1 of 18

Enhancing Common Ground in a Command and Control Task��(ICCRTS Paper 72)

Andrew Leggatt, Huw Gibson, Simon Attfield, George Raywood-Burke, Nicola Turner

2 of 18

Study purpose

Empirically test one of theoretical principles of DSM.
Gain insight into whether we can enhance DSM performance in a significant manner.
To evaluate novel measures of DSM.

3 of 18

Sensemaking in C2

Sensemaking plays a critical role in command and control (C2) systems by helping decision-makers process complex and ambiguous information to make informed decisions under pressure.

Organise information

Seek and gather information

Adapt and anticipate

Prioritise and focus

Collaborate negotiate

Understand

Explain

Decide

4 of 18

Nine Principles for Distributed Sensemaking

Provide sufficient cues for sufficient sensemaking.
Support low-cost information workflows.
Represent information quality and provenance.
Promote expertise and associated domain knowledge.
Allow time to acquire information to build an evidence-based and coordinated situation picture.

4

Use strategies for the negotiation of sense.
Where appropriate, use strategies for frame enumeration and elimination.
Provide explanatory context for actions, orders and requests.
Minimise the costs of achieving and maintaining common ground.

Attfield, Minocha, Elliott, Fields, Baber, Hutton, Leggatt, Harryman (2021). Nine Principles for Supporting Distributed Sensemaking, Naturalistic Decision Making and Resilience Engineering Symposium 2021, Toulouse, France.

Elliott, Attfield, Minocha, Fields, Hutton, Baber (2020) Enhancing Sensemaking: Supporting Distributed Groups in the Future Operating Environment. 25^th International Command and Control Research & Technology Symposium (ICCRTS).

5 of 18

Common Ground (Herbert Clarke)

Shared knowledge, beliefs, and assumptions.
What are we doing? How do we do it? What’s happening? What do we call things? What’s important here?
For communication and coordination to be effective, parties need mutual understanding or ‘common ground’.
Arises in virtue of e.g. shared environment, experiences, expertise, culture.
Parties in a communication aim to minimize the overall effort needed to achieve the communication goal.
Different technologies and practices may enhance or limit the achievement of common ground.

Common Ground

?

6 of 18

Study Outline

Groups of three participants collaborated in a series of C3Fire simulations that involved fighting the spread of fires through rural and built-up environments.
Aim of task was to work as a team putting out fires and rescuing people.
Participants worked at separate workstations and could not see each other’s screen.
Two manipulations used to affect participants’ Common Ground

The extent to which positions of units under the control of each participant as well as locations of fires were included in a shared (plan) view of the simulation environment (Situational Common Ground).

7 of 18

C3Fire Task

Grid with basic geographical information visible to all participants on their individual laptops.
Participants instructed fire trucks, water tankers and rescue vehicles.
Dependencies meant that participants needed to coordinate.
Interface displayed clock, wind speed/direction, information about units under their control (e.g. water levels and the number of simulated people at unit locations, passengers on rescue vehicles).
Newspaper articles with intelligence at each run.
Goal = save as many people as possible
Fire could start at any location and would spread.

8 of 18

9 of 18

Manipulating Situational Common Ground�(within participants)

Situational common ground = shared understanding in virtue of the situation or context in which the communication is taking place.
Includes the physical setting, social context, and immediate events.

Shared Unit

See locations of all units, plus any fire occurring within one grid square of all units.

God’s Eye

See locations of all units plus all fire.

Unit Only

See locations of own units only, plus fire occurring within one grid square of own units.

10 of 18

Manipulating Cultural Common Ground�(between participants)

Cultural common ground = cultural practices, norms, values, and shared histories that arise in virtue of being members of the same cultural, social or institutional group.
Broad in scope and tricky to manipulate in the short term.
Structured method using (Post-its) in which participants discuss, shared, and co-develop mental models, goals or conceptual frameworks relating to the task, between task runs.

11 of 18

Greco-Latin Square Counterbalancing

Within Participants Manipulation

U = Unit only

S = Shared unit

G = Gods eye

Scenario

(1) = Scenario 1

(2) = Scenario 2

(3) = Scenario 3

Between Participants Manipulation	Group	Run
		1	2	3
Post-it task	1	U(1)	S(2)	G(3)
	2	S(3)	G(1)	U(2)
	3	G(2)	U(3)	S(1)
	4	U(1)	S(2)	G(3)
	5	S(3)	G(1)	U(2)
	6	G(2)	U(3)	S(1)

Individual reflection	7	U(1)	S(2)	G(3)
	8	S(3)	G(1)	U(2)
	9	G(2)	U(3)	S(1)
	10	U(1)	S(2)	G(3)
	11	S(3)	G(1)	U(2)
	12	G(2)	U(3)	S(1)

12 of 18

Study Design

Training

runs

C3Fire

experimental run 2

C3Fire

experimental run 3

C3Fire

experimental run 1

Thirty-six participants (27 male, 9 female) as 12 teams of three people.
Three scenarios (counterbalanced for order)
Three interface conditions counterbalanced for order and scenario.

6 teams

(Cultural Common Ground)

6 teams

(Individual Reflection)

Performance measure (number of casualties)

Sensemaking Assessment Questionnaire (SMAQ)

Measures

Distributed Sensemaking Worksystem Screening Tool

(DSM WST)

Post-it task

Individual reflection

Post-it task

Individual reflection

Post-it task

Task runs

13 of 18

A Typical Day

Time	Activity
09.30-09.50	Introductory briefing
09.50-11.00	Training Part 1
c. 11.00-11.15	Break
11.15-12.30	Training Part 2
12.30-13.00	Lunch
13.00-14.05	Experimental run 1
14.05-15.05	Experimental run 2
15.05-15.20	Break
15.20-16.20	Experimental run 3
16.20-17.00	Wrap-up session

14 of 18

Number of simulated casualties by trial run

Mean number of simulated casualties by trial run and cultural common ground intervention.

15 of 18

Number of simulated casualties by condition

Mean number of simulated casualties by situational common ground intervention (Unit Only vs Shared Unit vs God’s eye) and cultural common ground intervention (No Post-it vs Post it)

16 of 18

Sensemaking Assessment Questionnaire (SMAQ) by condition.

Mean SMAQ ratings by situational common ground intervention (Unit Only vs Shared Unit vs God’s eye) and cultural common ground intervention (No Post-it vs Post it).

17 of 18

Sensemaking Assessment Questionnaire (SMAQ) by trial.

Mean SMAQ rating by trial run and cultural common ground intervention.

18 of 18

Discussion

Test one of theoretical DSM principles (e.g. common ground)

Cultural common ground – helping people to understand each other’s personal models and convergence
Situational common ground

Enhancing DSM with interventions – cultural common ground

Provides performance improvement immediately (R1)
Similar levels of improvement to additional experience – accelerating the acquisition of expertise?
Continues to add performance even with greater experience

Enhancing DSM with interventions – situational common ground (more common information)

Provides significant improvement in team performance.

SMAQ showed a trend in the suitable direction with the performance data (under powered experiment)
C3Fire provides a suitable experiment environment to conduct C2 studies
Evaluated measures of DSM – see later papers.