sentence processing

Glossary of Grammatical and Rhetorical Terms

sentence processing
Linguist Roger P.G. van Gompel reports that modern work on sentence processing "started with the development of theories that explain how sentence structure is built up word-by-word by incrementally incorporating each word into the previously built syntactic structure" (Sentence Processing, 2013). (Michaela Begsteiger/Getty Images)


Sentence processing is the study of the ways in which sentences are produced and understood.

One basic finding of linguists is that sentence processing is incremental--that is, "humans structure and interpret the words of an utterance as they are perceived rather than store them as a list to be combined later" (Weber et al.).

See Examples and Observations below. Also see:

Examples and Observations

  • "The goal of any theory of sentence processing is to determine how people arrive at the desired interpretation of a given sentence, attempting thus to investigate the very nature of the human language processor. Sentence processing is a rapid and automatic process, one which 'is closely linked in time to the input' (Garrett, 1990: 136, Pickering, 1999). This means that sentence processing proceeds in an incremental fashion: once each word is encountered, it is integrated to the sentence analysis. Sentence processing necessarily involves a syntactic analysis of the sentence, usually referred to as parsing, which allows the reader or listener to identify the syntactic role of words within each sentence, and an analysis which involves the use of semantic and pragmatic information to arrive at a plausible interpretation. Whether these two analyses are two distinct processes in terms of when they occur during sentence comprehension and the information they have access to is a controversial issue and one that has given rise to different theoretical proposals."
    (Despoina Papadopoulou, "Introduction." Cross-Linguistic Variation in Sentence Processing: Evidence from RC Attachment Preferences in Greek. Springer, 2006)

  • "[T]he human language processing faculty [is able] to integrate diverse cues, including linguistic context, intonation, world knowledge, and visual context . . .. [P]urely linguistic constraints, long thought to identify the core of sentence comprehension mechanisms, can in fact be overridden by highly contextual aspects of the situation, such as the intonation contour of a particular utterance, semantic expectations supported by the visual scene, and indeed events going on in the scene itself."
    (Andrea Weber, Matthew W. Crocker, and Pia Knoeferle, "Conflicting Constraints in Resource-Adaptive Language Comprehension." Resource-Adaptive Cognitive Processes, ed. by Matthew W. Crocker and Jörg Siekmann. Springer, 2010)

  • Theories of Sentence Processing
    "Early work in sentence processing mirrored the emphasis on the syntactic component in grammatical theory, by focusing on the large-grained structural properties of interpretations as they are developed incrementally. For example, the most widely known statements of preference, Minimal Attachment and Late Closure (Frazier, 1978), rely solely on general properties of the size and locality of incremental additions to a partial syntactic structure. Beginning with the work of Ford, Bresnan, and Kaplan (1982), there has been a gradual shift to emphasizing more and more the influence of individual lexical information within such structure-based accounts. For example, Pritchett (1992) examines the role of the lexical head of a phrase in determining its basic properties, and the impact of the argument structure of a verb on preferred structural representations. However, while some lexical properties are assumed to influence processing decisions in this type of structure-based account, they are viewed as secondary to structural information.

    "The recent lexicalist constraint-based approach in sentence processing takes the lexical basis of language comprehension much further, suggesting that perhaps all of the relevant processing distinctions can be traced to distinctions in lexical information . . .. On this view, interpretation is an incremental process of satisfying constraints associated with lexical entries."
    (Suzanne Stevenson and Paolo Merlo, "Words, Numbers and All That: The Lexicon in Sentence Understanding." The Lexical Basis of Sentence Processing: Formal, Computational and Experimental Issues, ed. by Paola Merlo and Suzanne Stevenson. John Benjamins, 2002)

  • Lexical Processing and Sentence Processing
    "[T]here is a fundamental difference between lexical and syntactic processing: the lexemes in a language, being finite in number, are stored in the mental lexicon. Sentences, however, typically are not stored in any kind of mental list (if they were, then we would be unable to produce any new sentences, i.e. sentences that we have never heard or read before). Indeed, sentence repetition and sentence recognition experiments have shown that normally syntactic structures are extremely transient: memory for syntax is unreliable only half a minute after a sentence has been heard or read . . .. Hence, whereas word recognition can be described as a retrieval process with the goal of finding an entry in the mental lexicon, sentence processing does not involve accessing and retrieving entries from a mental repository.

    "If the representations of sentences are not retrieved from a mental store, this means that they are constructed on-line (in a step-by-step fashion) in accordance with syntactic principles or rules. It follows that sentence comprehension involves segmenting the sentence into relevant processing units and constructing a syntactic representation for the sentence (the technical term for this is parsing)."
    (Linguistics: An Introduction, ed. by Andrew Radford, Martin Atkinson, David Britain, Harald Clahsen, and Andrew Spencer. Cambridge University Press, 1999)