π Lecture 04 β DIMENSIONS OF WORD KNOWLEDGE
π Overview: This lecture explores the multiple dimensions of word knowledge in psycholinguistics, including phonological, syntactic, and morphological components. It examines how we store and retrieve phonological information, how syntactic categories organize our mental lexicon, and how morphological awareness can be taught to improve language and literacy skills.
ποΈ Topics Covered
The lecture covers five main topics: phonological knowledge and the tip-of-the-tongue phenomenon, conceptual and empirical issues surrounding phonological theory (including four main views), syntactic knowledge including open-class and closed-class words, children's acquisition of syntactic knowledge comparing generative and usage-based approaches, and morphological knowledge including inflectional and derivational morphemes along with evaluation of morphological awareness intervention effectiveness.
π Lecture Summary
Topic-019: Phonological Knowledge
Phonological knowledge refers to our knowledge of the pronunciation or sound structure of words. For example, we know when two words are homophones, which are words that are spelled differently but sound alike (such as bare and bear). This knowledge is evidenced by the tip-of-the-tongue (TOT) phenomenon, which occurs when we are not quite successful at retrieving a particular word but can remember something about how it sounds. William James described this state vividly: "There is a gap therein, but no mere gap. It is a gap that is intensely active. A sort of wraith of the name is in it, beckoning us in a given direction." Brown and McNeill systematically studied the TOT phenomenon for the first time by presenting definitions of infrequent words (such as sextant) and asking subjects to produce the defined word. When subjects were in the TOT state, they retrieved but rejected similar-sounding words such as secant. Thus, we sometimes activate words by their sounds, and when we make speech errors, we sometimes substitute a similar sounding word for the intended word.
π Definition β Homophones: words that are spelled differently but sound alike (e.g., bare and bear)
π Definition β Tip-of-the-tongue (TOT) phenomenon: a state where we are not quite successful at retrieving a particular word but can remember something about how it sounds
π Example: Brown and McNeill presented definitions of infrequent words such as sextant and asked subjects to produce the defined word. When subjects were in the TOT state, they retrieved but rejected similar-sounding words such as secant.
Topic-020: Phonological Knowledge: Conceptual and Empirical Issues
Conceptual and empirical issues address the central question: "Is phonological knowledge different from linguistic knowledge in general?" Burton-Roberts, Carr, and Docherty have shown that there are at least four main views concerning the goals and methods of phonological theory:
(a) Isolative view β held by Bromberger and Halle, and endorsed by Chomsky. This view considers phonology to be different from the rest of the language faculty, belonging to the Periphery rather than the Core.
(b) Functional view β held by Burton-Roberts, and supported by the work of Ohala, Lindblom, Bybee, and Hayes, among others. This view considers phonology to be grounded in the sensorimotor apparatus.
(c) Formal view β represented by Hale and Reiss, and supported by research on sign language phonology. This view considers phonology to be like the rest of language faculty and grounded in the mind. The extreme version excludes phonetic substance entirely from phonology.
(d) Unified view (Integrational) β held by Pierrehumbert, Beckman, and Ladd. This view considers phonology to be a laboratory science and argues for the adoption of scientific methodology to test formal claims. This approach seeks a bind between phonology, psycholinguistics, and sociolinguistics.
π‘ Why this matters: These four views represent fundamentally different assumptions about whether phonological knowledge is innate or learned, mental or physical, and whether it can be studied in isolation or must be integrated with other scientific disciplines.
Topic-021: Syntactic Knowledge
Our knowledge of words includes the syntactic category, or part of speech, to which they belong. Two words belong to the same syntactic category when they can substitute for one another in a sentence. For example:
(1) The aging pianist stunned the audience.
We can replace aging with any number of words, such as wealthy, poor, fat, solemn, and so on. Although the substitutions may change the meaning, the sentence remains grammatical. One advantage of using syntactic categories is that we can formulate grammatical rules in terms of categories rather than lexical items. The rule is that adjectives may modify nouns, not that aging may appear before pianist.
Traditional grammatical theory recognizes eight syntactic categories: noun, verb, adjective, adverb, pronoun, preposition, conjunction, and interjection. From a psychological vantage point, these are placed into two groups:
- Open-class words (content words): include nouns, verbs, adjectives, and adverbs. We have all learned a large number of these, and that number continues to grow.
