Title: Learning Across the Expert-Novice Continuum: Cognition in the Geosciences
1Learning Across the Expert-Novice Continuum
Cognition in the Geosciences
- Heather Petcovic
- Western Michigan University
2Julie Libarkin Zach Hambrick Sheldon
Turner Nicole LaDue Tara Rench
- Heather Petcovic
- Kathleen Baker
- Caitlin Callahan
- Magdalena Wisniewska
Joe Elkins
Thank You IUGFS
This work is supported by the National Science
Foundation under Grants No. DRL-0815764 (PI
Petcovic) and DRL-0815930 (PI Libarkin). Any
opinions, findings, and conclusions or
recommendations expressed in this material are
those of the author(s) and do not necessarily
reflect the views of the National Science
Foundation.
3WHY STUDY EXPERT AND NOVICE GEOCOGNITION?
4THE EMPIRICAL STUDY
- PURPOSE examine how cognitive processes that
underlie geological thinking and skills change
from novice (undergraduate student) to expert
(professional geoscientist)
Education Experience
Domain Content Knowledge
Working Memory Capacity
General Spatial Ability
Geological Expertise
Outdoor (Field) Problem-Solving
Outdoor Comfort (Novelty Space)
5THE EMPIRICAL STUDY
- METHODS (1) use existing or develop new
cognitive tasks and measures (N40) (2) recruit
novice through expert volunteers to complete
suite of lab and field tasks (N29)
Construct Task or Measure
Experience Geologic Experience Questionnaire (General and Mapping)
Domain Content Knowledge Geoscience Concept Inventory (GCI)
General Spatial Ability Paper Folding, Form Board (ETS Toolkit) Space Relations
Working Memory Capacity General Matrix Span, Arrow Span
Working Memory Capacity Geologic Block Diagram Test
Outdoor Comfort Novelty Space Survey
Field Problem-Solving Geologic Mapping Task Bedrock geologic map (scanned and digitized)
Field Problem-Solving Geologic Mapping Task GPS track during mapping (ArcGIS)
Field Problem-Solving Geologic Mapping Task Post-mapping Interview (and audio logs, N8)
6METHODS LAB AND FIELD
- LABORATORY STUDY
- Adopt, adapt, or develop new tasks and measures
- Recruit subjects (N40)
- Novice no geology experience
- Intermediate undergrad major or grad student
- Expert professional geoscientist
- Quantitative analysis
- FIELD STUDY
- Develop task/measure suite
- Recruit and select subjects (N29)
- Novice undergrad major
- Intermediate grad or young professional
- Expert professional
- Data collection Aug 2009
- Rocky Mts., Montana
- Cognitive tests
- Intro to local rocks
- Mapping task
- Data analysis
7MEASURES EXPERIENCE
- GEQ-G
- Geoscience education and experience survey
- Demographics
- Degrees and coursework
- Assistantships
- Teaching experience
- Original research
- Work experience
- Subject validation
- Scoring after AIPG certification
- GEQ-M
- Experience bedrock mapping
- Coursework, internship, teaching, work
experience, etc. - Related to bedrock mapping
Have you completed an undergraduate degree in
geology or earth science? If no, are you
currently enrolled as an undergraduate student?
Did you take a Field Geology methods
course? Do you now hold or have you ever had a
position to teach geological sciences at the
college level? Have you ever been employed as a
geoscientist or earth scientist (not including
internships and/or assistantships while you were
a student)?
8RESULTS EXPERIENCE
- FINDINGS
- GEQ-G
- 10 experts (gt5)
- 9 intermediates (2.5-5)
- 10 novices (lt2.5)
- GEQ-M
- 4 experts (gt7)
- 13 intermediates (1-5)
- 12 novices (lt1)
- Measures are significantly correlated
9MEASURES DOMAIN CONTENT KNOWLEDGE
- GCI
- Knowledge of geoscience content
- Modified from Libarkin and Andersons GCI
- 18 multiple choice
- 2 free-response
Draw a representation of the intersection between
a bedding plane and a fault on the stereonet
below.
- Why does the Earth have a magnetic field?
- The Earth contains crust of different composition
- The Earth has a gravitational force of attraction
- The Earth contains moving metal liquid
- The Earth has an orbit around the sun
- The Earth does not have a magnetic field
10RESULTS DOMAIN CONTENT KNOWLEDGE
- FINDINGS
- GCI discriminates experts from novices
- GCI and GEQ-G positive correlation
- GCI and GEQ-M significant positive correlation
- Experts have higher domain content knowledge
N29
s. rho .376, p .04
11MEASURES GENERAL SPATIAL ABILITY
- SPATIAL-VISUAL ABILITY
- Mental rotation, recognizing patterns and shapes,
vertical and horizontal frames of reference,
mentally manipulating a surface or volume,
spatial relationships - Paper Folding, Form Board (ETS Toolkit) Space
Relations
12RESULTS GENERAL (NON-GEOLOGIC) SPATIAL ABILITY
- FINDINGS
- Higher scores than non-geologists
- Experts do NOT score higher than novices
- Does not decrease with age
- Positive corr. with GCI
- Do high spatial ability students self-select into
geosciences?
