography and an Index.
In 1987, Jean Aitchison published the first edition of this text on the Mental Lexicon. It has since become a reference in the field, especially as an introductory text to the theoretical issues surrounding the representation and processing of words. In this book, theoretical issues such as semantic networks, lexical categories, phonological representations, neology and others, are presented in a straightforward manner not overburdened by psycholinguistic jargon. Theoretical notions are illustrated using text (often poems), analogy and visual props in the form of graphs, comic strips and diagrams. Other more "classic" types of data (slips of the tongue, aphasias, tips-of-the-tongue) are also included. The book is divided into 21 chapters, which are very short -- rarely spanning more than ten pages -- thus making the information quite easy to assimilate. The book is thus approachable and user-friendly. One caveat is the fact that all notes are at the end of the book in a Notes section. Since most of the notes are simply references, I believe it would have been more judicious to include them in the text, put them as footnotes, or include them at the end of each chapter.
Words in the mind is a seminal text on the mental lexicon, now in its third edition. The author notes in the preface that substantial changes have been made to previous editions, in order to incorporate numerous findings that have been made in this expanding field of study. For example, an additional chapter on meaning change has been added, and other modifications have been made to already-existing chapters. However, I was left with the nagging feeling that the text is dated and is lacking in essential information from the field. One area of strong interest -- morphological processing -- seems to have been overlooked in a number of ways. For one, in the chapter on word recognition (Organized Guesswork: Recognizing words), Aitchison states: "In Speech recognition [...] speakers flash up on their mental screen [...] any word that is consistent with what they hear, then make use of all the available evidence -- syntactic and semantic -- to narrow down the possibilities. The more information they are able to bring to bear on the situation, the faster they can come to a decision. [...] People start doing this as soon as they hear any part of a word."(p. 235) I believe that this simplification of research data may be misleading. Although it is possible that semantic and phonological information might help one recognize a word, these effects are not equal and are most probably not all active at the same time, from stimulus onset. In fact, it has been proposed that semantic and syntactic effects found in word recognition research can be attributed to response bias effects, especially since they can disappear when lexical decision is speeded up, and thus more automatic. In addition, it has been shown that semantically and syntactically ambiguous forms (bank and bug_N vs. bug_V) can activate their multiple meanings, even when context highly constrains a specific interpretation of a word, suggesting that semantics and syntax do not necessarily help us arrive at the appropriate meaning of a word (Cutler, 1995).
Research from visual word recognition also shows that these effects are neither necessarily simultaneous nor equal. Using primed visual recognition tasks, Feldman and Prostko (2002) showed that orthographic, morphological and semantic effects are of different magnitudes and different time courses across different tasks (lexical decision, naming, and go/no go naming). Morphological effects are systematically stronger and appear earlier than semantic effects, across tasks. Orthographic effects, also usually less strong than morphological ones, vary depending on the time between presentation of the prime and target, and depending on the task. In some cases, they can be null or inhibitory.
Finally, morphological priming effects are never equal to, and are usually quite different from, the addition of orthographic and semantic priming effects. It is likely that some of these type and time constraints on priming are indicative of the organization of the lexicon, and that morphological structure is the preferred mode of early lexical access, while semantic knowledge comes in at a later, more controlled stage. Finally, in the area of neuroimaging, different time courses are found for the processing of different language components. Specific event related brain potentials (ERPs) have shown activation at different time windows in language processing. Mismatch negativity responses (MNRs) are typically found early on (between 100 and 250 ms) in phonological processing (Dehaene-Lambertz, Dupoux, & Gout, 2000). Left anterior negativity (LAN) patterns appear, around 100-300 ms after stimulus, in the processing of syntactic violations. The N400 occurs when processing semantically incongruous words (e.g. I went to the zoo to see the judge), approximately 400 milliseconds after stimulus offset. A positive brain potential appears at around the same time if the sentence is appropriate. It thus seems to reflect semantic integration. Finally, the P600 is related to grammatical violation of syntax (agreement, for example) and the reanalysis of garden-path sentences, and appears typically 600 ms after stimulus offset. It has been argued to reflect some sort of repair process (Gunter, Frederici, & Schreifers, 2000).
The data here is probably too complex to be included as such in the book, but it seems to indicate that the time-course of activation of different linguistic components is important for language recognition. These data do not support the notion that all information is available or used immediately when a word is perceived.