- Closed-class words (function words): include determiners, pronouns, prepositions, conjunctions, and interjections. These are much smaller in numberβa few hundred in Englishβand are used over and over.
Neurologists have found that some patients suffer from agrammatism. Agrammatic patients frequently omit closed-class words (and inflectional endings) from their sentences while preserving open-class words somewhat better. They also process closed-class words differently than individuals without neurological damage.
π Definition β Agrammatism: a condition where patients frequently omit closed-class words and inflectional endings from their sentences while preserving open-class words somewhat better
π Example: In the sentence "The aging pianist stunned the audience," aging can be replaced by wealthy, poor, fat, or solemn, and the sentence remains grammatical because all are adjectives that can modify nouns.
Topic-022: Childrenβs Acquisition of Syntactic Knowledge
Children babble, pass through a single and multiword stage, and then start to produce entire sentences that increase in complexity. Two current approaches to language acquisition are introduced:
1. The Generative Approach β follows the theory of Universal Grammar advanced by Noam Chomsky. This theory assumes that children are "hardwired" with linguistic knowledge that gives them access to structural representations in the absence of experience.
2. The Usage-Based Account (constructivist approach) β promoted by Elena Lieven, Michael Tomasello, and others. This theory argues that children learn sentence structure through experience. It assumes the child has no specialized knowledge of language or syntax and must learn this on the basis of positive input alone. This is a slow process because children must gradually build up knowledge of constructions.
Chomsky proposed three questions that need to be addressed:
- What constitutes knowledge of language?
- How is knowledge of language acquired?
- How is knowledge of language put to use?
The theories depart in their perspective on whether acquisition is guided partly by innate knowledge (nature) or whether all knowledge is learned through experience (nurture)βknown as the "nature versus nurture" controversy. A widely shared assumption is that exposure to language and interaction with speakers is essential for acquisition to proceed. Speakers provide positive input in the form of utterances and their corresponding meanings. The dispute is whether positive evidence alone is sufficient for children to achieve mastery of adult grammar.
π‘ Why this matters: This debate represents one of the most fundamental controversies in all of cognitive scienceβwhether the human mind comes pre-equipped with specialized language knowledge or whether language is learned entirely through general cognitive mechanisms and experience.
π Definition β Positive input: linguistic input provided by speakers (caretakers, siblings, etc.) to the child in the form of utterances and their corresponding meanings
Topic-023: Morphological Knowledge
Morphemes are the smallest unit of meaning in a language. Some words consist of just a single morpheme. Morphemes that are also words are called free morphemes. Bound morphemes are those that are attached to free morphemes to create new words. There are two different kinds of bound morphemes:
- Inflectional morphemes: added to a free morpheme to express grammatical contrasts in sentences. Examples include the plural morpheme for nouns (cat/cats) and the past tense morpheme for verbs (jump/jumped).
- Derivational morphemes: bound morphemes added to free morphemes that create new words. For example, -ness turns good (an adjective) into goodness (a noun). Some derivational morphemes change not only the syntactic category but also pronunciation. For example, -ion changes decide (a verb) into decision (a noun), and the second /d/ in decide becomes the /Κ/ in decision.
When a word contains both inflectional and derivational morphemes, the derivational morphemes are applied first. Consider the word neighborhoods. The root word is neighbor, and both the derivational morpheme -hood and the inflectional morpheme -s are applied. The derivational is applied first, so the resulting word is neighborhoods, not neighbors hood.
Regarding vocabulary size, our ability to form various alternative forms effectively means there is no limit to the number of new words. Using the criterion of whether it would be possible to determine the meaning of a word with a morpheme by knowing its root, it is estimated that the average high school graduate knows about 45,000 words. The number is likely somewhat higher in college graduates and for those who do a lot of reading.
