Home / Introduction / Motivation / Methods / Applications / More Information

II. Behavioral Processes. These methods describe the sequences of behaviors necessary to perform tasks in the work domain, as well as rule-based decisions that determine when particular sequences are activated and how sequences interact. They are not suited for analyzing highly cognitive tasks, and they are of limited use in analyzing systems that support a flexible range of tasks in which the workers do not have clearly defined roles. But the techniques can be used as the starting point in cognitive task analyses by representing the tasks workers perform, and then zeroing in on aspects of those tasks that are cognitively demanding and require further analysis. 
   
  A. Task Analysis: Methods for producing detailed descriptions of the way a task is currently performed or could be performed.

References
    1. Behavioral Task Analysis: Decomposition of a task into a series of observable behaviors that are required to complete the task

References
    2. Operational Sequence Diagrams: (Also called procedural task analysis) A traditional task analysis technique in which the behavioral operations, inputs, outputs, and simple rule-based decisions necessary to perform a task are diagrammed in a flow chart in the order that they are to be carried out. The diagram may also be supported by a text description.

References
    3. Timeline Analysis: A detailed form of task analysis in which a temporally ordered sequence of actions necessary to achieve a task is constructed, along with duration estimates of each action.

References
    4. Operator Function Model (OFM): A hierarchic/heterarchic network of nodes that describe/prescribe the role of the operator in a complex system. At the topmost (heterarchic) level of an OFM are high-level functions that an operator performs, as well as the conditions that cause each of these functions to be initiated. These high-level functions are broken down into a supporting hierarchy of sub-functions. Connections between nodes indicate conditions that may initiate, terminate, or sequence activities. OFMSpert is the computer implementation of an OFM.

References
    5. Link Analysis: An annotated diagram showing the frequency of operator movements between system components is conducted. The diagram is then used to analyze the relationships between system components to optimize their arrangement by minimizing movement times and distances.

References
   
  B. Computational Task Simulation: These techniques are the analog of Computational Cognitive Modeling, but instead focus only on modeling the overt behaviors necessary to perform tasks, rather than the underlying cognitive activities that drive task performance. Typically, human performance data that have been previously collected are provided as input to the simulation. The simulation can either simulate graphically the environment and workspace, or dynamically "run" the task in real or fast time as a way of estimating complete cycle times, error likelihoods, workload, etc. These techniques can be used to assess potential contributions of alternative configurations of tasks, equipment, and team organizations. They can also aid in the design and analysis of tasks by assessing how the characteristics, interactions, and sequences of tasks can impact operator workload. Further, they can be used to assess the effects of proposed changes to an existing system on operator workload and productivity without the need for person-in-the-loop testing.
    1. IMPRINT: (Improved Performance Research INtegration Tool): An event-based task network in which a mission is decomposed into functions that are further decomposed into tasks. The tasks are linked together in a network that represents the flow of events. Task performance time and accuracy, as well as expected error rates are entered for each task (this data can be gained either empirically or from published research studies). IMPRINT can also incorporate the effects of training, workload, personnel characteristics, and environmental stressors on performance. IMPRINT models can be used to compute expected performance in terms of speed and accuracy.

References
    2. CART (Combat Automation Requirements Testbed): CART is an evolving software tool under development by Air Force Research Laboratory and is used to develop human performance models and subsequently integrate them with military simulations to address human performance modeling questions. CART is built on IMPRINT, and as such it is a task networking modeling environment that allows for human behavior to be represented in terms of the tasks and functions an operator performs. CART extends IMPRINT in two key areas: 1.) it has a goal orientation feature, and 2.) it provides HLA (High Level Architecture) and COM (Common Object Model) interfaces. 

References
    3. Micro Saint (System Analysis of Integrated Network of Tasks): Micro Saint is a general purpose, discrete-event simulation software tool that is commercially available. Micro Saint models of task execution yield estimates of times to complete tasks and task accuracies, as well as estimates of workload and task load (i.e. the number of tasks an operator has to perform over time). 

References
    4. WinCrew: WinCrew is used for constructing system performance models for existing or conceptual systems when a central issue is whether the humans and machines will be able to handle the workload. WinCrew also can be used to predict operator workload for a crew given a design concept. Additionally, WinCrew can simulate how humans dynamically alter their behavior under high workload conditions, including the dropping of tasks based on task priority, task time, and accuracy degradation.

References
    5. Integrated Performance Modeling Environment (IPME):  
5. Integrated Performance Modeling Environment (IPME): IPME provides a full-featured discrete-event simulation environment built on the Micro Saint modeling software. It provides a means to model performance shaping factors (i.e. environmental factors such as temperature, humidity, time, etc. that impact task performance), performance shaping functions (define how performance shaping factors affect performance), a scheduler to simulate operator loading, and an algorithm for estimating operator workload. Its modeling framework is based on Perceptual Control Theory (PCT, see Section V, Part A).

References