Word Recognition Flashcards
What is Spoken word recognition?
identifies phonetic structure in the incoming speech signal, allowing the signal to be mapped onto representations of known words in the listener’s LEXICON
Characteristics of speech make word recognition difficult:
The Segmentation Problem and The invariance problem
The Segmentation Problem
-speech is a continuous stream of sound. No white spaces between words
Possible word-constraint
Norris et al., (1997) suggests that possible word-constraint could ease word recognition by limiting the number of lexical candidates activated by a given input.
Norris et al., (1997) used a word spotting task to demonstrate that adults find words such as “apple” more easily in a possible condition (i.e., “vuffapple”) rather than in an impossible condition (i.e., “fapple”)
Any segmentation that results in impossible words is likely rejected for instance, “fapple”
The invariance problem
The invariance problem refers to the challenge that listeners face when confronted with acoustic variability in speech sounds as they attempt to map these sounds to few phonological categories.
Speech sounds (e.g., phonemes) are shaped by external factors
Co-articulation – the articulation of two or more speech sounds together, so that one influences the other
“b” in “bill”, “ball”, “bull” and “bell” is acoustically different
segmentation
It is difficult to delimit a stretch of speech as belonging to a single perceptual unit. As an example, the acoustic properties of the phoneme “d” will depend on the production of the following vowel (because of co-articulation).
Warren, (1970)
Knowledge about the context in which a word occurs (semantics, syntax) can help word recognition
Warren, (1970) demonstrated that if a sound, such as a cough or tone, replaces a speech sound, listeners believe they hear the missing sound or phoneme. This illusory percept of the missing phoneme is referred to as phoneme restoration.
The phonemic restoration effect refers to the tendency for people to hallucinate a phoneme replaces by a non-speech sound (e.g., a tone in a word). This illusion can be influenced by preceding sentential context providing information about the likelihood of the missing phoneme.
TRACE (McClelland & Elman, 1986) : theory of SWR
TRACE model – take all the various sources of information found in speech and integrate them to identify single words. The TRACE model is based on the principles of interactive activation.
Information can flow bottom-up and top-down across all three levels
TRACE model can explain context effects
This is because it allows higher-level information to affect lower-level information (top-down effects)
Cohort Model (Marslen-Wilson, 1984) : theory of SWR
According to the Cohort Model, a word is recognised by a successive reduction in the number of possible word candidates as each new phoneme is perceived. When only one candidate remains in the cohort of possible words, the word is recognised.
In the original Cohort Model, the selection stage is influenced by
Words beginning with “T” phoneme
Trespass
Trench
Trestle
Trend
The semantic or syntactic context (a word can be recognised before its uniqueness point if context supports only one candidate in the
In this sentence context “trespass” has a recognition point that comes before the uniqueness point
This is due to the semantic or syntactic context of the sentence
What is Visual word recognition?
In visual word recognition, a whole word may be viewed at once and recognition is achieved when the characteristics of the stimulus match the orthography (spelling) of an entry in the mental lexicon.
Interactive activation model (McClelland & Rumelhart, 1981) : VWR theory
- words are represented as nodes in a network that are connected by inhibitory links.
- Word units – letter units – feature units
- A model in which perception results from excitatory and inhibitory interactions of detectors for visual features, letters, and words.
Dual-route cascaded model (Coltheart et al., 2001) : VWR theory
Model of reading that can perform the 2 tasks most used to study reading: lexical decision and reading aloud
Route 1: Grapheme-phoneme conversion
- Conversion rules used to convert each grapheme into a phoneme
- Good for regular words and nonwords
- Bad for irregular words
Route 2: Lexicon and semantics
Route 3: Lexicon only
- Good for reading all familiar words
- Bad for reading unfamiliar words
- Orthographic input (spelling) lexicon stores the spelling of all words you know
- Activates meaning and phonology
Limitations of the DRC model:
- Semantics is important in reading, but DRC is vague about how or why
- Underestimates phonological influence
- Doesn’t explain how reading is learned
Connectionist triangle model (Harm & Seidenberg, 2004) : VWR theory
- Proposes that normal individuals rely on semantic information for correct reading of words with atypical spelling-sound relationships
- Connected speech consists of mapping from a representation of the speech input to a lexical or semantic representation of the words contained in a speech stream
Two possible routes to take you from spelling to sound:
- Direct pathway from spelling to phonology
- Indirect pathway from spelling to phonology via semantics
Conclusion
- All theories of reading involve orthography, phonology, and semantics to some degree, but these key components interact in different ways in different theories.