π Definition β Morphemes: the smallest unit of meaning in a language
π Definition β Free morphemes: morphemes that are also words (can stand alone)
π Definition β Bound morphemes: morphemes that are attached to free morphemes to create new words
π Definition β Inflectional morphemes: bound morphemes added to express grammatical contrasts in sentences (e.g., plural -s, past tense -ed)
π Definition β Derivational morphemes: bound morphemes that create new words and may change syntactic category or pronunciation (e.g., -ness, -ion)
π Example: In neighborhoods, the root is neighbor, the derivational morpheme -hood is applied first, then the inflectional morpheme -s is applied. The result is neighborhoods, not neighbors hood.
π Example: The derivational morpheme -ion changes decide (verb) into decision (noun) and changes pronunciation: the second /d/ in decide becomes /Κ/ in decision.
Topic-024: Evaluating the Effectiveness of a Morphological Awareness Intervention
Intervention should begin with an introduction of the concept of morphology and provision of many relevant examples. This should include emphasis on the importance of morphology, explanation of target patterns (inflectional or derivational), and provision of multiple examples. Depending on the grade and cognitive level, jargon terminology may be taught directly (e.g., derivative, suffix) or simpler terms may be used (e.g., base word, word ending). The rationale should be provided that students will learn about morphemes to increase vocabulary skills and/or become better readers and spellers. A base word can stand all by itself and tells us what the word is aboutβit is the "power" of the word.
Goodwin and Ahn found morphological awareness instruction to be particularly effective for children with speech, language, and/or literacy deficits. School-age children who received explicit instruction improved significantly in phonological awareness, morphological awareness, vocabulary knowledge, and related literacy areas of reading and spelling.
Several features of effective morphological awareness instruction include:
- Integration of problem solving or a motivating "detective" theme
- Explicit focus on morphological meaning units
- Incorporation of morphological awareness in contextual or language-/literacy-related instruction
Individualized morphological awareness instruction that integrated a problem-solving detective theme and included an explicit focus on morphology in a contextualized manner appeared to improve the language and literacy skills of an eight-year-old boy with deficits. Clinically significant improvements were found in phonological awareness, vocabulary, and reading comprehension.
Instruction in morphological awareness improves phonological awareness because morphologically based instruction incorporates the awareness of phonemes by linking this information to meaning. For example, when speaking of the relationship between music and musician, the base word music is used to ascertain the meaning of the derived form musician, despite the shift in phonology (letter c from /k/ to /Κ/ in the derived form). Thus, morphological awareness may help to mediate or scaffold a child's understanding of the phonological relationship between words.
Explicit vocabulary instruction based in morphological awareness may be especially important for children with language and literacy deficits. Intervention should focus on both recognition of meaning and patterns in words and include production activities such as word building. A link to reading and spelling provides a functional context for students to apply their newly learned morphological awareness skills.
π Example: The relationship between music and musician: the base word music is used to ascertain the meaning of musician, despite the shift in phonology where the letter c changes from /k/ to /Κ/.
β Key Takeaways
Phonological knowledge involves understanding word sounds and is evidenced by the tip-of-the-tongue phenomenon and homophone recognition. Syntactic knowledge organizes words into open-class (content words) and closed-class (function words) categories, with agrammatic patients showing selective difficulty with closed-class words. The nature vs. nurture debate in language acquisition contrasts the generative approach (Universal Grammar, innate knowledge) with the usage-based constructivist approach (learning through experience and positive input). Morphological knowledge involves understanding morphemes as the smallest units of meaning, with inflectional morphemes expressing grammatical contrasts and derivational morphemes creating new words while applying before inflectional morphemes. Morphological awareness instruction, particularly with a problem-solving detective theme and contextualized focus, effectively improves phonological awareness, vocabulary, reading, and spelling in children with language and literacy deficits.
π§ Quick Revision Questions
- What is the tip-of-the-tongue phenomenon and how did Brown and McNeill study it systematically?
- What are the four main views concerning phonological theory, and how does each view differ in its understanding of phonological knowledge?
- What is the difference between open-class words and closed-class words, and what happens to these word types in agrammatic patients?
- How do the generative approach and the usage-based constructivist approach differ in their explanations of how children acquire syntactic knowledge?
- What is the difference between inflectional and derivational morphemes, and why must derivational morphemes be applied before inflectional morphemes when both are present in a word?
