Challenges in sharing information effectively: examples from command
Diane H. Sonnenwald
Swedish School of Library and Information Science
Göteborg University & University College of Borås
Högskolan i Borås, Borås, Sweden
Introduction. The goal of information sharing is to change a person's image of the world and to develop a shared working understanding. It is an essential component of collaboration. This paper examines barriers to sharing information effectively in dynamic group work situations. Method. Three types of battlefield training simulations were observed and open-ended interviews with military personnel were conducted. Analysis. Observation notes and interview transcripts were analysed to identify incidents when group members erroneously believed they had shared information effectively and were collaborating successfully, i.e., a deceptively false shared understanding had emerged. These incidents were analysed to discover what led to these unsuspected breakdowns in information sharing.
Results. Unsuspected breakdowns in information sharing emerged when: differences in implementations of shared symbols were not recognized; implications of relevant information were not shared; differences in the role and expression of emotions when sharing information was not understood; and, the need to re-establish trust was not recognized. Conclusion. The challenges in information sharing identified here may extend to other high stress, unique and complex situations, such as natural disasters. Recommendations for more effective information behaviour techniques in dynamic group work situations are presented.
The goal of sharing information is to provide information to others, either proactively or upon request, such that the information has an
impact on another person's (or persons') image of the world, i.e., it changes the person's image of the world, and creates a shared, or mutually compatible working, understanding of the world (Berger and Luckmann 1967.)
Information sharing includes providing information, confirming the information has been received, and confirming that the information is jointly understood. Information sharing is an important component of information behaviour. It is an essential activity in all collaborative
Davenport work, and helps to bind groups and communities together (e.g., & Hall 2002.) When working together group, or team, members must continually provide information to others and to some degree mutually understand and use information others provide. When information is not effectively shared, collaborative group work fails.
Recent research in information sharing has focused on identifying types of and motivations for sharing information (e.g., Talja 2002; Davenport
& Hall 2002), use of technology in sharing information (e.g., Walsh et
2000), and the process or stages of information acquisition for al.
subsequent sharing (e.g., Rioux 2005.) In comparison, this paper
investigates challenges in sharing information effectively in group work contexts. How and why do group members fail to effectively develop a shared understanding of information they provide to one another? To explore these issues, I examined information sharing in the context of command and control (command and control) at the battalion level in the military. command and control is an example of a dynamic work context, in which decisions are made and tasks completed under stringent time constraints in order to manage changing emergency situations. In command and control, team members are well aware of their need to provide useful information to others and their need to use information provided by others. This is part of command and control training and practice. Yet during the command and control process, problems sharing information effectively occur - sometimes with devastating consequences.
Observation data from three types of battlefield training simulations and interviews with military personnel with command and control experience were analysed to identify challenges to sharing information effectively. Four types of challenges were identified: recognizing differences in the meanings of shared symbols; sharing implications of information; interpreting the emotions used in sharing information; and, re-establishing trust after incorrect, critical information is shared. These challenges were not caused by errors that can be solely assigned to either the information provider or recipient. Rather, they appear to be caused by mutual misunderstandings that emerge from typical practices found in our everyday use of language and symbols and in everyday work and cultural practices in organizational settings.
In this paper, the theoretical framework used to investigate information sharing is discussed, followed by a brief description of command and control that highlights the role and importance of information sharing in command and control. Next, the data collection and analysis methods used to identify challenges to effectively sharing information are described. These challenges are then presented using examples found in the data. In conclusion, information sharing techniques to overcome these challenges are proposed.
provides a framework in which to investigate The concept of common ground
information sharing. is the information, knowledge and Common ground
beliefs, which a group (of two or more) have in common and their awareness that the group has this information and knowledge in common (Clark 1996;
Olson & Olson 2000.) Olson and Olson (2000) suggest that common ground
is essential for successful collaboration, and discuss how various communication media support and/or impede common ground. Carroll et al.
(2006) propose that common ground is an important construct for analysing activity awareness in group work.
Group members implicitly and explicitly reflect on their common ground, seeking and providing evidence of it (Clark 1996; Schutz & Luckmann 1973,
1989.) The theory of common ground (Clark 1996) in conjunction with
activity theory (Engeström 1987) suggest that evidence of common ground
may be implied from circumstantial evidence, such as physical appearances, language, dress, organizational membership, organizational roles, cultural and/or community affiliations, and objects in the current environment. Evidence of common ground may also emerge from episodic evidence, such as events and actions that group members take part in or observe. In particular, evidence may emerge through joint communicative actions that are interpreted through various cultural, situational and individual lenses.
Each joint communicative action has a presentation and acceptance phase (Clark 1996.) In the presentation phase, Person A presents information, or a signal, to Person B. Person A further assumes that if B provides appropriate evidence of understanding, then A can believe that B has understood the information presented. In the acceptance phase, Person B accepts the information from Person A by providing evidence to A that B believes he or she understands the information. Person B assumes that A will then believe that B understands the information.
