Visual and Spoken Word Recognition and Lexical Organization

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/24

flashcard set

Earn XP

Description and Tags

Flashcards covering visual word recognition factors, reading models, the organization of the mental lexicon, pre-lexical speech processing, and spoken word recognition models.

25 Terms

1
New cards

What is the word superiority effect, and how does phonotactic legality influence nonword recognition (e.g., "Wrod" vs. "Wlod")?

Words or chunks that already exist in the mental lexicon are recognized faster than nonwords due to a top-down processing pathway from words to their constituent letters. A legal nonword like "Wrod" is processed more efficiently than an illegal nonword like "Wlod" because "wr" is an allowable consonant cluster in English, demonstrating that the mental lexicon contains knowledge of permissible sound sequences.

2
New cards

How does word length affect recognition speed in lexical decision and naming tasks?

Word length displays a mixed, U-shaped effect: shorter words of 33–55 letters yield faster responses, whereas longer words of 88–1313 letters produce longer response times.

3
New cards

How does word frequency impact recognition, and what confound is introduced by Zipf's effect?

Higher frequency words are recognized more easily and reliably than low-frequency words. Because high-frequency words tend to be shorter according to Zipf's effect, comparing words like "dog" and "elephant" introduces confounding factors stemming from both length and frequency.

4
New cards

What is the regularity effect, and how does it interact with word frequency?

The regularity effect describes how words with basic spelling-sound correspondences (e.g., "ea" in "bleach") are recognized more easily than irregular words (e.g., "ea" in "spread"). Low-frequency words are strongly affected by regularity, whereas high-frequency words are typically unaffected.

5
New cards

What defines an orthographic neighbor, and what are examples for the target word "work"?

Orthographic neighbors are words that share all letters with the target word except one, with the differing letter occurring in the identical position. For example, "word", "ward", and "ford" are all orthographic neighbors of "work".

6
New cards

How does high neighborhood density influence the identification of a low-frequency target word?

Having high neighborhood density helps activate and detect a low-frequency word faster; however, in tasks such as the progressive demasking task, longer exposure to the target word is required to uniquely identify it among its competitors.

7
New cards

What are the core properties and key weaknesses of Morton's logogen model?

Words are represented by logogens (units that collect orthographic, phonological, and contextual input in parallel) that trigger recognition once an activation threshold is reached. Its primary weaknesses are failing to properly define what a logogen is and failing to account for neighborhood density effects, as it assumes no interaction between logogens.

8
New cards

How is word retrieval structured in Forster's search model, and why can it not explain neighborhood density effects?

It is a serial search model where separate access files (orthographic, phonological, syntactic/semantic) organized by factors such as frequency are examined one by one to reach a master file. Because search is strictly serial rather than simultaneous, words cannot be compared at the same time, leaving neighborhood density effects unexplained.

9
New cards

What structure and connection types are assumed by the Interactive Activation (IA) / TRACE model?

It uses a parallel processing architecture across three levels—words, letters, and letter features—governed by a combination of excitatory and inhibitory links. This accounts for both the word superiority effect and the Ganong effect.

10
New cards

In the dual-route model of reading, what distinguishes the direct route from the mediated route?

The direct route accesses the lexicon directly from written input for whole-word pronunciation of familiar words without spelling-to-sound conversion. The mediated route relies on rule-based grapheme-phoneme conversion (GPC) to translate written strings into pronunciation for nonwords and unfamiliar forms.

11
New cards

What are the four sequential steps in the mediated (grapheme-phoneme conversion) reading route?

  1. Written word input; 2. Grapheme identification; 3. Phoneme identification; 4. Stored word in the mental lexicon.
12
New cards

What is the pseudohomophone effect, and why does a nonword like "blud" take longer to reject in a lexical decision task?

Pseudohomophones are nonwords that sound identical to real words when pronounced (e.g., "blud" for "blood"). They take longer to reject because mandatory grapheme-to-phoneme conversion triggers an auditory match that interferes with the real lexical entry, necessitating an extra visual check against the orthographic representation.

13
New cards

Why are lowercase words recognized faster than uppercase words, and how does overall word shape affect proofreading typo detection?

Lowercase letters produce more distinctive word outlines than uppercase letters, speeding recognition. When a typing mistake retains the overall word shape (e.g., "tesf" instead of "test"), readers overlook the typo more frequently during proofreading than when the overall shape is disrupted (e.g., "tesc").