π Lecture 05 β Organization of the Internal Lexicon
π Overview: This lecture explores how word meanings are organized and represented in the human mind, known as the internal lexicon. It examines semantic knowledge, network models, and measurement tools that help psycholinguists understand the cognitive architecture of meaning. Understanding these concepts is crucial for grasping how humans process, store, and retrieve lexical information.
ποΈ Topics Covered
The lecture begins by examining semantic knowledge including sense and reference, then introduces semantic network models as graphic representations of knowledge. It covers hierarchical network models and spreading activation models as competing theories of lexical organization. The lecture concludes with the Semantic Differential Scale for measuring word meanings and an analysis of semantic barriers in communication.
π Lecture Summary
Topic-025: Semantic Knowledge
This section explores what it means to know the meaning of a word and how meaning is represented mentally. Linguists, philosophers, and psychologists have identified several important aspects of word meaning, beginning with the distinction between sense and reference.
The reference of a word concerns the relationship between words and things in the world; the things in the world are called the referents of the word. This aspect of meaning is crucial for determining the truthfulness of utterances. For instance, when understanding sentence (2): "There is a brown cow grazing in the field," we grasp its truth conditions β the conditions under which the sentence may be said to be true. There must be a cow, it must be brown, and it must be grazing in the field. Reference concerns what the world should be like if a given utterance is true. However, abstract words like justice, plausibility, and relativity have meaning but lack clear referents.
Meaning is developed through the following ways: Synonymy (words with similar meanings), Hyponymy (hierarchical relationships like animal-dog), Meronymy (part-whole relationships), Taxonomic Relations (classification systems), Attributive Relations (property associations), Functional Relations (purpose-based connections), Denotation (literal dictionary meaning), and Connotation (associated emotional/cultural meanings).
π Definition β Reference: The relationship between words and things in the world; what the world should be like if a given utterance is true.
π Definition β Referents: The actual things in the world to which a word points.
π Definition β Truth Conditions: The conditions under which a sentence may be said to be true.
π Example: For "There is a brown cow grazing in the field" β truth conditions require (1) a cow exists, (2) it is brown, (3) it is grazing, and (4) it is in the field. All conditions must be met for the sentence to be true.
Topic-026: The Concept of a Semantic Network
A semantic network or net is a graphic notation for representing knowledge in patterns of interconnected nodes and arcs. Computer implementations were first developed for artificial intelligence and machine translation, but earlier versions have long been used in philosophy, psychology, and linguistics. Common to all semantic networks is a declarative graphic representation used either to represent knowledge or to support automated systems for reasoning about knowledge.
Six common kinds of semantic networks are identified:
Definitional networks emphasize the subtype or is-a relation between a concept type and a newly defined subtype. Also called a generalization or subsumption hierarchy, they support the rule of inheritance for copying properties defined for a supertype to all subtypes. Since definitions are true by definition, information in these networks is often assumed necessarily true.
Assertional networks are designed to assert propositions. Unlike definitional networks, information in an assertional network is assumed contingently true, unless explicitly marked with a modal operator. Some assertional networks have been proposed as models of the conceptual structures underlying natural language semantics.
Implicational networks use implication as the primary relation for connecting nodes. They may represent patterns of beliefs, causality, or inferences.
Executable networks include mechanisms such as marker passing or attached procedures that can perform inferences, pass messages, or search for patterns and associations.
Learning networks build or extend representations by acquiring knowledge from examples. New knowledge may change the old network by adding and deleting nodes and arcs or by modifying numerical values called weights associated with the nodes and arcs.
Hybrid networks combine two or more of the previous techniques, either in a single network or in separate but closely interacting networks.
Some networks implement hypotheses about human cognitive mechanisms, while others are designed primarily for computer efficiency. The distinction between definitional and assertional networks parallels the distinction between semantic memory (general world knowledge) and episodic memory (personal experiences).
π‘ Why this matters: Semantic networks provide a bridge between computational models and human cognitive processes, helping researchers understand how the brain might organize millions of word meanings.
π Definition β Semantic Network: A graphic notation for representing knowledge in patterns of interconnected nodes and arcs.
π Definition β Rule of Inheritance: The principle that properties defined for a supertype apply to all its subtypes.
