The Prime Dynamics of System Computation
Yi-Ming, Tu, Wei-Young, Wang
Department of Management Information System
National Sun Yat-sen University, Kaohsiung, Taiwan, R.O.C.
ymtu @ mis.nsysu.edu.tw, wyoung @mis.nsysu.edu.tw
Abstract
From the perspective of socioculture cognitive system, the operation of teamwork is
considered as a cognitive computation system essentially. Distributed and related
person-technology interactions determine a cognitive system's properties. Each task
performer proceeds organizing activities in order to do the tasks. In the organizing
processes, he or she coordinates with other team performers, instruments, procedures,
etc. Such interactions influence the teamwork system’s cognitive properties and
determine it’s computational power, such as learning ability, flexibility, robustness, etc.
That is to say the dynamic behaviors of group performance will depend on its
underlying interactions. Based on Hutchins’s research in 1996, this research develops a
system dynamics model to explore and to understand the prime mechanism of teamwork.
Fortunately, this paper provides a significant contribution on the process of team
operations.
1. Introduction
The strength of team work supports us to accomplish certain tasks by operations of
distributed, parallel, interactive, and adaptive mechanisms (Hutchins, 1996; Williams,
1996; Ray D. and H. Bronstein, 1995). Today, teamwork is the major form for us to
solve complex problems. However, a team's work does not equal to the sum of each
member's work. Team is the emergence of its components and it has its own properties
and characteristics. The operation of a team's work reveals specific dynamic and
non-linear characteristics, which makes itself different from other teams. Teams are
systems composed of individuals. Among these individuals, various kinds of
interactions can be found. They are interactions between members, interactions
between members and environments, and interactions between members and media,
tools, and technologies. These interactions are substantial to a team's operation,
therefore, it is important to understand its operating processes.
This paper intends to examine the adaptive mechanism of a self-directed team (Ray
D. and H. Bronstein, 1995) from the cognition perspective. The mechanism represents
actions and related information feedback between key elements. As the mechanism
operates, multiple kinds of dynamic and time varying patterns are generated, which are
known as teams' behaviors. This research is based on the assumption that we can find
some common and fundamental mechanisms behind all teams from previous
researches of cognitive science, cognitive anthropology, and other related fields. This
paper attempts to model the mechanisms of computational structure, cognitive load
adjustment, detection of errors and rework, experienced learning and task governance
for a better understanding of teams.
2. Dynamic Behavior and Organizing Process
2.1.Dynamic Behaviors of Navigation Team
This research is concerned with a navigation team, which reorganizes itself to
overcome environmental changes (Hutchins, 1996). Major focus is on time varying
patterns of the team's performance and related cognitive properties, including cognitive
load, mutual understanding among members, knowledge redundancy about tasks, task
governance and sharing, and communication overhead among members. To understand
these patterns is the model's purpose.
2.2. The Goals of Organizing Process
In the analysis of the navigation team’s response to environmental change, there
exist interrelated and goal-seeking processes. The goals, implicit or explicit, of each
process function as motivations of the team’s actions (Forrester, 1961). In the
navigation team, process
s seek major goals of performance, task quality, functional
effectiveness, and cognitive load limitations. Among these goals, functional
effectiveness is the most important, because without functional effectiveness the team
can not move on and is in danger immediately. The figurel.1 to figure 1.3 depict the
main structure of the navigation team under the purpose of team’s adaptive mechanism.
Figure 1.1 illustrates a loop to maintain the alignment between desired performance and
the team’s real performance. The loop also connects to an explicit cognitive adjustment
loop, which represents team members’ efforts to change their ways of carrying out tasks
for the reason of cognitive economy. Those changes may be new instruments, new
procedures, or whatever depending on their environmental availability. Because those
changes are unplanned behaviors, they also affect local interactions between members.
Therefore, time is needed for the team members to rebuild their mutual understanding
about interactions. Besides, members may ignore some minor tasks and lower task
quality standards under extreme cognitive loads.
Cognitive Capacity
Desired Work Rate
Cognitive Load to task
Stress to Task
Skipped Subtasks
Real Work Rate
Potential Work Rate
Mismatch Rework
Introduce New Tools and Methods
Mismatch
Mutual Understanding_—"”
Fig. 1.1 Causal Loop Diagram of the Navigation Team’s Adaptive Process 1
Figure 1.2 presents the members’ self-regulating mechanism. Task governance is
the way of collaboration among team members. It can provide more efficient learning
environment for a new comer or higher quality of tasks performed by team members.
For good honors or relationships, self-regulations are also essential. Both task
governance and self-regulations take place when the team members have enough
cognitive resources and knowledge redundancy. Knowledge redundancy is the team’s
overlapping distributions of knowledge among the members. To increase knowledge
redundancy means that the team members have to spend more time to work together.
However, when knowledge redundancy is too low, the team has no ability to perform
task governance.
