New Window

This printer friendly version collects all 
the ICARE segments for this module on one page.
  
Clicking links may lead to confusion. 


Module 6
Cognitive Task Analysis

Introduction

Using Knowledge Maps to Make Instructional Design Decisions

This module gives you the opportunity to learn how to build knowledge maps. This module also provides you with a chance to practice using knowledge maps to make instructional design decisions. 

When you complete this module you should be able to:

  1. Interpret knowledge maps and see how they relate to lesson content.
  2. Use probe questions to construct knowledge maps.
  3. Use knowledge maps to guide lesson/curriculum development.

The Content-Performance Matrix

Ruth Clark describes two ingredients of instruction as information and the performance outcome. Her Content-Performance matrix, shown below, provides a framework for classifying facts, concepts, processes, procedures, and principles as performance outcomes, at the remember or application level. This matrix provides a way of categorizing knowledge and outcomes in ways to better organize instruction. Knowledge maps are one of the ways to display and analyze this information.
 

  Facts Concepts Processes Procedures Principles
Find   Define a class, or set of objects
or events
Develop a process Derive, create, a process,
procedure, or technique for achieving a goal
Discover
cause and effect relationships
Use/Apply   Classify new examples Solve a Problem, or
make an Inference
Perform
the Procedure
Solve a Problem. Make an Inference
Remember Remember the Facts Remember
the Definition
Remember
the Stages
Remember
the Steps
Remember
the Guidelines


The Spreadsheet: An Example of a Knowledge Map

The SITE model challenges designers to think about the opportunities that the technical context provides for learners to realize their goals. Knowledge maps provide a context for more detailed analysis of the knowledge learners need to develop in order to achieve their goals, or the goals of the enterprise or community. 

The map below explicitly identifies one goal as a final state of affairs desired by the learner: "mortgage cost calculated" (red oval). The map also lays out more detailed knowledge of actions and concepts  necessary to realize this state of affairs. The actions are represented by green trapezoids and the concepts are represented by spiky blue bubbles. These concepts and actions can be stated as learning outcomes that would enable calculation of the mortgage cost.  For example: Given a spreadsheet document the learner will be able to identify the rows, columns, and cells. Before proceeding, look at the diagram below and consider which areas represent procedural knowledge, and which areas represent conceptual knowledge. Mouse over the knowledge map to see additional details.

 

Study the map and write three learning outcomes, or goals and objectives, that describe knowledge required to achieve the goal. At least one of your learning outcomes should describe an action or skill, and at least one should provide a basis for measuring conceptual knowledge.  Highlight below feedback with your cursor to see our answer.
 

Feedback: To see our answers, triple click the white area below:

Here are some examples of instructional goals that might be appropriate for this domain.  Of course, the goals could vary quite a bit depending on information about the prior knowledge of the learners, their role in some enterprise, and so on.
1. Students will be able to enter formulas into a cell.
2. Students will be able to identify (point to) the major parts of a spreadsheet: rows, columns, and cells.
3. Students will be able to name the two major types of data that can be entered into a spreadsheet cell.

Cognitive Task Analysis and Knowledge Maps

Traditional task analysis elicits knowledge and skills needed for each individual subtask from subject matter experts. Cognitive task analysis delves deeper and illuminates a larger knowledge base of task-related knowledge. Cognitive task analysis considers the procedures and concepts. In addition, a cognitive task analysis clarifies the interrelationships among concepts and procedures in a specific knowledge domain.

Designers use a variety of cognitive task analysis methods. Some methods focus on data collection, others on data representation, and others on specific task analysis (for more info. see Sallie Gordon's Book Systematic Training Program Design: Maximizing Effectiveness and Minimizing Liability (1994, pp. 70-110). The knowledge map combines data collection and data representation. Sallie Gordon advocates conceptual graph structures (a type of knowledge map) for cognitive task analysis (Gordon, p.99).

Data collection for developing knowledge maps involves obtaining  knowledge from documents, verbal protocols, question probes, and observation of task performance. During the data collection process, the designer classifies the knowledge and uses specified symbols to represent the knowledge on the knowledge map. The conceptual graph process of collecting and representing knowledge capably captures all types of knowledge, from factual, rule, and explicit knowledge, to implicit, automated knowledge.

A knowledge map represents relationships between knowledge elements (goals, concepts, goal/actions) in a domain of learning. The learner can work their way through the domain of knowledge in a variety of ways, depending on their specific purpose, or their enterprise's purpose. Before proceeding, look at the diagram below and consider how you could move through the knowledge domain.  Mouse over the knowledge map to see additional details.

