1
The Mnemonic Value of Orthography for Vocabulary Learning
Linnea Ehri
Program in Educational
Psychology
CUNY Graduate Center
2
3
The Mnemonic Value of Orthography for Vocabulary Learning
Collaborator: Julie Rosenthal
Mnemonic value – improving memory
Orthography – spellings of words
Two Ways to read words
DECODING rume rane taik gote yung interpossism subharkible contorrention
MEMORY ocean iron yacht tongue sugar
5
UNFAMILIAR WORDS
By Decoding:
– Graphemes -> Phonemes
– Larger units: spelling patterns for syllables or morphemes
FAMILIAR WORDS
By Memory or Sight
– Note: All words when practiced become read from memory
Process of forming connections
Spelling glue
Pronunciation
Meaning
Knowledge of the grapheme-phoneme system provides the glue connecting spellings to pronunciations in memory
Examples of connections for regularly spelled words
S T O P
/s/-/t/-/a/-/p/
G I GG LE
7
/g/-/I/-/g/-/L/
CH E CK
/c/-/E/-/k/
B IR D
/b/-/r/-/d/
8
Examples of connections for irregularly spelled words
I S* L A N D S W* O R D
/ay/-/L/-/ae/-/n/-/d/
L I S T* E N
/L/-/I/-/s/-/t/-/e/-/n/
/s/ - /o/ - /r/ - /d/
S I G* N
/s/ - /I/ - /n/
9
Knowledge needed to form connections
Phoneme segmentation
– To analyze pronunciations into phonemes
Grapheme-phoneme correspondences
– To access constituents of the writing system ( the glue )
Grapho-phonemic matching
– To connect graphemes to phonemes within specific words
10
Connections for Specific Words are
Learned Quickly
Reitsma (1983)
- Taught 1 st graders to read words
- Minimum of 4 practice trials to read words from memory
Share (2004) – self teaching mechanism
- 1 exposure to words in text for 3 rd graders
- Memory for letters persisted one month
Application to Vocabulary Learning
11
Examined connection forming process as it contributes to vocabulary learning
Explicit Word Learning Task:
– Students rehearsed pronunciations and meanings of new words over several trials
Procedure:
– Initial study trial: words and meanings were introduced
Pictures and defining sentences
– Several test trials with feedback followed
Experimental Manipulation
–
–
Treatment condition : spellings of words were shown during study and feedback periods but NOT when recall of words was tested
Control condition : same except spellings of words were not shown
12
Hypothesis : Students will learn the pronunciations and meanings of words more readily when they are exposed to spellings of the words during study periods than when they are not exposed to spellings.
Explanation : Grapheme-phoneme connections are activated by spellings and will better secure the words in memory.
N = 20 2 nd graders, Mean age = 7yrs. 7 months
Pretests
Woodcock word identification: M = 2.2 grade-equivalent
CVC nonword reading (M=55%) and spelling (M=53%)
Word Learning Conditions
They were taught two sets of 6 concrete nouns and their meanings
–
–
One set: spellings of words accompanied learning
The other set: spellings did not accompany learning
Counterbalancing
Examples:
–
–
–
–
–
–
Gam – family of whales
Cur – a homeless dog
Sod – wet, grassy ground
Fet – big, fun party
Nib – tip of a pen
Yag – fake jewelry
Spelling Seen Condition :
Initial study trial : Student hears each word and a defining sentence, sees picture and written word, repeats word and sentence
(picture) gam
(picture) yag
(picture) sod
(picture) fet
(picture) nib
(picture) yag
An example:
Nib
Spelling Seen Condition :
Word Recall Test Trial : Student sees each picture and recalls word. Then word is seen, pronounced, and heard in a sentence. Students repeats the word and its sentence.