Cognitive Capacity Remained
ae Knowledge Redundency
Cognitive Load to Governance
Task Governance Cognitive Load to Task
Error Notification Ny Members Ratio
— Real Work Rate
Fig. 1.2 Causal Loop Diagram of the Navigation Team’s Adaptive Process 2
Figure 1-3 presents the mechanism of allocating human resources to achieve
desired performance. The adjustment of the member size influence the old/new
members ratio and the shared tasks’ performances. In the navigation team, the member
size is fixed. But when the desired performance is too low, members may transfer from
other tasks temporarily. The figure also shows how the team’s knowledge redundancy
changes. By tasks performance sharing, members have opportunities to learn to learn
the others’ tasks, especially, under the motivation of maintaining the team’s functional
effectiveness.
Task Governance
Members
Desired Members Old Members Ratio
/ ‘Communication Overhead
Knowledge Redundancy
Desired Work Rate \
Potential Work Rate ( \
Shared Tasks Performance
Functional Effectiveness
Cognitive Load to Task Leaming
Cognitive Capacity Remained ~ Shared Tasks Learning
Fig. 1.3 Causal Loop Diagram of the Navigation Team’s Adaptive Process 3
3.Model Behavior
This section discusses some variables’ patterns of the navigation team in a real
crisis. When the navigation team was navigating the board into a hub, the electronic
power of the board was fail. Almost instruments in the navigation team can not run
properly and the ship was very dangerous in that time. The time between the power
failed to the ship anchored and stopped was about 120 minutes. In the model, this
scenario is designed with desired work rate changes from 10.8 to 14.4 tasks/minutes at
time 50 and potential work rate decreases to 60% against normal situation. The
generated patterns are listed in figures 2-7.
Desired WA 2: Real WR 3: Work Rate
4 2000
3
5
1
: 10.00
i
:
;
i 0.00
100 20075 40050 60025 00.00
Tine
Fig. 2 Desired WR: desired work rate (tasks/time, at time 50,from 10.8 to 14.4),
Real WR: real work rate (task/time), Work Rate: potential work rate (task/time, at
time 100,decrease to 60%)
1.00 200.75 400.50 600.25 800.00
Time
Fig. 3 MU: mutual understanding of operations between members (MU unit)
1: Errors 2: Mismatch
gs
y
;
das
;
Pr,
10 ‘on ‘oom ons vaio
Time
Fig.4 Errors: errors remained (error unit), Mismatch: mismatched remained
(mismatch unit)
1: Coo Task
1 1.00
1.00 200.75 400.50 600.25 800.00
Time
Fig. 5 Cog Task: cognitive load to perform tasks (cognitive unit), ETG: cognitive
load to perform governance (cognitive unit)
1.00 200.75 400.50 600.25 800.00
Time
Fig. 6 TQ: quality of task performed (index)
1.00 200.75 400.50 600.25 800.00
Time
Fig. 3 KR: overlapping distributions of knowledge among the members of the team
(knowledge redundancy unit)
4.Structure Design for Team Learning
By means of model developing and simulation, we can explicitly understand how a
team could concentrate on and adapt itself to the environmental changes succ
ssfully.
Adaptability is the most important capability of a self-directed team, and it is also the
main merit for self-directed team to exist. Such a capability does not occur by accident,
but is obtained from long-term structure. The sudden environmental change is a
temporary event to the navigation team, but it reveals the team’s capabilities to cope
with certain environmental variations. The capability is determined by the way a team
responds.
The competence of a system’s reaction to the environment is determined by a routine
structure, together with shared and distributed mental models (Senge, 1990). By the
model we understand more about why a team can adapt successfully by self-organizing.
The structure also indicates how a system gets the abilities needed to cope with
uncertainties of environment. To understand the structure is to be aware of how the
team learns. With the understanding and awareness, one can design and improve the
structure by model experiments.
This paper illuminates that mutual understanding and knowledge redundancy are
more important in this case. They are basis of a successful self-organizing of team. The
behaviors of mutual understanding and knowledge redundancy are worthy of further
endeavors.
5. Conclusion
This paper examines a team’s self-organizing process in a dynamic environment with
a quantitative model. With the mathematics model we can understand the interactions
between elements more deeply than other method. Some operating mechanisms such as
cognitive load adjustment and task governance are found to be more fundamental and
common for self-directed teams. This research is just a beginning by approach to
understand why teams can cope with the changing environment and how they to get and
where they to get the abilities they need.
References
Forrester, J.W.(1961). Industrial Dynamics, MIT Press, Mass.
Hutchins, E.(1996). Cognition in the Wild, MIT Press, Mas
Ray D. and H. Bronstein.(1995). Teaming up: making the transition to a self-directed,
team-based organization, McGraw-Hill, New York.
Senge, P.M.(1990). The Fifth Discipline-The Art and Practice of the Learning
Organization, Doubleday, New York.
Williams, H.(1996). The Essence of Managing Groups and Teams, Prentice Hall, New
York.