This thumbnail sketch of a knowledge map lays out three different paths the learner might take through the domain. The paths might vary depending on their prior knowledge of the domain.  For example, a learner who can capably use the manual or the help menu might opt for the blue path. A learner who was already familiar with basic concepts might take the red path.

A Little Bit of History

The type of knowledge maps that we are using in this module are called conceptual graph structures (CGS), and they were originally developed for use in a research context. People have used them in the field of computer science for many years. Since then, conceptual graph structures have been modified for use in many domains including: knowledge engineering, artificial intelligence mapping, and, most importantly for instructional designers' knowledge acquisition. Gordon (1994) modified these conventions specifically for use in developing technical training. Brock Allen and EDTEC 544 students further modified these conventions, adding color and shape.

 

Use the I-CARE menu on the left
to advance to the Connect segment.

Module 6
Cognitive Task Analysis

Connect

Interpreting Conceptual Graph Structures

Building conceptual graph structures is a process of making connections to represent knowledge structures. The following lesson explains the nodes and links that comprise a conceptual graph structure, and gives you a chance to practice interpreting a context illustrated by a conceptual graph structure.

EDTEC 544 Conventions

The nodes and links used in this module are conventions used in Ed Tec 544. There are other ways to construct conceptual graphs.  

Nodes

Conceptual graph structures employ seven types of nodes:  Goal, Goal/Action, Concept, State, Event, Style, and Information.  Definitions and examples of each follow in the table below.

 

NODE

DEFINITION

EXAMPLE

A circumstance, situation or state of affairs desired by a person or agent.

An action performed by a person or agent to attain a goal; an action not readily distinguishable from a goal.

A set of objects, entities or ideas that share a common name and common characteristics or properties.

A stable configuration of a system; a relatively stable situation or circumstance.

A change in situation or circumstance; a shift between stable states of a system.

Qualitative descriptors of goal/actions-such as speed, force or manner.

A source of additional information.

Interpreting a CGS: Eggs Served

The nodes below describe a context, a situation, a set of possibilities.  Take a minute to read the nodes out loud.  Talk to yourself about the nodes.  Describe the situation.  What happened?  What could happen?  What will happen?  What are the concepts involved in the situation?  What is the desired state of affairs?  Write a short paragraph that tells the story.

Now look again.  Is there a different story you could tell?

Read our story.

Interpreting a CGS: The Nail is Flush

The nodes below describe a contextual situation.  Think about the definitions of the nodes.  Are these items goals, goals/actions, concepts, states, events, styles, or information
 

Nail is Flush Nodes

When you're ready for feedback, click on the graphic to see the types of nodes.

Take a minute to read the nodes out loud.  Describe the situation.  What happened?  What could happen?  What will happen?  What are the concepts involved in the situation?  What is the desired state of affairs?  After you have thought through the nodes, write in the form below a short story based on the actions.  Highlight below feedback with your cursor for feedback.

Write a short analysis of the above nodes here.

Feedback: For feedback, triple click the white area below:
This set of nodes describes a cognitive scenario.  The goal in this scenario is "nail is flush".  Again, the goal has been described as a completed act.  The person wants the nail to be flush.  The act of hammering the nail helps the person accomplish the goal.  The state, "nail sticks up" is what causes the event "snagged socks".  The "snagged socks" initiates the goal "nail is flush".  The goal/action then is hammering the nail.  The style for hammering the nail is "gently".  The concept involved in this scenario is the category "nail".  This could be any number of kinds of nails.

Interpreting a CGS: The TV is Silent

The nodes below describe a contextual situation.  Think about the definition of the nodes.  Are these items goals, goals/actions, concepts, states, events, styles, or information
 

TV is Silent Nodes

When you're ready for feedback, click on the graphic to see the types of nodes.

Take a minute to read the nodes out loud. Describe the situation. What happened?  What could happen?  What will happen?  What are the concepts involved in the situation?  What is the desired state of affairs?  After you have thought through the nodes, write in the form below a short analysis and story of the scenario.  Highlight below feedback with your cursor for feedback.
 

Write a short analysis of the above nodes here.

Feedback: For feedback, triple click the white areas below:
 

TV is Silent:  An Analysis

This set of nodes describes a cognitive scenario.  The goal in this scenario is TV is silent.  Again, the goal has been described as a completed act.  The person wants the TV to be silent.  The act of throwing the hammer helps the person accomplish the goal.  The state, "TV is blaring" is what causes the goal of "TV is silent." The goal/action to accomplish the goal of TV is silent is "throw hammer."  The style for throwing the hammer is "violently".  The event that occurs after the hammer is thrown violently is "TV explodes" The concept involved in this scenario is the category "hammer." This could be any kind of hammer.

