Intelligent Tutoring Applied to Bioinformatics: A Learning Sciences Perspective

1
Intelligent TutoringApplied to BioinformaticsA
Learning Sciences Perspective??

• Jim Pellegrino Susan Goldman
• University of Illinois at Chicago

2
Overview

• How People Learn - Some of what we know its
  possible implications applications to the
  bioinformatics learning instructional context
• Knowing What Students Know - Understanding the
  nature of assessment its role in the design and
  implementation of an intelligent tutoring project
• Questions and Issues - What do we know need to
  know? Whats sensible and feasible in the context
  of this ITR project?

3
(No Transcript)
4
Advances in Sciences ofThinking Learning

• The most important cognitive principles about
  thinking and learning are derived from study of
  the nature of competence and the development of
  expertise in specific curriculum domains.
• Characteristics of expertise
• Knowledge organization
• Metacognition
• Multiple paths to competence
• Preconceptions and mental models
• Situated knowledge and expertise

5
Characteristics of Expertise

• Experts have well-organized knowledge
• their knowledge is organized to support
  understanding and it is conditionalized for
  use.
• they have fluent access to their knowledge and
  recognize patterns and chunks.
• they have domain-specific problem solving
  strategies
• expertise is acquired over time and depends on
  multiple, contextualized experiences.
• Questions --
• What are examples and features of expertise and
  its consequences in this domain?
• What assumptions can be made about the necessary
  conditions time course of acquiring expertise?

6
Knowledge Organization

• Effective knowledge organization in areas such as
  genetics bioinformatics means that persons
• have a deep foundation of factual and procedural
  knowledge,
• understand facts, ideas and procedures in the
  context of a conceptual framework,
• organize knowledge into schemas that facilitate
  retrieval and application
• Questions --
• What defines the key conceptual, procedural
  knowledge schemas for areas of genetics?
• What must be done to define core competencies?
• What do students need to know to appropriately
  use the biologists workbench?

7
Metacognition

• Competent performers consciously keep track of
  their own thinking and adjust their understanding
  while they learn or solve a problema process
  called metacognition.
• self-aware learners can explain which strategies
  they used and why
• less competent students monitor their thinking
  sporadically and ineffectively.
• Questions --
• How does metacognition develop for specific
  genetics content areas or problem scenarios?
• What does this monitoring look like?
• What is specific to areas and functions of the
  biologists workbench?

8
Multiple Paths to Competence

• Not all persons learn in the same way or follow
  the same paths to competence.
• problem solving strategies become more effective
  over time and with practice
• the growth process is not a simple, uniform
  progression, nor is there movement directly from
  erroneous to optimal solution strategies.
• Questions --
• What does this look like for specific areas of
  genetics?
• What specific patterns exist in the growth of
  understanding and competence?
• About genetics?
• About the Workbench?

9
Preconceptions Mental Models

• Students come to the classroom with knowledge
  representations containing pre-conceptions about
  how the world works.
• If their initial understanding in a domain is not
  engaged they may fail to grasp new concepts
  information that are the focus for learning.
• Questions --
• What are the preconceptions and mental models
  that apply to the domain of genetics?
• Which are serious concerns for future learning?
• How can these be identified and externalized?

10
Situated Knowledge Expertise (1)

• Knowledge frequently develops in a highly
  contextualized and inflexible form, and often
  does not transfer very effectively.
• Transfer depends on the development of an
  explicit understanding of when and how to apply
  what has been learned.
• Questions --
• What constitutes evidence of transfer in areas of
  genetics and bioinformatics?
• How context bound is knowledge of genetics and
  bionformatics and how much does current practice
  constrain transfer?
• To what extent is training for transfer part of
  the teaching learning process?

11
Situated Knowledge Expertise (2)

• There are important relationships among learners
  and the contexts in which they learn which define
  major parts of knowing and expertise.
• Expert performers, through interactions with
  peers, build communities of practice and
  understanding which are distributed and build on
  the learning of others.
• Questions --
• What are the communal and participatory practices
  that constitute part of the domains of
  bioinformatics?
• How is community established and supported in
  areas of bioinformatics?

12
Technology Is A Means To An End Tools to
support the creation and enactment of more
powerful learning environments
13
Principle 1 -- Instruction should be Knowledge
Centered and organized around meaningful problems
with appropriate goals.

