MURI:

1
MURI Training Knowledge and Skills for the
Networked Battlefield ARO Award No.
W9112NF-05-1-0153 Alice Healy and Lyle Bourne,
Principal Investigators Benjamin Clegg, Bengt
Fornberg, Cleotilde Gonzalez, Eric Heggestad,
Ronald Laughery, Robert Proctor,
Co-Investigators
2

• Project goals
• As defined in the MURI proposal executive summary
• Construct a theoretical and empirical framework
  for training that can
• Predict the effectiveness of different training
  methods for military tasks
• Point to ways to optimize training for specific
  tasks.

3

• Technical approach
• Work will proceed in three interrelated efforts
• Experimentation and training principle
  development
• Breadth, interaction, and generality of training
  principles
• Acquisition and retention of basic skill
  components
• Effects of levels of automation, individual
  differences, and team environments on performance
• Multi-dimensional taxonomic analysis
• Analyze task types, training methods, performance
  measures
• Using training principles, relate task training
  to performance
• Computational modeling
• Derive from data and training principles
• Monitor and assess for performance and
  reliability
• Use to extend the research findings to militarily
  relevant tasks

4

• MURI personnel and their project roles
• Overview and Coordination CU, Alice Healy,
  Principal investigator
• Lyle Bourne, Co-principal investigator
• Experimentation/Principles
• Training Principles in CU, Healy and Bourne
• simple and complex tasks
• Basic Components of Skill Purdue, Robert Proctor,
  Co-investigator
• Levels of automation, CSU, Benjamin Clegg,
  Co-investigator
• individual differences, Eric Heggestad,
  Co-investigator
• and team environments
• Taxonomies CU, William Raymond, Research
  Associate
• Modeling
• ACT-R CMU, Cleotilde Gonzalez, Co-investigator
• IMPRINT CU, Ron Laughery, Co-investigator
• Model Assessment CU, Bengt Fornberg,
  Co-investigator

5

• Experimentation and
• training principle development
• Breadth, interaction, and generality of training
  principles
• Generality of principles across tasks Strategic
  use of knowledge
• Multiple principles in a task serial position,
  list length, and chunking effects
• Principles in complex, dynamic environments
  Training difficulty hypothesis
• Acquisition and retention of basic skill
  components
• Stimulus-response compatibility in response
  selection
• Transfer of response associations to different
  tasks or environments
• Effects of levels of automation, individual
  differences, and team environments on performance
• Impact of levels of automation on performance
• Interaction of automation level with individual
  ability (task and transfer)

6

• Taxonomic analysis
• Task types
• Feature decomposition for cognitive tasks,
  aligned with IMPRINT taxons
• Model of data entry
• Training methods
• Developing taxonomy from experiments, needed
  military training coverage
• Performance measures
• - Developing taxonomy from prior research
• Training principles
• 30 training principles summarized
• Aligning training principles with other dimensions

7

• Computational modeling
• Cognitive modeling using ATC-R
• Cognitive model of data entry task
• Model interpretations for observed phenomena
• Model predictions for optimizing skill retention
• IMPRINT modeling
• Researchers being trained in modeling with
  IMPRINT
• Model assessment
• IMPRINT and ACT-R suited for intended modeling
  purposes
• Data mining (e.g., using radial basis
  functions) may also be useful