Topic-027: Hierarchical Network Models
A semantic network is an interconnected web of concepts connected by various relations. In the hierarchical network model, knowledge of words is stored in the form of a semantic network with some words represented at higher nodes (more general categories) and others at lower nodes (more specific categories). For example, "animal" would be at a higher node than "bird," which would be higher than "canary." Properties are stored at the highest possible node to avoid redundancy β for instance, "has wings" is stored with "bird" rather than with every specific bird type.
Although the hierarchical network model can explain some experimental results, it is too rigid to capture all of our tacit knowledge of the lexicon. It cannot easily account for typicality effects (some category members are considered "better" examples than others) or for the fact that people can quickly verify properties that are not stored at the most efficient hierarchical level.
π‘ Why this matters: The hierarchical model was an early influential attempt to model lexical organization, but its limitations led to the development of more flexible spreading activation models.
π Definition β Hierarchical Network Model: A semantic network where words are stored at different levels (nodes) in a hierarchy, with general categories at higher nodes and specific instances at lower nodes.
Topic-028: Spreading Activation Models
Spreading activation models are network models that are not strictly hierarchical. In these models, activation spreads from one node to neighboring nodes, with the strength of activation decreasing as it travels further from the original node. When a word is accessed, activation radiates outward through the network, activating related concepts. This explains priming effects β why hearing "doctor" makes "nurse" faster to recognize (because activation spreads from the "doctor" node to the closely related "nurse" node).
Spreading activation models of the lexicon that incorporate conceptual, syntactic, and phonological knowledge appear to offer the most realistic picture currently available of the internal lexicon. These models can account for a wider range of experimental findings than hierarchical models, including typicality effects, context effects, and the speed of lexical access.
π‘ Why this matters: Spreading activation models better capture the dynamic, context-sensitive nature of human word processing and have become the dominant theoretical framework in psycholinguistics.
π Definition β Spreading Activation: The process by which activation radiates from one node to neighboring nodes in a semantic network, with decreasing strength as distance increases.
π Example: If a person hears the word "bread," activation spreads to related nodes like "butter," "toast," "bakery," and "food." This is why "butter" is recognized faster immediately after hearing "bread" compared to hearing an unrelated word like "lamp."
Topic-029: Advantages and Disadvantages of Semantic Differential Scale
Osgood's Semantic Differential was an application of measuring the semantics or meaning of words, particularly adjectives, and their referent concepts. The respondent is asked to choose where their position lies on a scale between two polar adjectives (for example: "Adequate-Inadequate," "Good-Evil" or "Valuable-Worthless"). Semantic differentials can be used to measure opinions, attitudes, and values on a psychometrically controlled scale.
Osgood and colleagues performed a factor analysis of large collections of semantic differential scales and found three recurring attitudes that people use to evaluate words and phrases: evaluation, potency, and activity. Evaluation loads highest on the adjective pair 'good-bad'. The 'strong-weak' adjective pair defines the potency factor. The adjective pair 'active-passive' defines the activity factor. These three dimensions of affective meaning were found to be cross-cultural universals in a study of dozens of cultures.
The Semantic Differential question scale offers a bipolar pair of adjectives between which the respondent must choose along some form of scaling (typically a five-point scale). The pairs of words need not be opposites and may be used to discover fine differences in viewpoint.
Advantages: Provides a standardized, quantifiable measure of word meanings; reveals underlying dimensions of affective meaning; cross-culturally valid; useful for measuring attitudes and opinions.
Disadvantages: Limited to adjectives and evaluative dimensions; may not capture all aspects of meaning; respondents may be influenced by social desirability; the bipolar structure may force choices that don't reflect actual perceptions.
π Definition β Semantic Differential: A measurement tool where respondents rate concepts on scales between two polar adjectives to measure connotative meaning.
π Definition β E-P-A (Evaluation-Potency-Activity): The three universal dimensions of affective meaning: Evaluation (good-bad), Potency (strong-weak), and Activity (active-passive).
π Example: To measure attitudes toward "democracy," a respondent might rate it on scales like: good-bad, strong-weak, active-passive. A score of 7 on "good" and 1 on "bad" would indicate a very positive evaluation.