13MEASURES GENERAL SPATIAL WORKING MEMORY CAPACITY
- WORKING MEMORY CAPACITY
- Capacity to both store and process information in
active memory - Correlates with other measuresof intelligence
-
- Arrow Span, Matrix Span
- Sets of 2-6
- Briefly view letter
- Determine if inverted
- Briefly view arrow
- Was letter inverted? Y/N
- Draw arrow
14RESULTS GENERAL SPATIAL WORKING MEMORY CAPACITY
- FINDINGS
- No better than non-geologist population
- Experts do NOT perform better than novices
- Negative correlation with age
- Weak positive correlation with GCI
- Experts do not have greater general WMC
15MEASURES GEOLOGIC WORKING MEMORY CAPACITY
- DOMAIN SPECIFIC WMC
- Block diagram test ubiquitous in teaching
- Geologic and non-geologic
- Range of difficulty
- Test
- View block
- Reproduce on tablet PC
16RESULTS GEOLOGIC WORKING MEMORY CAPACITY
- FINDINGS
- People overall do better on geologic blocks
- Geologists do NOT do better on geologic blocks
- Experts are faster, can be more sloppy, give up
quickly, and tend to correct non-geologic blocks
NOVICE (Non-Geo)
INTERMEDIATE
EXPERT
17MEASURES OUTDOOR COMFORT
- NOVELTY SPACE SURVEY
- Modified from Elkins
- Measures 3 of the 4 dimensions of novelty space
- Cognitive (knowledge and skills relevant to task)
- Geographic (familiarity with the outdoor setting)
- Psychological (familiarity with the task
expectations) - 5 point Likert scale
- I know a lot about geology.
- I have difficulty identifying geologic
structures in the field. - I know where we are going on this field project.
- I am unfamiliar with this field area.
- I know what is expected of me during this field
project. - I am uncomfortable working alone in the field.
18RESULTS OUTDOOR COMFORT
- FINDINGS
- Cognitive novelty decreases with expertise
- Geographic and social novelty decrease slightly
with expertise - Experts are more comfortable with mapping task
19MEASURES GEOLOGIC PROBLEM-SOLVING
- THE BEDROCK MAPPING TASK
- Walk-through introduction to rock types
- Air photo and topo maps
- 1 km by 500 m area
- Unlimited time
- 4 major rock units
IMAGE REMOVED DUE TO ONGOING RESEARCH
20MEASURES BEDROCK MAPS
s. rho -.47, p .01
- MAP SCORING RANK ORDER BASED ON
- Correct ID of units
- Accuracy of contacts
- Accuracy of thickness
- Correct structural interpretation
IMAGE REMOVED DUE TO ONGOING RESEARCH
21RESULTS BEDROCK MAPS
- CORRELATIONS
- Map rank and GCI significant correlation
- Map rank and WMC (general and geologic)
correlation - Map rank has no correlation with spatial
visualization - Map performance mainly driven by domain content
knowledge
22MEASURES GPS TRACKS
- PARAMETERS
- Total time, distance, no. stops
- Mean speed, intersections
- Up/down ratio, area seen
- ANALYSIS
- Cluster analysis based on track characteristics
to identifysimilar track types - Visual inspection and qualitative analysis to
identifysignature clusters - Overlay tracks on maps
23RESULTS GPS TRACKS
3
2
4
1
24RESULTS GPS TRACKS
- SIX SIGNATURE CLUSTERS
- Reflect different mapping strategies
Looping coverage
Back forth
Wandering path
Linear
Back forth w /loops
Linear w/ loops
25RESULTS GPS TRACKS
Mapping Expertise
Expert
Intermediate
4
Novice
Wandering Path
2
Participants
Looping Coverage
Back Forth
Linear
Back Forth w Loops
27
5
Linear w Loops
26
28
25
29
3
1
26RESULTS GPS TRACKS AND MAPS
- ANALYSIS (N9)
- Time spent within units
- Time spent along contacts between units
- Accuracy of map units and contacts (overlaid on
key)
IMAGE REMOVED DUE TO ONGOING RESEARCH
27RESULTS GPS TRACKS AND MAPS
- FINDINGS
- Mapping rank is moderately correlated with of
intersections significant correlation with
accuracy (Unit 1 and Unit 3) - Mapping expertise moderately correlated with area
seen and of intersections some moderate
correlation with accuracy and time spent
28RESULTS MAPPING TASK
- FINDINGS
- Maps and tracks reflect individual strategies
- Experts tend to make better maps but so do newly
trained novices - Experts tend to be more efficient with their
route, and more effective covering more area in
shorter time - Experts tend to be more accurate and consistent
with placement might be due to ability to
orient themselves in the field - Not one mapping strategy produces a good map,
however some are better than others - Good back forth, linear
- Bad back forth w/ loops
- Novices more likely to adopt poor strategy
29OVERALL PRELIMINARY FINDINGS
- EXPERTS
- Have more domain-specific content knowledge
- Have reduced cognitive novelty
- Were not significantly better spatial-visualizers
than undergrads - Do not seem to have greater geologic or general
WMC? - Produce better maps (though newly trained novices
also produce good maps)! - Overall, individual differences in strategies
used to produce successful maps
30IMPLICATIONS FOR TEACHING
- Domain content knowledge is critical
- Early instruction in spatial ability
- Purposeful movement in the field physical
instruction? - No one best strategy for mapping success (but
avoid bad strategies)