Title: An Introduction to Strategic Reading Comprehension
1An Introduction to Strategic Reading Comprehension
2Overview
- Shallow versus deep comprehension
- Levels of representation and processing
- Theoretical frameworks
- Strategies
- Interdisciplinary projects with computer
technologies
3Wading in Shallow Waters
- Shallow knowledge
- Shallow comprehension strategies
- Minimal comprehension training
- Shallow tests
- Shallow classroom questions
- Unspectacular comprehension calibration of
learners - Shallow theories
- Shallow policy makers
4Restoration of Hope
- The escape of the Brazilian boa constrictor
earned Harry his longest-ever punishment. By the
time he was allowed out of his cupboard again,
the summer holidays had started and Dudley had
already broken his new video camera, crashed his
remote control airplane, and, first time out on
his racing bike, knocked down old Mrs. Figg as
she crossed Privet Drive on her crutches.
5- A mushy, brown peach is lifted from the garbage
and placed on the table to pinken. It pinkens,
it turns hard, it is carried in a shopping sack
to the grocers, put on a shelf, removed and
crated, returned to the tree with pink blossoms. - In this world, time flows backward.
6Levels of Discourse Structure
- Graphemes and phonemes
- Morphemes and words
- Syntactic composition
- Linguistic style and dialect
- Explicit propositions
- Referents of referring expressions
- Common ground (shared knowledge)
- Discourse focus versus presuppositions
- Situation models (space, causality, goal
structures) inferences - Embedded dialog
- Configuration of multiple agents
- Genre, registers, rhetorical structures
- Plot configurations
- Local and global coherence
- Point of message -- theme
- Goals and attitude of author
7What is the ideal grain size?
- Crude but teachable to teachers
- Word/sentence interpretation
- Links between sentences
- Links to readers background knowledge
- Theoretical natural categories (Kintsch,
Graesser) - Surface code
- Propositional textbase
- Situation model
- Text genre and rhetorical structure
- Pragmatic communication
- Psychometric categories (Oakhill, VanderVeen)
- 3-5 construct limitation
- Evidence centered design (Mislevy)
- ETS, College Board
- Dozens at a fine grain
- Requires computers
- Challenges of validation
8Example Theories and Models in Discourse
Psychology
- Text grammars (Van Dijk)
- Bridging and given-new (Clark)
- Scripts, plans, and schemas (Bower, Schank
Abelson) - Structural affect theory (Brewer, Lehnert)
- Structure-building framework (Gernsbacher)
- The Reader model (Just Carpenter)
- Constructionist theory (Graesser, Singer,
Trabasso) - Construction-integration model (Kintsch)
- Landscape model (van den Broek)
- Event indexing model (Zwaan)
- Memory-based resonance model (OBrien and Myers)
- Embodied theoretical frameworks Indexical
Hypothesis (Glenberg)
9Five Metaphors for Text Understanding (Graesser
Britton, 1996)
- Assembling a multilevel representation
- The construction of a coherent structure
- A dynamical system
- Managing working memory
- Inference generation
10Theoretical Components
Levels of representation
Reader goals and assigned task
Knowledge and skill
Metacognition
Memory and attention
Cohesion and coherence
Information sources
11What do the theories and models truly predict and
explain?
- Directly articulated in the model
- Naturally follows from the model, but not
directly articulated - Ad hoc assumptions or parameters to accommodate
data - Impossible to accommodate or out of scope of
model - Do we draw the line at 2 or 3?
12(No Transcript)
13Observations on analysis of theoretical
predictions
- Very few 2s even though they are perhaps most
interesting - Status of predictions is often murky with respect
to 3 versus 4 - General theories of memory and cognition do not
go the distance in explanatory adequacy. We need
theories of discourse and pedagogy. - Problems with kitchen sink manipulations in
informing us about effective theoretical
components
14Strategies and Strategy Interventions
- iSTART Interactive Strategy Training for Active
Reading and Thinking (McNamara 30) - LSAT RSAT diagnoser with iSTART (Magliano
Millis) - Reciprocal teaching method (Palincsar Brown)
- Concept mapping (Vitale Romance)
- PALS Peer-assisted Learning Strategies (Fuchs)
- CORI Concept Oriented Reading Instruction
(Guthrie) - Text Structure (Williams)
- Structure Strategy tutor (Meyer)
- Question Asking and Answering (King, Guthrie,
Gholson) - 3D Readers (Johnson-Glenberg)
- Joke City (Yuill)
- Indexing and embodiment (Glenberg)
- Universal Learning Environments (Dalton
Proctor) - Summary Street (Caccamise, Kintschs)
15Good Job!
