The Mental Lexicon

by

Niels Schiller

• LAST MODIFIED: 27 October 2021
• DOI: 10.1093/obo/9780199772810-0282

Introduction

The term Mental Lexicon, sometimes seen as an online version of a dictionary, is often taken to refer to the part of our language system that hosts the word forms we know and their corresponding meanings. Unlike a dictionary, words in the Mental Lexicon are not organized strictly alphabetically (although speech errors such as malapropisms, e.g., appointment –> apartment, suggest that words having a form relationship are somehow relatedly represented) but stored according to principles of meaning, form, frequency, and possibly other aspects. Form refers to the phonological make-up of words, but also to orthography (spelling) or gestures. Moreover, words must be mentally represented and organized in an easily accessible way to support efficient processing—both in comprehension and production. However, how are words represented and connected to each other in the Mental Lexicon? It is clear that our brain must have the capacity to store word forms and their meanings because understanding and producing words are important skills of the human language user and both processes involve form and meaning. Theories of language comprehension and language production include ideas about the organization and representation of words in the Mental Lexicon. Comprehension and production is one important dichotomy reflected in this article. Another contrast is the difference between representation and processing. Representation of units in the Mental Lexicon, be it bundles of words, individual words, morphemes, syllables, phonemes, or features, is rather static. When we learn a word, it is important to store the correct pronunciation, spelling, or gestures with the particular meaning. Processing, however, is more dynamic. When we hear or produce a word, a number of (dynamic) cognitive processes are being executed that operate on (static) representations in the Mental Lexicon. When we hear an acoustic signal corresponding to words we know, our language comprehension system can recognize those words and map them to corresponding meanings resulting in speech comprehension. A similar process takes place when we read text written in a language we can read. Visual word forms are being recognized and mapped onto meanings leading to text comprehension. If word forms are not stored in our brain, we cannot recognize them and thus we are not able to understand what is being said or written down. This happens when we listen to speech or read words in a language we do not know. When we produce speech, we start with the intention to convey a certain message making access to stored information in the Mental Lexicon necessary. The exact information that is being stored in the Mental Lexicon and how it is stored and accessed, is still being investigated. This bibliography will especially be concerned with features of form representation of words, i.e., morphology and phonology, and less so with the representation of meaning, i.e., semantics.

Levels of Representation of Words

How are words represented in the Mental Lexicon? According to De Saussure 1916, words have a form and a meaning. It seems obvious that these two aspects of words must be stored in the Mental Lexicon since the form-meaning relationship is—for the most part—arbitrary in language. The meaning of a word corresponds to the concept it activates in our mind. However, words also have other features that do not strictly belong to form nor meaning but need to be stored in the Mental Lexicon, such as morpho-syntactic features or word frequency. Take, for instance, grammatical gender or the difference between count and mass. Our Mental Lexicon must have stored that garlic is a mass noun (i.e., cannot be counted) while onion is a count noun because there is nothing in the form of the word or the visual appearance of the object that may hint to the difference between count and mass. Similarly, we must have stored in the Mental Lexicon that German Löffel (spoon) has masculine gender, Gabel (fork) has feminine gender, and Messer (knife) has neuter gender because there are only subtle and probabilistic cues in the form of these words and few semantic cues that predict grammatical gender assignment (however, see Köpcke and Zubin 1983 for gender assignment rules). Jescheniak and Levelt 1994 provides evidence that processing features such as lexical frequency are stored at the lexical level. It is well-known that words that occur relatively often in a language are recognized and produced faster than words that occur less often—when all other factors, such as length, concreteness, age of acquisition, familiarity, etc. are controlled for. This frequency effect must be included in models of the Mental Lexicon. Aitchison 1994 and Altmann 1999 are examples of monographs that give a comprehensive overview to a large variety of aspects regarding the Mental Lexicon. Taft 1991 is a monograph focusing on written word-form processing (spelling and reading).

• Aitchison, J. 1994. Words in the mind: An introduction to the Mental Lexicon, 2d ed. Oxford: Blackwell.

  This monograph is a basic and easy to understand introduction to how humans handle the words they use in their cognitive system. Large parts of the book are concerned with word meaning (semantics) and how children acquire words, relatively few chapters touch upon formal aspects of words.

• Altmann, G. 1999. The ascent of Babel: An exploration of language, mind, and understanding. Oxford: Oxford Univ. Press.

  This monograph is a psycholinguistic description about how the mind produces and comprehends language. The book describes the whole process of language acquisition, comprehension, and production, and pays a lot of attention to the representation and processing of words in the Mental Lexicon.

• De Saussure, F. 1916. Cours de linguistique générale. Paris: Payot.

  English translation: “Course in general linguistics.” This posthumously published monograph is considered to be the birth of linguistic structuralism. This work is especially known for making explicit the difference between two aspects of words that are linked with each other, i.e., form and meaning.

• Jescheniak, J. D., and W. J. M. Levelt. 1994. Word frequency effects in speech production: Retrieval of syntactic information and of phonological form. Journal of Experimental Psychology: Learning, Memory, and Cognition 20.4: 824–843.

  DOI: 10.1037/0278-7393.20.4.824

  This article investigates the locus of word frequency effects in speech production. The most crucial experiment tests the naming latencies for homophonic word forms through a translation naming paradigm. Results show that low-frequency homophones (e.g., nun) behave like high-frequency control words, presumably because they inherit the frequency from their high-frequency twin (e.g., none). This suggests that the word-frequency effect originates at the word-form level, and not higher up in the system such as the lemma or the concept level.

• Köpcke, K.-M., and D. Zubin. 1983. Die kognitive Organisation der Genuszuweisung zu den einsilbigen Nomen der deutschen Gegenwartssprache. Zeitschrift für Germanistische Linguistik 11.2: 166–182.

  DOI: 10.1515/zfgl.1983.11.2.166

  English translation: “The cognitive organization of gender assignment in monosyllabic nouns of contemporary German.” This article provides experimental evidence for gender assignment regularities in German monosyllabic nouns. It is demonstrated that gender assignment, which has been described as arbitrary in the past, follows several semantic and phonological regularities or rules. Just like any other rule, these rules have exceptions as well. This is just one paper by these authors on this topic in German linguistics.

• Taft, M. 1991. Reading and the Mental Lexicon. Hove, UK, and London: Lawrence Erlbaum.

  This monograph focuses on lexical processing during reading. More specifically, the book summarizes the research on visual word recognition in healthy adults. Models of visual processing are evaluated with the help of a number of well-established effects. Moreover, the influence of semantic, syntactic, morphological, and phonological factors on visual lexical processing are extensively discussed.