Reading 2 Flashcards
why is phonemic awareness important for reading
- PA is essential for alphabetic insight “cracking the code”
- if children crack the alphabetic code they can use knowledge of spoken words for self-teaching - PA not spontaneously acquired through exposure to spoken words
e.g. Byrne et al (2000) :
– Teach children to identify BAT/ MAT
– Test ability to distinguish novel words eg BUG/MUG
– performance no better than chance EVEN THOUGH they can generalise for words (little/big) and morphemes (book, books)
e.g. only 25% of pre-literate children can match 1st phoneme
➔Children do not attend to individual letters of words; learn words as whole patterns not sets of letters/ phonemes
➔cannot generalise knowledge to new words
• If children fail to acquire phonemic awareness
– must use a “whole word” strategy➔ successful till ~Grade 2
BUT encounter difficulties when memory load increases
byrne & freebody (1988) study: chinese vs phoenician
“Chinese”: Irreg word> non word
“Phoenician”: non word> Irreg word
reading+listening comprehension:
Grade 2: Chinese> Phoenician
Grade 3: Phoenician > Chinese
reason: phoenician able to read for meaning, self-teach
stages of reading
logographic: words recognised by idiosyncratic global features
–> memory load, PA
alphabetic: letter-sound relationships learned, new words can be sounded out
–> practice, amalgamation of lexical codes
orthographic: words recognised automatically from orthographic features
–> attention can be directed to comprehension
why is automatic word identification important for reading comprehension
Reading requires many simultaneous processes
• access meanings of words
• maintain meanings in memory
• integrate meanings together in the light of current goals
• relate meanings to existing knowledge
Attentional resources are limited
• attention must be divided between simultaneous tasks
• if attention is overloaded, performance is poor
• to limit attentional requirements we automate “low level” components of the task eg driving a car, playing instrument, playing sport…
Automating a task requires
• that the task is “invariant”
• lots of practice
➔the component of reading that can be automated is word identification
simple view of skilled reading
reading = decoding x comprehension
word decoding: skilled readers have developed precise orthographic representations that support automatic word identification –> functionally modular
–>
limited attentional resources can be directed towards comprehension processes
Eye movements of skilled readers
Reading consists of sequence of fixations (av. ~200 ms) and saccades (~8 chars)
• fixate about 80% of content words (most function words eg if, to, of.. are short)
Skilled readers’ perceptual span ~18 letters: asymmetric ~15 right/3 left of fixation;
opposite for Hebrew
Visual acuity is only sufficient for letter discrimination in foveal vision (3-6 letters around fixation)➔lexical processing restricted to foveal vision
peripheral information used to guide where to land next fixation
Readers extract peripheral information about spaces, letter shapes, BUT NOT lexical or semantic information (?)
➔skilled readers read in a relatively word-by-word manner
➔ even skilled readers cannot read words in peripheral vision
➔ Skilled readers are NOT highly selective in their sampling of text
implications of eye movement
Skilled readers show more efficient eye movements than less skilled readers BUT
eye movements are more like poor readers when they read difficult material
- Inefficient eye movements are a consequence not a cause of poor reading
- Training eye movements will nor improve reading
- Efficient lexical retrieval is the foundation of skilled reading: lexical retrieval drives eye movements
- Words provide the building blocks for comprehension processes
a more complex simple view of skilled reading
The lexical system is functionally modular:
words are usually identified automatically from bottom-up information with little top-down influence of context
orthographic lexicon:
word-specific representations + subword orthorgraphic-phonological connections
comprehension process: inference + situation model + text representation + parse
top-down knowledge:
language/linguistic system (semantics, morphology, syntax, phonology) + general knowledge
does context change lexical access? i.e. is lexical access modular?
- Words ARE identified more easily in context
• Prior context facilitates identification of brief stimuli
• Semantic priming: lexical decision to nurse faster when preceded by doctor than butter - Two ways context could facilitate lexical access:
• Automatic spreading activation in lexical/semantic memory
• Attentional strategies eg guessing, prediction - Two process model of semantic priming (Posner & Snyder, 1975)*
• Automatic processes: fast, obligatory, parallel
– Pre-activate related items with no cost for unrelated
➔facilitation without interference
• Attentional processes: slow, attentionally mediated, serial
➔ increase benefit for related items as well as cost for unrelated
➔extra facilitation but also interference
Neely (1977) semantic category priming at different prime target intervals
participants told primes from furniture category followed by birds
four conditions: Expected semantically related eg FRUIT apple Expected semantically unrelated eg FURNITURE eagle Unexpected semantically related eg FURNITURE chair Unexpected semantically unrelated eg FRUIT horse
short interval (250ms): automatic priming, facilitation from semantically related prime both expected and unexpected
long interval (700ms): attentional effects, facilitation from expectancy (related or unrelated), inhibition from unexpected (both related or unrelated)
Stanovich (1986) study of reading development
priming in sentence context
TASK: Read the final word aloud
predictable: the banker locked the hidden safe
neutral: they said it was the hidden safe
incongruous: the train pulled into the hidden safe
• 4th and 6th grade children show inhibition as well as facilitation but adults only show (small) facilitation consistent with automatic spreading activation
➔Reliance on contextual prediction reduces with reading skill
stanovich & west (1983) skilled readers study
similar study, text presented in normal clear text or degraded text
• Facilitation for predictable completions; No inhibition for incongruous in normal text
➔ priming effects reflect automatic processes
• Inhibition for incongruous only observed when stimuli are degraded (effect expected from attentional rather than automatic)
➔adults only use active prediction when identification is difficult
implications of priming from sentence context
• Skilled readers DO NOT use contextual prediction to identify words
➔ Little evidence that context changes lexical access process
• Increasing skill is associated with reduced reliance on contextual prediction
➔Opposite to prediction of top-down model
context effects on ambiguous words: are both meanings of ambiguous words accessed even when context favours one?
- cross-modal priming (swinney)
2. lexical ambiguity and reading skill (gernsbacher)
cross-modal priming (swinney) in skilled reader
Auditory: ….The man was not surprised when he saw several spiders, roaches and other bugs
Visual lexical decision: ant, spy, sew
– Priming of BOTH meanings for up to 1 sec after homograph; after 1.5 secs, priming only for context-relevant
➔Context-irrelevant meaning initially activated, then suppressed
• BUT subsequent data contentious – Activation of context-irrelevant meaning depends on • Homograph dominance • Degree of context bias • Reading skill
