•                                                                                72

3.1   Introduction: Top-Down Processing as the Driver of Comprehension

As we outlined in the previous chapter, linguistic processing enables the listener to capture the input and to represent it in short-term memory.The listener then uses this representation to construct meaning and interact with the input and the speaker. This coordination between objectively oriented and subjectively oriented processing is the essence of the experience of listening.

While linguistic processing provides the foundation of comprehension, it is semantic processing that drives the process of creatingmeaning. Semantic pro- cessing begins effectively once syntactic processing is complete, when there is a symbolic representation of the inputin terms of semantic roles. At this point, the listener begins to integrate this linguistic representation with their own experi- ential knowledge—their knowledge of words, of concepts, and of related ideas and contexts—and constructs meaning.

DOI: 10.4324/9781003390794-5


56  Defining Listening


We can say that semantic processing is a comprehension-building activity through which the listener interprets language subjectivelybased on pre-existing knowledge and expectations. Semantic processing involves using higher-level cognitive processes, which involve the more advanced cognitive functions of abstract thinking, reasoning, contextual understanding, decision-making, and memory encoding (Toulmin, 2006).

In this chapter, we explore these higher-level processes in three domains that encompass the listener’s comprehension-building ability:lexical access, schema activation, and inferencing.


3.2   Comprehension Building through Lexical Access

Semantic processing begins as soon as word recognition is achieved through linguistic processing. This initial step is called lexical access. Lexical access refers to retrieving relevant connections associated with recognized words. This retrieval involves activating the appropriatelexical representations stored in the memory, from the network referred to as our mental lexicon. Neurologically, activating a word in the mental lexicon involves the firing of specific neural circuit pathways between various brain regions where memory associations are stored.

This firing of linked neural circuits is referred to as spreading activation. When the neural representation of a word is activated, the network recruits semantic information from various “tagged” memory stores about the word, associated linguistic features, and concepts related to the word (Font et al., 2014). These are aspects of depth of knowledge of the recognized word—its semantic and syntactic associations. An individual can activate knowledge associated with their experience of hearing and reading the word in various contexts (Beglar & Nation, 2013; Qian & Lin, 2019; Schmitt, 2014).

This depth of knowledge is accessed along one of two dimensions: paradig- matic and syntagmatic relations (Chiu & Lu, 2015; Peters &Weller, 2008). The organization and interconnectedness of words in our mental lexicons through these three relations facilitate the retrieval of words and contribute to the richness of the listener’s comprehension (Hasan & Rahman, 2020; Sims et al., 2022; Ullman, 2001; Wray, 2016). (See Table 3.1.)

Paradigmatic relations provide a network of semantic connections between words, enriching and deepening the sense of the target lexical item. Activating paradigmatic relations while listening is necessary to understand comple- mentary or contrasting aspects of the speaker’s meaning and nuance, to ini- tiate inferencing and reasoning processes, and to predict upcoming concepts (Carston, 2021; Chiu & Lu, 2015; Green, 2008).

Syntagmatic relations provide a network of sequential connections that enable the speaker and listener to conventionalize frequentcommunica- tion patterns. For the listener, activating conventional syntagmatic relations


Semantic Processing  57


Table 3.1 Paradigmatic and Syntagmatic Relations


  • Hyponymy/Hypernymy: Hyponymy represents a hierarchical relationship between words, where one word (hyponym) is more specific and falls under the domain of another word (hypernym) that is more general. For instance, “salmon” and “goldfish” are hyponyms of the hypernym “fish,”which is itself a hyponym of the word “animal.”
  • Synonymy: Synonymous words have similar meanings or can be used interchangeably in certain For example, “awful,” “terrible,”and “horrible” are all synonyms of “bad.”
  • Antonymy: Antonyms are words with opposite For example, “exit” which is an antonym of “entrance.”
  • Meronymy/Holonymy: Meronymy refers to the relationship between a whole and its parts. For example, “finger” is a meronym of the whole “hand.” The reverse relationship is called holonymy, i.e., “hand” is a holonym of “finger.”
  • Homonymy: Homonyms are words that have the same pronunciation and spelling but differ in For instance, “bank” can refer to a financialinstitution or the edge of a river.
  • Homophony: Homonyms are words with the same pronunciation but different

spellings and meanings. For example, “to,” “too,” and “two” are all homophones.

  • Polysemy: Polysemous words have evolved from the same origin but have multiple related meanings. For example, “run” can mean to move faster than walking (run

a race), flee (run!), fill with water (run the bath), etc. (Oxford English Dictionary claims “run” has over 600 distinct meanings, many with particlesattached, such as “run up the bill.”)

