Psyc 124: Psychology Of Reading

Front 1

Frequency result

Back 1

NL high < NL low

Front 2

Regularity result

Back 2

NL regular < NL irregular

Front 3

Frequency x Regularity Interaction Effect

Back 3

NL high & reg = NL low & reg = NL high & irreg
and
NL low & irreg > NL all others

Front 4

Pseudowords Result

Back 4

NL pseudo > NL all others

Front 5

3 Reasons Pseudowords > All Others

Back 5

1. No target word, takes longer to reach activation
2. No partial activation of neighbors with similar spelling
3. Relies solely on rule route

Front 6

Fast Incorrect Regularizations Result (Irregular words)

Back 6

NL regular incorrect < NL irregular correct

(Cannot be accounted for if there is not strong lateral inhibition in Analogy model)

Front 7

Position x Regularity Interaction Effect

Back 7

NL irregular @ phoneme 3 < NL irregular @ phoneme 2

(Irregular phonemes closest to front of word take longest-- no difference for regular words)

Front 8

Analogy Model: Word Activation - Frequency Result

Back 8

NL hi < NL lo (Regular and irregular)

Front 9

Analogy Model: Word Activation
(HF/LF regular/irregular words)

Back 9

Regular:
HF = no activation of neighbors, inhibits almost completely
LF = slight activation of neighbors because target cannot inhibit neighbors as much
Irregular:
HF = Same
LF = Same

Front 10

Analogy Model: Phoneme 2 Activation
(HF/LF regular words)

Back 10

HF = mirrors word activation because vowel pronunciation is regular
LF = partially activated neighbors support same vowel pronunciation, mirrors word activation BUT rises even faster than before (still not faster than HF)

Front 11

Analogy Model: Phoneme 2 Activation
(HF/LF irregular words)

Back 11

HF:
1. target activates correct irregular pronunciation
2. neighbors (incorrect + regular) are suppressed
3. target word wins
LF:
1. target activates correct irregular pronunciation
2. neighbors activate incorrect regular pronunciation
3. lots of neighbors, regular pronunciation rises faster
4. eventually target word wins

Front 12

Two DRC Model Routes

Back 12

1. Lexical route (1.2.3.4, wait for slowest phoneme to finish before beginning)
2. Rule-based route (1>2>3>4, begin and then figure out slowest phoneme)

Front 13

DRC Model: Regular Activation
1. How final decision is made
2. Information from 2 routes
3. Where phoneme activation line is

Back 13

1. Final decision = lexical + rule
2. Information from 2 routes is consistent, speeds up activation
3. Phoneme activation is IN FRONT OF pure lexical activation line, wait for slowest phoneme

Front 14

DRC Model: Irregular Activation
1. How final decision is made
2. Information from 2 routes
3. Where phoneme activation line is

Back 14

1. Final decision = lexical – rule
2. Information from 2 routes is inconsistent, slows down activation
3. Phoneme activation is BEHIND the pure lexical activation line, wait for slowest phoneme-- usually irregular in 2nd position

Front 15

WW Criterion:
1. Minimum info needed
2. Regularity effect
3. IPD effect

Back 15

1. Minimum info is whole word, vowels are slowest, must wait for slowest one
2. NL reg < NL irreg
3. IPD reg < IPD irreg (no IPD effect because pronunciation for the whole world is known, so you don't need to elongate the IP)

Front 16

IP Criterion:
1. Minimum info needed
2. Regularity effect
3. IPD effect

Back 16

1. Minimum info is pron for IP, elongate IPD until 2nd phoneme retrieved, wait for 1st consonant
2. NL reg = NL irreg (no reg effect: all consonants reach threshold at same time)
3. IPD reg < IPD irreg (pronunciation for vowel unknown, so you elongate the IP to wait for its activation)

Front 17

Dependent Variables

Back 17

Naming Latency
IPD (Initial Phoneme Duration)

Front 18

Independent Variables

Back 18

Correct/Incorrect pronunciation
Regular/Irregular
Hi frequency/Lo frequency
Pseudowords/Non-pseudowords
Plosive/Non-plosive

Front 19

Prerequisites for Reading (5)

Back 19

1. Letter recognition
2. Directional orientation
3. Eye movement control
4. Word consciousness
5. Linguistic/phonological awareness

Front 20

Ways of Teaching Reading (3)

Back 20

1. Whole-word (flashcards, context of sentence)
2. Phonics (rule-based)
3. Meaning (in-between the two, using a bunch of words that sound similar in a sentence)

Front 21

Stages of Reading Development (4)

Back 21

1. Linguistic guessing (Based on pictures/content)
2. Discrimination not guessing (guess based on 1st letter)
3. Sequential decoding (decoding using combos of letters)
4. Hierarchical decoding (context sensitive rules of pronunciation-- a letter in context of other letters like c pronounced /s/ before an i/e, or c pronounced /k/ before a/o/u)