Topic-030: Semantic Barriers
The meaning of words, signs, and symbols might differ from one person to another, and the same word might have hundreds of meanings. When a message is sent by a sender to a receiver, it might be interpreted wrongly in a communication process, causing misunderstandings. This can happen due to different situations that form the semantic (of, relating to, or arising from the different meanings of words or other symbols) of the sender and the receiver, known as the semantic barrier.
Semantic barriers arise due to differences in language, education, culture, and place of origin (dialect or accent), or most likely, people's experiences. For example, the word "cool" might mean "low temperature" to one person but "excellent" or "fashionable" to another. These barriers can lead to miscommunication, confusion, and conflict in both personal and professional settings.
π‘ Why this matters: Understanding semantic barriers is essential for effective communication across cultural, educational, and linguistic divides, and highlights why psycholinguistic research on meaning is practically important.
π Definition β Semantic Barrier: Misunderstandings in communication arising from different interpretations of words, signs, and symbols due to differences in language, education, culture, dialect, or experience.
π Example: In British English, "boot" refers to a car's trunk; in American English, "boot" refers to footwear. A British speaker saying "Put it in the boot" might confuse an American listener expecting footwear storage rather than car storage.
β Key Takeaways
The internal lexicon organizes word meanings through complex networks rather than simple dictionaries. Hierarchical network models were early attempts to represent this organization but proved too rigid; spreading activation models better capture the dynamic, interconnected nature of lexical access where activation radiates between related concepts. Semantic knowledge includes both reference (word-world relationships) and sense (meaning relationships like synonymy and hyponymy). Osgood's Semantic Differential reveals that affective meaning universally organizes along three dimensions: evaluation, potency, and activity. Finally, semantic barriers highlight that meaning is not fixed but shaped by individual experience, culture, and contextβmaking effective communication dependent on shared understanding.
π§ Quick Revision Questions
- What is the difference between sense and reference in semantic knowledge?
- What are the six types of semantic networks, and how do definitional and assertional networks differ?
- Why are spreading activation models considered more realistic than hierarchical network models for explaining lexical access?
- What are the three universal dimensions of affective meaning identified by Osgood's Semantic Differential, and which adjective pairs define them?
- What causes semantic barriers in communication, and why do the same words lead to different interpretations?
π Lecture 6 β Models of Lexical Access
π Overview: This lecture explores how individuals access and retrieve words from their mental lexicon, focusing on two major competing theoretical frameworks: serial search models and parallel access models. It also examines key variables that influence lexical access, the nature of lexical ambiguity, and the stages involved in speech production. Understanding these models and processes is crucial for comprehending how humans produce and comprehend language in real time.
ποΈ Topics Covered
The lecture begins by introducing the two major classes of models of lexical access: serial search models and parallel access models, with detailed accounts of Forster's autonomous search model, Morton's Logogen model, and Marslen-Wilson's Cohort theory. It then discusses the differences between Logogen and Cohort models, followed by an examination of variables that influence lexicon access, including word frequency, phonological variables, syntactic category, morphological complexity, and semantic priming. The topic of lexical ambiguity is addressed, explaining how ambiguous words are processed. Finally, the lecture appraises the models of lexical access and outlines the four stages of lexis in speech production: conceptualization, formulation, articulation, and self-monitoring.
π Lecture Summary
Topic-031: Models of Lexical Access
Lexical access is an area in psycholinguistics that studies the activation or retrieval process of the mental lexicon for people who can speak two languages. Lexical retrieval is the process of getting from an abstract concept to a spoken word. There are two major classes of models that detail how lexical entries are retrieved during reading and listening tasks: serial search models and parallel access models.
Serial search models believe that when we encounter a word, we look through all lexical entries to determine whether the item is a word or not and then retrieve the necessary information about a word (i.e., its semantics or orthography). They propose that lexical access occurs by sequentially scanning one lexical entry at a time.