  • Ambiguity: Ambiguous words or phrases have multiple possible meanings or interpretations due to lexical, syntactic, or semantic For example,“Visiting relatives can be boring” has syntactic ambiguity; “I’ll meet you at the bank” has lexical ambiguity; “I saw her duck” has semantic ambiguity.


  • Collocations: Collocations are words that frequently occur together and can be processed as chunks, such as “prices fell,” “heavy rain,” “break a promise.”
  • Situational utterances (also known as conventionalized expressions): Full

utterances that have assumed a conventional pragmatic meaning, e.g., “How can

I ever repay you?”; “Here, let me get that for you.”; “You can say that again!”; “Oh,

that’s an interesting choice.”

  • Ritualized texts: Often used phrases, sayings, and clichés can be interpreted as a single lexical item. For example, “Houston, we have a problem,” a phrase uttered by astronaut Jim Lovell during the Apollo 13, mission is used to signify a serious unexpected problem or complication in a situation andto suggest a kind of coolness under pressure.
  • Polywords: Multiword expressions that have assumed a specific meaning, g., “fixed income,” “work-life balance,” “fish and chips.”
  • Fixed phrases: A sequence of words that are commonly used together and have a specific meaning that may not be predictable from the meanings of the individual words; g., “by sheer coincidence,” “a golden opportunity,” “the more the merrier,” “barking up the wrong tree.”



  • Meta-messages (also known as rhetorical markers): These contain conventional signals about the intended interpretation of a message, g., “for thatmatter,” “as far as I can tell,” “by the way,” “to be honest,” “no offense,” “just kidding.”
  • Derivations: Derivational relations involve forming new words through affixation or word-formation processes. For instance, happiness, happily, and happier are all derived from the root (lemma) form “happy” by adding grammaticalizing
  • Lexicalized sentence frames: Frequently used sentence frames that have a conventional meaning when slots are filled in, g., “I’m sorry if I …”“Do you mind if I …” “Based on (…), I would say that (…).”
  • Literary devices: These include metaphor, hyperbole, onomatopoeia, irony, metonymy. Words and concepts can be related analogously, such as “time is money” or “love is a journey,” providing a nuanced understanding of a familiar word or phrase.


is essential for automatizing word recognition. Knowledge of syntagmatic relations enables the listener to predict upcoming words and to retroactively recognize words that may have been ambiguous. Several types of syntagmatic relations are associated with depth of word knowledge.

The spreading activation in neural pathways of paradigmatic and syntagmatic relations involves both an expansion and a retraction cycle. Initially, spreading activation makes connections to the most strongly activated associations (i.e., those used most frequently or those encodedwith highly charged emotional situ- ations in the past). Activation spreads along the semantic network links, poten- tially including all of theparadigmatic and syntagmatic relations that the listener knows unconsciously. As new words are recognized, activation of associations with previous words retracts to allow for associations with more recently recognized ones (Collins & Loftus, 1975; Kumar, 2021; McNamara, 2005).

The strength of activation is influenced by priming effects: the impact of the initial activation (i.e., the prominence in the speaker’s utterance), the strength of the connections between words themselves (i.e., the linguistic and cultural conventions that connect the words), and the recency and frequency of prior activations (i.e., the listener’s familiar connections with similar words and concepts). Activation isalso inhibited by the recognition of incoming contextual information that renders many connections irrelevant (Christensen & Kenett, 2021; Fernandino et al., 2022; Quilty-Dunn, 2021; Ritvo et al., 2019).

In recent psychological research, the term spreading activation is often used alongside terms such as semantic priming (Grimmer et al., 2022), semantic memory access (Kumar, 2021), and lexical network access (Pranoto & Afrilita, 2019). All are referring to the essential cognitive process of recruiting connections in our mental lexicon to allow us to utilize the richness of lexical knowledge as we listen.

About The Author

, Lexical Access, Lateral Communications
Michael Rost, principal author of Pearson English Interactive, has been active in the areas of language teaching, learning technology and language acquisition research for over 25 years. His interest in bilingualism and language education began in the Peace Corps in West Africa and was fuelled during his 10 years as an educator in Japan and extensive touring as a lecturer in East Asia and Latin America. Formerly on the faculty of the TESOL programs at Temple University and the University of California, Berkeley, Michael now works as an independent researcher, author, and speaker based in San Francisco.

More Posts You May Find Interesting