• Meaningful problems help overcome the inert
  knowledge problem
• Increases motivation to learn and interest
• Challenge is in creating rich and complex
  problems that support sustained inquiry and the
  development of understanding
• Technology can help bring complex problems into
  the instructional setting and make it manageable
• Learners can exercise control
• What problems are possible and at what price?

14
Principle 2 -- Instruction must be Learner
Centered and provide scaffolds for solving
meaningful problems and supporting learning with
understanding.

• Having complex and interesting problems is not
  enough
• Because of problem complexity and student
  inexperience scaffolds are needed
• Multiple forms of scaffolding -- modeling,
  coaching, guides, and reminding
• Technology can provide resources to assist
  problem solution, tools, hints and ways to
  visualize process and relationships.
• What scaffolds and supports will be needed in
  this learning/instructional context?

15
Principle 3 -- Instruction should be Assessment
Centered and provide opportunities for practice
with feedback, revision and reflection.

• Critical for metacognitive development
• The novices dilemma -- requires scaffolds for
  monitoring and self-regulation skills
• Cycles of feedback, reflection revision provide
  opportunities for practice with feedback which is
  critical for learning
• Technology assists in providing practice
  opportunities allows for embedding diagnostic
  assessment and feedback on conceptual knowledge
  problem solving
• At what level can such assessment be implemented
  given issues of natural language interpretation?

16
Principle 4 -- The social arrangements of
instruction must promote Community Centered
processes such as collaboration and distributed
expertise, as well as independent learning.

• Thinking and understanding is the product of
  multiple minds in interaction
• Working together facilitates problem solving
  capitalizes on distributed expertise
• It helps making thinking visible and provides
  opportunities for feedback revising thinking
• Technology supports collaboration through
  communal databases and discussion tools
• Is there to be a communal aspect of this learning
  and instructional context? How and in what form?

17
(No Transcript)
18
(No Transcript)
19
Assessment as a Process of Reasoning from Evidence

• cognition
• model of how students represent knowledge
  develop competence in the domain
• observations
• tasks or situations that allow one to observe
  students performance
• interpretation
• method for making sense of the data

interpretation
observation
cognition
Must be coordinated!
20
Understanding Cognition

• Challenge I Articulating Multiple Explanations
  of Thought Behavior -- What we want and need to
  understand
• Behavior ranges from micro-processes of rapid
  perception to macro-processes like problem
  solving and negotiation
• Time periods over which behavior and learning
  unfolds can vary tremendously
• Challenge II Multiple Levels of Explanation --
  The way we focus the explanation
• Cognitive Accounts of Individual Processes and
  Knowledge Representations
• Situated/Sociocultural Accounts of Collective
  Processes and Distributed Knowledge
  Representations

21
Rationalist Sociocultural Perspectives

• Rationalist Focused on the nature of competence
  and the development of knowledge in specific
  curriculum domains or topic areas like genetic,
  bioinformatics etc..
• Individual cognition -- the mind of the
  individual
• Sociocultural Focused on the nature of practice
  and forms of participation in communities of
  practice like biology, genetics, bioinformatics,
  etc.
• Distributed cognition -- the collective mind

22
Contrasting Complementary Emphases
23
Why Cognitive Models of Knowledge in Content
Domains are Critical

• Tell us what are the important aspects of
  knowledge that we should be assessing.
• Give deeper meaning and specificity to standards
• Give us strong clues as to how such knowledge can
  be assessed
• Suggest what can and should be assessed at points
  proximal or distal to instruction
• Can lead to assessments that yield more
  instructionally useful information
• Chosen well assessments tied to domain-based
  models can guide a student-tutor interaction and
  support the learning process

24
Translating Science into Engineering Principles
of Assessment Design

• Assessment design should be based upon a model of
  student learning and a clear sense of the
  inferences about student competence that are
  desired for the particular context of use.
• Design is recursive process -- starting with the
  Student Model
• The student model suggests the most important
  aspects of student achievement that one would
  want to make inferences about and provides clues
  about the types of tasks that will elicit
  evidence to support those inferences.

25
Some Questions Issues

• Who are the learners and what do they already
  know?
• About genetics?
• About the Workbench?
• What are the types of scenarios they are asked to
  work with?
• How large and/or constrained is the problem
  and/or solution space?
• What do we know about how they approach such
  problems and their degree of success?
• What do they learn?
• How is it assessed now?
• How much do we care about transfer? Of what type?