Another area that could have been expanded on concerns the roles of productivity and transparency of derivational processes in word processing. In the chapter on word structure (Bits of Words: The internal architecture of words), Aitchison discusses the issue of words with different suffix types (# or +, using a Kiparskian level morphology, 1982), and of how these are processed by the language user. Aitchison proposes that plus-type affixed words are probably not decomposed because of their idiosyncrasy, but that evidence bearing on the hash-type suffixed words is not conclusive. She states that "[t]hey are probably firmly fixed in common words, otherwise there might be far more errors, such as *goodism instead of goodness", and goes on to mention "[s]o far, experimental evidence on this point has not been mentioned. Unfortunately, experiments which have tried to examine this question are somewhat inconclusive, partly because they have failed to distinguish between different languages, different suffixes types and differences between frequency of use."(p. 134) Apart from the fact that the research cited to support this claim all date from 1993 or before, I would like to mention that there has been a wealth of experiments on derivational processes that pay attention to factors such as language, suffix type and productivity. Tsapkini, Kehayia and Jarema (1999) have specifically investigated the effects of phonological change between the stem and derived form in English word recognition (simple visual, primed visual and cross-modal primed lexical decision). They have shown that word recognition is more rapid when derivational processes are phonologically transparent, while opaque forms show slowed recognition. Bertram, Laine and Virkkala (2000) studied effects of frequency and productivity in word acquisition by young Finnish-speakers. They found that morpheme type and frequency had an effect on word recognition. Overall, children were better at recognizing multi-morphemic words than monomorphemic ones matched for length and frequency, showing that a morphological analysis strategy can help word recognition. They found that this effect was stronger in low-frequency words. They also found that in high-frequency words, recognition of words with less productive suffixes is poorer than for other word types. Based on data from Dutch and Finnish, Bertram, Schreuder & Baayen (2000) proposed that three factors influence whether a word will be stored or decomposed during lexical access. These are word formation type, productivity and affixal homonymy (the same form serving multiple semantic or syntactic uses). There is much recent data in English and in other languages on suffix types and their processing.
These two highlighted examples are unfortunate, since morphological processing is one of the areas in psycholinguistics where there are strong debates, interactions and collaborations between linguists and psycholinguists. It is a hot topic area where research is ongoing and active. To present it as succinctly as the author has done does not serve the field and it does not highlight the excitement generated by the research. Nor does this book reflect the current state of knowledge on this issue.
In conclusion, I would only recommend using this book with additional course materials, most probably more recent articles on selected topics. The text can be used as a starting point for discussion but needs to be approached with a critical mind.
Bertram, R., Laine, M., & Virkkala, M. M. (2000) The role of derivational morphology in vocabulary acquisition: Get by with a little help from my morpheme friends. Scandinavian Journal of Psychology. 41(4):287-296.
Bertram, R., Schreuder, R., & Baayen, R. H. (2000) The balance of storage and computation in morphological processing: The role of word formation type, affixal homonymy, and productivity. Journal of Experimental Psychology: Learning, Memory, & Cognition, 26:489-511.
Cutler, A. (1995) Spoken word recognition and production. In Miller & Eimas (eds.) Speech, language, and communication. San Diego, CA: Academic Press. (pp. 97-136.)
Dehaene-Lambertz, G., Dupoux, E., & Gout, A. (2000) Electrophysiological correlates of phonological processing: A cross-linguistic study. Journal of Cognitive Neuroscience, 12(4):635-647.
Feldman, L. B. & Prostko, B. (2002) Graded Aspects of Morphological Processing: Tasks and Processing Time. Brain and Language, 81:12-27.
Gunter, T. C., Frederici, A. D., & Schreifers, H. (2000) Syntactic Gender and Semantic Expectancy: ERPs Reveal Early Autonomy and Late Interaction. Journal of Cognitive Neuroscience 12(4):556-568.
Kiparsky, P. (1982) Lexical morphology and phonology. In I. S. Yang (ed.), Linguistics in the Morning Calm, 2 (pp. 3-91). Seoul: Hanshin.
Tsapkini, K., Kehayia, E. & Jarema, G. (1999) Does phonological change play a role in the recognition of derived forms across modalities? Brain and Language, 68(1-2):318-323.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER Phaedra Royle holds a Ph.D. in Linguistics from the Université de Montréal. Her interests lie in psycholinguistics, language disorders (Specific Language Impairment), language acquisition and morphology. Her thesis investigated lexical access in language-impaired French-speaking adolescents and adults. She is presently carrying out postdoctoral research on early language acquisition in French-speaking children with and without language delay at the School of Communication Sciences and Disorders, McGill University.