One of the earliest and most influential serial models is Forster's autonomous search model. In this model, the word recognition system is divided into several different components. One is devoted to the orthographic (spelling) properties of a word and another to the phonetic properties. Each of these is organized in descending order of frequency. Thus, more frequent words are searched before lower-frequency words. When the input is matched to one of the items in one of the two bins, a pointer to an entry in the master lexicon is retrieved, along with other properties like its syntactic function. Forsterβs model assumes that the lexicon is autonomous or independent of other systems involved in language processing. Activation of words is not directly influenced by syntactic or semantic factors; such factors affect the general cognitive system, and information from the lexicon is fed into this system. The revised model has separate comparators for each file bin.
In contrast, parallel access models propose that perceptual input about a word activates lexical items directly, and that multiple entries can be activated at once. A number of potential candidates are activated simultaneously, and the lexical item which shares the most features with the targeted stimulus is the one that is chosen. Examples include Marslen-Wilsonβs cohort model, McClelland & Seidenberg's connectionist model, and Mortonβs logogen model.
π‘ Why this matters: At the present time, there is greater acceptance toward the parallel access models than the serial search models when explaining lexical access.
π Definition β Lexical Access: The activation or retrieval process of the mental lexicon. π Definition β Lexical Retrieval: The process of getting from an abstract concept to a spoken word. π Definition β Serial Search Models: Models proposing that lexical access occurs by sequentially scanning one lexical entry at a time. π Definition β Parallel Access Models: Models proposing that perceptual input activates lexical items directly and that multiple entries can be activated at once.
Topic-032: Difference Between Logogen and Cohort Model
An important aspect of Mortonβs Logogen Model is that both sensory and contextual information interact in such a way that there is a trade-off relationship between them. The more contextual information input to a Logogen from top-down sources, the less sensory information is needed to bring the Logogen above threshold for activation. This feature enables it to account for facilitation effects of syntactic and semantic constraints on speed of lexical access, as well as word frequency and word apprehension effects. In the presence of constraining prior contexts, the time needed to activate a logogen from the onset of the relevant sensory information will be less because less sensory information will be necessary to bring the logogen above its threshold value.
In contrast to Logogen Theory, which assumes activation of only a single lexical item after its threshold value is reached, Cohort Theory views word recognition as a process of eliminating possible candidates by deactivation. According to Marslen-Wilson and Welsh, the set of word-initial cohorts consists of the entire set of words in the language that begin with a particular initial sound sequence. A word is recognized at the point that it can be uniquely distinguished from any other words in the word-initial cohort set. This is known as the βcritical recognition pointβ of a word. Upon first hearing a word, all words sharing the same initial sound characteristics become activated in the system. As the system detects mismatches between the initial bottom-up sensory information and the top-down information, inappropriate candidates are deactivated.
In Cohort Theory, as in Logogen Theory, word recognition and subsequent lexical access are viewed as a result of a balance between available sensory and contextual information. When deactivation occurs on the basis of contextual mismatches, less sensory information is needed for a single word candidate to emerge. Once word recognition has occurred, the perceptual system carries out a much less detailed analysis of the remaining input.
In Mortonβs model, each word (or morpheme) in the lexicon is represented as a logogen, which specifies the wordβs various attributes (semantic, orthographic, phonological, etc.). The logogen is activated in either of two ways: by sensory input or by contextual information. As orthographic or phonological features are detected, they are matched to the logogen. The logogen functions as a scoreboard; when the counter rises above a predesignated threshold, the item is recognized. For example, in the sentence "Her closest relative was appointed as her legal guardian," the word guardian can be anticipated due to expectations created by earlier words, temporarily lowering its threshold. Thus, it is easier to recognize guardian in this context than in isolation. In the original model, information about associations between words is contained in a separate cognitive system. These two routes work in parallel; when many sensory features are detected (loud presentation) or when a word is expected, the corresponding word will be activated.
π Example: For "Her closest relative was appointed as her legal guardian," the logogen for guardian receives activation from both sensory features (the sound of the word) and contextual features (the sentence meaning), temporarily lowering its threshold and making it easier to recognize.
π Definition β Logogen: A representational unit in the lexicon that specifies a word's attributes and is activated by sensory or contextual input. π Definition β Cohort: The set of all words in a language that begin with a particular initial sound sequence. π Definition β Critical Recognition Point: The point at which a word can be uniquely distinguished from any other words in the word-initial cohort set.
