LEXICAL STORAGE
المؤلف:
John Field
المصدر:
Psycholinguistics
الجزء والصفحة:
P158
2025-09-10
368
LEXICAL STORAGE
Book: Psycholinguistics
Author: John Field
Page: 158
The way in which lexical items are organised in the lexicon so as to ensure rapid access.
Most current models assume that words are linked in a complex network which reflects semantic relationships such as partial synonymy, antonymy and hyponymy. Similar links enable us to make collocational associations; these may be established as a consequence of repeated co-occurrence (heavy + smoker, highly + emotional) or of semantic association (fish + chips). It seems probable that a whole sequence like fish and chips is stored as a single chunk.
Lexical items are similarly associated by form. This has obvious benefits for understanding language; but evidence from Slips of the Tongue (SOT) indicates that it also assists in language production. A word substituted in error frequently has formal resemblances to the target word (average for avarice). The SOT evidence suggests that important criteria for characterising word forms are:
number of syllables: sleep! speak; obsolete! absolute
location of stress: unanimously! anonymously; comprehensive! contraceptive
initial syllable: syllables! cylinders; Protestant! prostitute
final syllable or rime: decimal! dismal; Alsatian! salvation
The last two constitute what is sometimes termed the bathtub effect, with the first and last syllable of a word more robust and more likely to be retained in a Slip of the Tongue (antidote! anecdote). The analogy is to the head and knees of somebody in a small bath.
A simple view of lexical storage might suggest that words are in some way stored ‘together’ according to semantic and to formal criteria. This would assume an extremely large store with much replication. For example, the word CABBAGE would be stored with POTATO and PEA in a subset of vegetables; in a set of thousands of disyllabic words which bear initial stress; in other relatively large sets of words beginning with /k æ / and CA; and in rather smaller sets ending in /Id Ʒ / and-AGE. One solution is to assume that reception and production are separate processes, the former accessing words grouped by meaning (with cross-checks to form) and the latter accessing words grouped by form (with cross-checks to meaning).
However, a network model offers a more convincing account. Influenced by connectionist theory, such a model assumes that words are not stored together in sets but are linked by connections of varying strengths. Thus, there are strong semantic links between CABBAGE and POTATO and strong formal links between CABBAGE and CARRIAGE. Strong links might also be the consequence of regular co-occurrence– associating CABBAGE with (for example) CHI NESE or RED. These links are strengthened as more and more examples are encountered of the two words used in conjunction.
Importantly, network models incorporate the principle of spreading activation. It is well documented that hearing or seeing a word such as DOCTOR will prime (speed up recognition of) associated words such as NURSE, PATIENT or HOSPITAL. This effect is represented in the form of activation– best envisaged as a kind of current which diminishes in strength as it travels further from the initial word. Thus, DOCTOR primes NURSE, but is unlikely to prime PLUMBER, even though the term is also a job.
An interesting finding in priming has been that concrete words do not appear to prime closely associated abstract words; and vice versa. This has suggested to some commentators that there may be separate lexical stores for the two types.
The notion of words as linked by a network of forms and meanings is an important one when considering language acquisition. Learning a new lexical item is not just a matter of mastering the form of the item and associating it with a sense or range of senses. The item also has to be linked to the whole network of previously learned words. If a child learns the word TERRIFIED, it has to (a) form a connection with HORRIFIED and TERRIER which are similar in form; (b) form a connection with AFRAID and SCARED which are similar (but distinct) in meaning.
Psycholinguistic research into second-language vocabulary acquisition and into the vocabulary of bilinguals has especially concerned itself with the question of whether the individual operates with a single unified store or two parallel ones. If the former is the case, words in L1 might be linked to those in L2 by shared meaning or by similarity of form. There is some evidence of formal links. With native Spanish speakers studying in the USA, the word RED in an English language task was shown to prime the word RED in a Spanish language task, even though in Spanish the word means ‘net’. The same effect did not obtain between English FROG and Spanish rana.
See also: Lexical access, Lexical entry, Semantic network
Further reading: Aitchison (2003)
الاكثر قراءة في Linguistics fields
اخر الاخبار
اخبار العتبة العباسية المقدسة