Front 22

Dyslexia (2 kinds)

Back 22

1. Developmental
2. Acquired

Front 23

Developmental Dyslexia

Back 23

A disorder manifested by difficulty in learning to read despite conventional instruction, adequate intelligence, and socio-cultural opportunity

Front 24

Criteria for Developmental Dyslexia Diagnosis (3)

Back 24

1. Reading 2+ grade levels below grade level
2. 90 < IQ (not due to low intelligence)
3. Not due to emotional/motivational problems

Front 25

Types of Acquired Dyslexia (2)

Back 25

1. Central
2. Peripheral

Front 26

Central Acquired Dyslexias (4)

Back 26

1. Non-semantic
2. Surface
3. Phonological
4. Deep

Front 27

Non-semantic Dyslexia
1. What they can/can't do
2. Deficit in what route

Back 27

1. Can't identify spelling of some animals with picture; can read words, pseudo-words and irregular words
2. Deficit in semantic route

Front 28

Surface Dyslexia
1. What they can/can't do
2. Deficit in what route

Back 28

1. Homophone errors, regularization of irregular words
2. Deficit in lexical route

Front 29

Phonological Dyslexia
1. What they can/can't do
2. Deficit in what route

Back 29

1. Can't produce pseudo-words (sometimes said as real words), function words
2. Deficit in sub-lexical/rule route

Front 30

Deep Dyslexia
1. What they can/can't do
2. Deficit in what route

Back 30

1. Semantic errors, visual errors, cannot produce pseudo-words/function words (over-reliance on semantic route)
2. Deficit in lexical and rule route

Front 31

Biological Bases for Developmental Dyslexia (3)

Back 31

- Family History
- Twin Studies (more prevalent in identical)
- Planum Temporale (no asymmetry in Wernicke's area)

Front 32

Advantages of Reading on Computers (9)

Back 32

1. Saves trees/money/space
2. Harder to lose
3. Easy to manipulate text/quote
4. Links to relevant material
5. Easy to search for key words/phrases
6. Zoom options
7. Text to speech capability
8. Translations
9. Access to multimedia

Front 33

Disadvantages of Reading on Computers (6)

Back 33

1. Rotated pages can make it harder to read
2. Some annotations (esp. drawings)
3. Scrolling (difficult to keep place)
4. Eye strain
5. Not everyone has access
6. Requires power source

Front 34

Ways to Overcome Problems Reading on Computers (4)

Back 34

1. Minimize glare
2. Increase contrast
3. Increase temporal resolution (refresh)
4. Increase spatial resolution (pixels/inch)

Front 35

Text to Speech Benefits (5)

Back 35

1. Get information when you cannot read it
2. Speak & spell toys for children
3. Online information (both static and real-time)
4. Foreign languages
5. Disabilities (deaf)

Front 36

Parts of a Synthesizer (2)

Back 36

1. Natural Language Processing (breaking down language)
2. Digital Signal Processing (generating the sound)

Front 37

Parts of Natural Language Processing (3)

Back 37

1. Text analyzer
2. Generating phonetic sequence: letter → sound mapping
3. Prosody generation: changes in pitch loudness and duration

Front 38

Text Analyzer Parts (4)

Back 38

1. Pre-processor: segment speech into list of words
2. Morphological analyzer: decompose words into morphemes
3. Contextual analyzer: determine POS of words, adjust pronunciation
4. Syntactic prosodic parser: find clause/phrasal units, interject pauses

Front 39

Digital Signal Processing Parts (2)

Back 39

1. Rule based (Klatt): formant synthesizers, difficult to achieve natural sound
2. Concatenative synthesizers: sounds more natural, takes diphones from real speech and generate a full syllable by concatenating (piecing together) two diphones with same overlapping vowel (“ba” + “ag” → bag -- vowel is same for 2 diphones)

Front 40

Normal Reading Rate

Back 40

200-300 words/minute

Front 41

Claimed Speed Reading Rate

Back 41

~2000 words/minute

Front 42

Why Speed Reading is Impossible

Back 42

People can only distinguish within 1-2 degrees of fixation point (speed reading people make educated guesses about content-- cannot comprehend)

Front 43

WW Results for Irregular Phoneme @ Position 3

Back 43

All predictions are the same as position 2
Regularity Effect: NL reg < NL irreg
No IPD Effect: IPD reg = IPD irreg
 * You are waiting for the whole word anyways – it doesn’t matter if the irregularity is as position #2, 3, or 4

Front 44

IP Results for Irregular Phoneme @ Position 3

Back 44

Same: NL reg = NL irreg
Change: IPD reg = IPD irreg
* The long wait for the vowel happens AFTER the the 1st and 2nd phoneme (after the IPD)