Topic-033: Variables That Influence Lexicon Access
The lecture lists the following variables that affect lexicon access:
- Word Frequency
- Phonological Variables
- Syntactic Category
- Morphological Complexity
- Semantic Priming
Topic-034: Lexical Ambiguity
Lexical ambiguity refers to the form of ambiguity in which a single word may be interpreted to have more than one meaning. Foss was the first to apply the phoneme-monitoring technique to the study of lexical ambiguity. He presented listeners with sentences containing ambiguous words, such as "The man started to drill before the truck arrived."
π Definition β Lexical Ambiguity: A form of ambiguity in which a single word has more than one meaning. π Definition β Phoneme-Monitoring Technique: A research method where listeners detect a specific phoneme in a sentence, used to study processing of ambiguous words.
Topic-035: Appraising Models of Lexical Access
Lexical access is influenced by a variety of factors, including the frequency of a word, its phonological structure, its syntactic category, its morphological structure, the presence of semantically related words, and the existence of alternative meanings of the word. Common words and meanings appear to be in a state of greater readiness than less-often-used words and meanings. We rely on morphological structure when encountering unfamiliar words. Considerable research has investigated how we access lexically ambiguous words. Some research suggests that we briefly consider all meanings of an ambiguous word. However, when a preceding context primes the most dominant meaning of a word, lexical access may be selective.
π‘ Why this matters: This appraisal shows that while some models are supported by certain findings (e.g., frequency effects), others (e.g., models of ambiguity resolution) are more complex and context-dependent, highlighting the need for hybrid or more sophisticated models.
Topic-036: Stages of Lexis in Speech Production
Stage 1 β Conceptualization: This is when a speaker spontaneously thinks of what they are going to say. It is an immediate reaction to external stimuli and is often based on prior knowledge. No premeditation goes into these words, and they are all formulated based upon the speakerβs knowledge and experience at hand. Examples range from answering questions to the immediate verbiage produced as a result of stubbing your toe.
Stage 2 β Formulation: This is when the speaker thinks of the particular words that are going to express their thoughts. It occurs almost simultaneously with the conceptualization stage, but this time the speaker thinks about the response before responding. The speaker is formulating their words and deciding how to reply to the external stimuli in the best manner. Where conceptualization is more of an instant response, formulation is a little delayed.
Stage 3 β Articulation: This is when the speaker physically says what they have thought of saying. This is a prepared speech or planned wordage. The words may have been rehearsed, such as when someone practices a presentation or rehearses a lie. It involves the training of physical actions of several motor speech organs such as the lungs, larynx, tongue, lips, and other vocal apparatuses. The articulation stage uses these organs multiple times for the same word patterns.
Stage 4 β Self-Monitoring: This is when the speaker reflects on what they have said and makes an effort to correct any errors in their speech. This is the action of reflecting on what you said and making sure that what you said is what you meant.
π Definition β Conceptualization: The stage of speech production where the speaker spontaneously thinks of what to say. π Definition β Formulation: The stage where the speaker selects the specific words to express their thoughts. π Definition β Articulation: The stage where the speaker physically produces the speech using motor speech organs. π Definition β Self-Monitoring: The stage where the speaker reflects on their speech and corrects errors.
β Key Takeaways
The most critical points to remember from this lecture are the fundamental distinction between serial search models (sequential scanning of entries) and parallel access models (simultaneous activation of multiple candidates). Forster's autonomous search model exemplifies the serial approach, while Morton's Logogen model and Marslen-Wilson's Cohort theory are key parallel models. The Logogen model features a threshold mechanism where sensory and contextual information combine to activate a word, whereas the Cohort theory involves deactivation of mismatching candidates from an initial set. Finally, lexical access is influenced by multiple variables (e.g., frequency, morphology, ambiguity), and speech production progresses through four distinct stages: conceptualization, formulation, articulation, and self-monitoring.
π§ Quick Revision Questions
- What is the core difference between serial search models and parallel access models of lexical access?
- In Forster's autonomous search model, what is the role of separate comparators in the revised version?
- How does the Logogen model explain the facilitation effect of context on word recognition?
- According to Cohort Theory, what is the "critical recognition point" and how is it determined?
- List the four stages of speech production and briefly describe what happens in each stage.