16- Question Taxonomy
- (Graesser and Person,1994 Gholson Craig)
- LEVEL 1 SIMPLE or SHALLOW
- 1. Verification Is X true or false? Did an
event occur? - 2. Disjunctive Is X, Y, or Z the case?
- 3. Concept completion Who? What? When?
Where? - 4. Example What is an example or instance of a
category?). - LEVEL 2 INTERMEDIATE
- 5. Feature specification What qualitative
properties does entity X have? - 6. Quantification What is the
value of a quantitative variable? How much? - 6. Definition questions What does X mean?
- 8. Comparison How is X similar to Y? How is X
different from Y? - LEVEL 3 COMPLEX or DEEP
- 9. Interpretation What
concept/claim can be inferred from a pattern of
data? - 10. Causal antecedent Why did an event occur?
- 11. Causal consequence What are the consequences
of an event or state?
17Status of Strategy
- Strategy presses buttons in some circles
- Strategies should be integrated with concrete
examples and content - Strategies should not interfere with
comprehension - Strategies are always context-sensitive rather
than context-free
18Strategy parameters to worry about
- What level of representation is being tapped?
- What prerequisite knowledge or skills are needed
to apply the strategy? - What prerequisite knowledge or skills will yield
maximal gains through the strategy? - How much training is needed for mastery?
- Does the strategy need to be explicit and
conscious? Or is unconscious induction adequate? - What are the relevant genres, registers, and
content? - Can humans be trained or is a computer necessary?
19Claims
- Human teachers and trainers will not have enough
expertise and time to apply complex
context-sensitive strategies. - Computer technologies are in place for automation
at many levels.
20Interdisciplinary Approach
Dissecting text at multiple levels
Readers expertise at each level
Computational Linguistics
Verbal protocols
Corpus analyses
Psychological educational theories
Behavioral Tests of Predictions
21Behavioral Tests
- OFF-LINE
- Recall
- Summarization
- Retrospective think aloud
- Recognition tests
- Sentence verification
- Question asking answering
- Response signal paradigm
- Ratings
- Word sorting
- ON-LINE
- Self-paced reading times
- Eye tracking
- Word naming latencies
- Lexical decision latencies
- RSVP-SOA
- Physiological recordings
- fMRI
- Evoked potential
22COMPUTATIONAL LINGUISTICS AND CORPUS ANALYSES
- Lexicons (Wordnet, COMLEX)
- Brill part-of-speech tagger
- Syntactic parsers
- Penn Tree Bank
- Latent Semantic Analysis
- Biber discourse classifications
- MUC and TREC
23CohMetrix (McNamara, Louwerse, Cai, Graesser)
24QUAID (Graesser, Louwerse, Cai)
25Plate-Tech Tutor on Critical Stance
(Wiley, Goldman, Graesser)
26AutoTutor (Graesser, Hu, Louwerse, Gholson,
Person 100)
- Asks questions and presents problems
- Evaluates meaning of the learners answers
- Gives feedback on answers
- Display facial expressions with emotions
- Nods and gestures
- Hints
- Prompts for specific information
- Adds information that is missed
- Corrects bugs and misconceptions
- Answers student questions
- Holds mixed-initiative dialog in natural language
- Guides the learner in interactive 3-D simulations
27(No Transcript)
28When a car without headrests on the seats is
struck from behind, the passengers often suffer
neck injuries. Why do passengers get neck
injuries in this situation?
Question
Head
Parameter Controls
Describe what happens
29Conceptual Physics(Graesser, Jackson, et al.,
2003)
- Three tutoring conditions
- AutoTutor
- Read textbook control
- Read nothing
- 63 college students
30Multiple Choice Scores
31Learner emotions AutoTutor (Graesser,
Franklin, Gholson, Louwerse,
Picards Affect Computer Lab
at MIT)
-
-
-
- Frustration
- Confusion
- Excitement
- Boredom
32Visual IBM Blue eyes camera
Posture Body Pressure Measurement System
Auto Tutor
Pressure force sensitive mouse and keyboard
AutoTutor text dialog
33Fishics AutoTutor
34Closing Comments
- Rand Reading Study Group (Snow, 2002) emphasized
comprehension - Text
- Reader
- Activities
- Sociocultural context
- Advanced computer-based learning technologies
will be necessary to achieve this vision
35