14
New cards
<p>What semantic network model is shown in this diagram, and what is its principle of "cognitive economy"?</p>

What semantic network model is shown in this diagram, and what is its principle of "cognitive economy"?

The Hierarchical Network Model (Collins & Quillian, 1969, 1970, 1972). "Cognitive economy" assumes that information is stored without redundancy at only one place in the network—specifically at the highest possible node where the property applies to all subordinates.

15
New cards

What empirical problems were identified in the Collins & Quillian Hierarchical Network Model?

It cannot explain the typicality effect (e.g., verification times are faster for "a robin is a bird" than for "an ostrich is a bird") or the familiarity effect, both of which violate the model's prediction that verification speed depends strictly on hierarchical distance.

16
New cards
<p>What model of semantic memory is depicted in this diagram, and what factors determine the distances between nodes?</p>

What model of semantic memory is depicted in this diagram, and what factors determine the distances between nodes?

The Spreading Activation Model (Collins & Loftus, 1975). Distance between nodes in this web-like network is determined by taxonomic relations, typicality, degree of association between concepts, and frequency of use, with activation propagating bidirectionally and in parallel.

17
New cards
<p>What are the three levels of lexical representation illustrated in the Bock &amp; Levelt (1994) model?</p>

What are the three levels of lexical representation illustrated in the Bock & Levelt (1994) model?

  1. Conceptual Level (semantic knowledge and attributes); 2. Lemma Level (grammatical category and syntactic specifications like gender); 3. Lexeme or Sound Level (phonological forms and individual phonemic segments).
18
New cards

Why are phonemes preferred over whole words as the unit of pre-lexical analysis in alphabetic languages?

Continuous speech lacks explicit acoustic boundaries between words (the segmentation problem) and is subject to perceptual errors (slips of the ear). Smaller and fewer units like phonemes permit more rapid and computationally efficient processing.

19
New cards

What is a gating experiment, and why is "soon" recognized faster than competitors like "soup", "suit", or "sued"?

A gating experiment presents increasingly longer fragments of a recorded word for identification at each successive "gate". "Soon" is recognized earlier because anticipatory lowering of the velum during the vowel [u][u] signals the forthcoming nasal consonant before it is fully articulated.

20
New cards

How do the Metrical Segmentation Strategy (MSS) and the Possible Word Constraint (PWC) function during pre-lexical speech analysis?

The Metrical Segmentation Strategy (MSS) posits that English word searches initiate at each strong or stressed syllable. The Possible Word Constraint (PWC) requires speech to be exhaustively segmented into viable words without leaving impossible, unattached residual consonants (e.g., finding "see" is easier in /siʃæb/ than /siʃ/).

21
New cards

What are the four progressive stages of spoken word recognition?

  1. Contact (linking acoustic input to stored word candidates); 2. Activation (making candidate words accessible); 3. Access (retrieving all lexically stored information and linking lemmas to lexemes); 4. Recognition (identifying the specific intended word).
22
New cards

How do lexical access, word recognition, and comprehension differ temporally and functionally?

Lexical access occurs prior to the uniqueness point when stored lemma information becomes available. Word recognition follows when a single candidate is selected. Recognition does not equate to comprehension, as recognizing a word does not guarantee contextual interpretation.

23
New cards

How does Cohort Theory describe spoken word recognition?

It is a parallel, bottom-up model in which the initial phoneme of an acoustic input activates all possible words starting with that sound (the cohort). As successive phonemes are perceived from left to right, incompatible words are discarded until a single unique word remains.

24
New cards

What visual competitor categories are displayed alongside the target in a visual world paradigm experiment testing phonological priming?

Participants view the referent (target word), a cohort competitor (shares initial phonemes), a rhyme competitor (shares final phonemes), and an unrelated distractor while listening to the auditory stimulus.

25
New cards

What is meant by the "contingency of choice" in spoken word recognition, and how is a phonological neighborhood defined?

"Contingency of choice" asserts that recognizing a word (e.g., "cat") depends not only on detecting its sounds, but also on confirming that competing cohort alternatives (e.g., "cap", "can", "cash") were not spoken. A phonological neighborhood comprises words differing from a target by a single phoneme (e.g., neighbors of "money" include "mummy", "many", "meanie", and "honey").