TV is Silent:  A Short Vignette

Slam.  Clomp, clomp, clomp.
"Hey everybody, I'm home!"
The sound of a blaring tv greets dad as an answer.
Clomp, clomp, clomp.  "Where's that hammer? Ah Ha!"
Swing, swing, swing, swoosh.  Hammer flies through the air with violent speed.
BANG!  CRASH! Glass flies and the kids take cover under the sofa cushions.
Silence now reigns in the room.  Dad smiles.  "That's much better."
The kids peek over the sofa pillows, astounded at dad's seemingly erratic action.

 

 

Links

We've just worked with the nodes used in Conceptual Graph Structures. The following section of instruction focuses on the links that connect the nodes and describe the relationships between them. For a detailed explanation of the links you can use in conceptual graphs see the job aid on "links" in the Extend section of this module.

Eggs Served: Practice With Links

The nodes below describe a contextual situation.  Think about the definition of the nodes.  Are these items goals, goals/actions, concepts, states, events, styles, or information?  Click on the graphic to view the types of nodes.

How would you use the links to connect the nodes?  Map it out for yourself. Then, in the form below, write your story, describing how you would use the links to connect the nodes.  Highlight below feedback with your cursor for feedback. Remember, there are many ways to map ideas, and yours may be very different.
 

Eggs Served Nodes Eggs Served Links
 

Feedback: For feedback, triple click the white area below:

The goal "eggs served" is connected to the goal/action "beat eggs" with the "initiates" link. The style "rapidly" is connected to "beat eggs" with the "manner" link because the style node describes the manner with which the goal/action is performed. The state "uncracked eggs" is connected to the event "eggs break" with the "before" link. The goal node "eggs served" is connected to the concept node "eggs" with the "refers to" link.

The Nail is Flush: Practice With Links

Read the nodes and the links. In the form below the nodes and links, write a narrative describing how you would use the links to connect the nodes. Highlight below feedback with your cursor for feedback. Remember, there are many ways to map ideas, and yours may be very different.

Nail is Flush Nodes Nail is Flush Links

Feedback: For feedback, triple click the white area below:

The state node "nail sticks up" is connected to the event node "snagged socks" with the "consequence" link. The event node "snagged socks" is connected to the goal node "nail is flush" with the "initiates" link. The goal node "nail is flush" is also connected to the concept node "nail" with the "refers to" link. The goal node is also connected to the goal/action node "hammer nail" with the "initiates" link. The manner with which you hammer the nail is "gently" so the goal/action node "hammer nail" is connected to the style node "gently" with the "manner" link.

 
Use the I-CARE menu on the left
to advance to the Apply segment.

Module 6
Cognitive Task Analysis

Apply

Probe Questions

There are two parts to the apply section of this module:

Part one:  Probe Questions Explained
Part two:  The Zuckoozi Movies:  Probe Questions in Action

You will navigate to part two of this module by clicking on "to Zuckoozi Movies" at the end of this page.

Probe Questions Explained

Creating Conceptual Graphs:  Overview

Now that we know something about the conventions of conceptual graphing we're ready to apply this knowledge to solicit information from subject matter experts.

Creating a conceptual graph is an iterative process. Initially the designer may begin with translation of information from unstructured interviews or documents. Usually the resulting graph shows a general overview of the domain. This initial graph can then be consulted to create probe questions. The probe questions provide the structure for the interviews of the subject matter experts. The designer may map the information gathered from the structured interviews onto the initial conceptual graph. The conceptual graph is then reviewed for level of desired detail. More probe questions are created to facilitate the gathering of missing information. More interviews are conducted. Observations can be conducted to gather information about tasks that the subject matter expert has automated and is incapable of verbalizing. The process is repeated until the expert and designer decide the graph is complete (See Gordon, p. 101 for more information).

Creating Probe Questions

Sallie Gordon (1994) has developed a generic set of questions that correspond to each type of node on a conceptual graph. Specific questions can be created by combining the generic question with the content of the specific node. The designer can then use the questions to structure the interview with the SME. During the interview the designer can then record the answers to the questions directly on the conceptual graph. The job aid below lists the generic probe questions.