(picture)
(picture)
(picture)
(picture)
(picture)
(picture)
sod
Spelling Seen Condition :
Definition Recall Test Trial : Student hears and sees each word and recalls its meaning. Then meaning is given, and student repeats the word and its meaning.
nib gam cur fet yag
18
Procedures are the same as in the Spelling
Condition
– Except:
Spellings of words are never shown
Students pronounce the words extra times
Summary of word learning events
Each child learns one set of vocabulary words with spellings and another set without spellings
One initial study trial to introduce words
Followed by trials to test recall of words and definitions
Word recall trials are interleaved with definition recall trials
Maximum of 9 trials are provided to learn words and meanings
Note: Spellings of words are not shown when words are tested, so recall depends upon having the spellings of words in memory .
Mean number correct by trial
(Study with 2 nd graders)
3
2
1
0
5
4
6
Recall
Definitions / spell seen
Definitions / spell not seen
Words / spell seen
Words /spell not seen
1 2 3
Trial number
4 5
Recall of words and definitions during the learning trials
Percent of students reaching a criterion of 3 perfect successive trials within a maximum of 9 trials
100%
80%
60%
40%
20%
0%
Words Definitions
Recall Measure
Spell Seen
Spell Not Seen
Mean Percent Correct on Posttests
100%
80%
60%
40%
20%
0%
Words Spellings Definitions
Posttest Measures
Spell Seen
Spell Not Seen
23
Conclusion : 2 nd graders learned vocabulary words and their meanings better when they were exposed to spellings of the words than when they only practiced speaking the words
Explanation:
– Pronunciations were unfamiliar.
–
–
Grapho-phonemic connections better secured their representations in memory
Stronger base for attaching meanings
Second Experiment
N = 32 5th graders , Mean age = 10 yrs. 11 months
Pretests
Reading words and nonwords; spelling words; vocabulary test;
Reader Ability Groups (word reading task)
Higher Readers (7.3 GE) vs. Lower Readers (4.6 GE)
Word Learning Conditions
They were taught two sets of 10 concrete nouns and their meanings
–
–
–
One set: spellings accompanied word learning
One set: spellings did not accompany word learning
Maximum of 8 trials to achieve 3 perfect successive trials
Examples:
– Barrow: a small hill
–
–
–
–
Tandem: a horse-drawn carriage
Fribble: a foolish shallow person
Tamarack: a big tree found all over America
Proboscis: a really big nose
Word-recall training/feedback card in the spellings present vs. spellings absent conditions.
25
Tamarack
26
Spellings will help 5 th graders learn the pronunciations and meanings of new vocabulary words
Students with stronger orthographic knowledge (hi readers) will benefit more from spellings than students with weaker orthographic knowledge (lo readers)
Recall of Words by 5 th Graders (10 max)
Hi Readers,
Spell seen
Lo Readers,
Spell seen
Hi Readers,
Spell not seen
Lo Readers,
Spell not seen
Hi Readers
Lo Readers
Recall of Definitions by 5 th graders (10 max)
Spell seen
Hi Readers
Lo Readers
Spell not seen
Percent of high readers and lo readers reaching a criterion of 3 perfect successive trials within 8 trials
100%
80%
60%
40%
20%
0%
Words
Hi Rs
Words
Lo Rs
Defin
Hi Rs
Defin
Lo Rs
Spell Seen
Spell Not Seen
Mean Percent Correct on Posttests
100%
80%
60%
40%
20%
0%
Hi Rs Lo Rs Hi Rs Lo Rs Hi Rs Lo Rs
Recall Words Write Spellings Fill Cloze Sentences
Spell
No Spell
31
Seeing spellings helped 5 th graders learn the pronunciations and meanings of new vocabulary words more than not seeing spellings.
Students with strong orthographic knowledge benefited more from seeing spellings than students with weak orthographic knowledge
Matthew Effect: rich getting richer over time
Explanation: grapho-phonemic connections better secured pronunciations of words in memory; better specified pronunciations provided a stronger base for learning meanings.
Effect incidental: no attention directed at spellings; no instruction to decode words; automatic activation of mapping relations
32
Implications for Vocabulary Instruction and Learning
Grapho-phonemic instruction :
- It is important for students to acquire strong orthographic knowledge as they learn to read
Strategy instruction :
– When students encounter new vocabulary words, they should be taught to examine the spellings of the words as they pronounce them aloud or as they listen to someone else pronounce them
The End