Sample Probe Questions

Instructional designers use probe questions to collect information for conceptual graph structures. Note that the questions presented here are just examples. Make up your own!

Printing: To print this page, change the page layout in IE Explorer.  (Edit menu > Page Setup > Orientation > Landscape.

 

To Zuckoozie Movies (requires QuickTime)

 

Module 6
Cognitive Task Analysis

Reflect

 

The Content Performance matrix and Conceptual Graph Structures: Identifying the content to be taught

This part of the module provides you with another chance to practice identifying the content in a CGS. Look at the Mouse Moves CGS. Think about what content the Mouse Moves CGS represents. Refer to the Content Performance Matrix as a guide to help you classify the content.  Describe the concepts and procedures in the Mouse Moves CGS.  Write your thoughts in the form below the CGS.  Highlight below feedback with your cursor for feedback.

 

Feedback: For feedback, triple click the white area below:

Concepts:  highlight, icon, activate, icon, menu item, GUI objects, button

Procedures: activate object, move object

Layouts for Conceptual Graph Structures

Readability
Conventions for CGS Flow
Procedural relationships
Taxonomic relationships
Causal relationships
Compromising the conventions

Readability

Conceptual Graph Structures (CGS) challenge the limited possibilities of two-dimensional representation imposed by paper and computer screens. We can make CGSs easier to understand by following a few simple conventions for layouts.

Which CGS is easier to follow? They are exactly the same in terms of "link logic."

 
  Layout A: Jumbled Procedure, Messy Concept Hierarchy

 

  Layout B:  Streamlined Procedural Flow.  Clear Hierarchy of Concepts

Layout B is more readable because you can see the major structures and ideas quite easily. Notice, for example, the green trapezoids marching along to represent a sequence of goals and actions, and the hierarchical tree of cloud-like concepts descending from a topmost cloud.

 

Conventions on direction of flow
 

Here are the rules we're using to avoid traffic jams in CGS layouts.

 

Procedural Relationships
 

Procedural relations reflect sequences of actions, including means and outcome. They are ordered horizontally, from left to right because this is the direction people read English, Spanish, and other western languages. The conventions start with the goal, because people usually formulate a goal and then plan how to achieve that goal. This may seem counter-intuitive, because in a real-time sequence of events, the goal state is achieved after the learner executes the procedure.

But, we're concerned with mapping the way people think, or the way they need to think. So our map reads: "Hmmm, I want to make this facility accessible. What's the means for doing that? Oh, I need to unlock it with a key. What's the means for unlocking it? Oh, I need to insert the key?  Yada, yada, yada.

The means link indicates that an action is the "way" to accomplish a higher-level goal or goal/action.

The before link means that one action is performed before another.  This is a way of indicating that particular action must be performed before another action.

Often a goal can be achieved by multiple paths of action.  In the layout below, we can see that there are three "ways" to get to the goal state.

 

Note the small branch detailing the "means" for swiping the card. This demonstrates how easy it is to add additional detail to a CGS. Essentially any node can be "unpacked" in this way, depending on the need of the designer to establish, verify, or communicate additional assumptions about how the learner might need to think in order to accomplish the goal.

 

Taxonomic relationships

Taxonomic structures depict hierarchical classification systems or levels of abstraction, including "has property", "has part," and "instance." The ET544 conventions order these relations vertically with the broader or more abstract categories at the top.

Hierarchy: Concept--instance-->Concept

Hierarchy: Concept--has part-->Concept

Causal relationships

Causal structures represent cause and effect relationships. The ET544 conventions represent these relations diagonally between the procedural and taxonomic axes.

Compromising the conventions

Although the conventions for direction of flow make CGSs more readable at a glance, these conventions often require extra room. If it's important to conserve space to fit a CGS in a smaller space, you may decide to judiciously "bend" the conventions on flow to make for a more compact layout.

 
Use the I-CARE menu on the left
to advance to the Extend segment.

Module 6
Cognitive Task Analysis

Extend

  Please be sure to complete the FEEDBACK form (look in the left menu bar).

The CGS and Lesson Development

Identification of Instructional Design Strategies:

The Spreadsheet

Review the Spreadsheet CGS, noting the specific concepts and procedures to be taught.  Then consult Ruth Clark's book and write about what instructional strategies she suggests you should use to teach the concepts and procedures represented by the Spreadsheet CGS.

 

feedback

Job Aids

Click on the links below to read job aids associated with conceptual graph structures.

nodes  
links / syntax  

 

Concept Mapping Tools

SemNet is a software program (Mac only) for building concept maps.