To increase our understanding of information sharing, common ground suggests that we should analyse joint communicative actions. Thus, this paper focuses on instances of joint communicative actions in which all participants erroneously believed they had a shared understanding. In these instances, episodic evidence of a shared understanding was provided through ongoing and previous joint actions in conjunction with circumstantial evidence. Yet after the collaborative group work failed participants came to understand there was no, or a flawed, shared understanding. Analysis of these failures suggests how episodic and circumstancial evidence of common ground provided in joint communicative actions can be misinterpreted and ways to avoid such misunderstandings in the future.
Command and control
Command and control (command and control) at the battalion level is
performed by teams of experts who are usually organized into three groups: the command group, the administrative logistics operations centre (hereafter, 'logistics centre'), and the tactical operations centre (Harrison 1995; McIlroy 1995; Jarett 1995.) The command group,
specifically, the Commander, assigns and gives the battlefield mission to the tactical operations centre. The logistics centre, led by an S1
, collaborates with the Personnel Officer and S4 Logistics Officer
Commander and tactical operations centre in planning and preparing for the battle to ensure that soldiers will have the personnel services and supplies when and where needed. Additional details on the organizational structure of the command group and logistics centre can be found in Sonnenwald & Pierce (1996.) Information behaviour in the tactical
operations centre is the focus of this paper.
The tactical operations centre typically consists of four elements, or subgroups: the S3 Plans and Operations, S2 Intelligence, Fire Support, and Signal elements. Each element is led by an officer. The S3 Operations Officer is responsible for planning and supervising battle preparation and battle execution. That is, during a battle the S3 Operations Officer is responsible for creating and maintaining situational awareness of the battlefield, including enemy movements, civilians in the area, and the logistics situation, and making decisions regarding battlefield actions for his troops. The S3 Operations Officer, who usually holds the rank of major, is assisted by two or three captains, including an assistant S3 Operations Officer, and two to four non-commissioned officers (NCOs.) The S3 Operations Officer may also be assisted by a Sergeant Major who provides detailed technical knowledge about the terrain and other aspects of the situation.
In addition, a nuclear, biological and chemical (NBC) NCO and liaison personnel from other tactical operations centres are part of the S3 Plans and Operations element. The NBC officer is commonly referred to as the 'bugs and gas' person and is responsible for advising the S3 Operations Officer on NBC threats. The liaison persons represent their tactical operations centres and help coordinate collaborative efforts among those centres.
The S2 Intelligence element, led by the S2 officer, is responsible for gathering and interpreting intelligence information on the enemy, including enemy equipment, enemy movement, estimates of enemy strength and locations, possible enemy targets, and the enemy's potential course of action. They may also provide information on the weather and terrain. Fire support, led by the Fire Support Officer, plans fire missions and provides (or calls for) fire support during the battle. The Signal element, led by a Signal Officer provides telecommunications support for the battalion. They may go into the battle area in advance to set up telecommunications networks, as well as work to maintain those networks and keep them secure during the battle.
On occasion, other elements with specialized expertise may be assigned to work with the tactical operations centre. For example, an engineering unit may be assigned to help the battalion traverse obstacles, such as rivers, in the terrain.
As discussed in the literature (e.g., Alberts & Hayes 2006; Sonnenwald
& Pierce 2000), officers and their staff in the tactical operations centre share information, ideally, developing a shared working understanding of the mission and battle and working in a coordinated fashion to achieve the mission. In a formal command and control task analysis, it was determined that 58% of command and control tasks require everyone's participation and an additional 25% of tasks require participation from everyone except one person (Sonnenwald & Pierce 1996.) As one participant
reported: 'Everyone plays a role—feeding or drawing information from the
Personnel in command and control units are taught this and train for it throughout a variety of courses and battle simulations. However, even though information may be shared among command and control personnel, a common understanding of the information may not exist. This problem is so severe it can make the difference between winning and losing a battle. Research methods
Data from observation of battlefield simulations and interviews were used in this research. Although originally collected for another study (Sonnenwald & Pierce 2000), the variety and richness of the data provided multiple examples of information sharing that illustrate its complexity and texture. In particular, examples of situations where participants thought they had effectively shared information but later learned this was not the case were selected for analysis.
Three simulated battlefield training exercises were observed. One battlefield training exercise took place at a military battle simulation centre in the U.S. The centre trains hundreds of military personnel annually from around the world in command and control, leadership development, and battle management skills. The training exercise observed was part of an officers advanced course. The students were army captains who had been together in the training course for six months at the time of the exercise. They had participated in multiple field training exercises and three other similar simulation exercises. Course instructors reported this group had demonstrated a high level of performance on previous tasks and exercises. The opposition in the battle was played by simulation centre personnel. During the exercise, I observed command and control in the air assault tactical operations centre including their preparation for the battle, participation in the battle, and after action review session.
A second battlefield simulation exercise was observed at another U.S. military base. The purposes of these exercises were to test the use of proposed new battlefield equipment and to provide training for a military unit. The enemy and other specialized support services, such as weather forecast, were played by instructors and officers recruited from the appropriate military branches. During this exercise, I observed the command and control preparation for the simulated battles, activities during several battles, and their after action review sessions. The third type of battlefield simulation exercise observed took place at the National Training centre at Fort Irwin, CA. I observed the same military unit I observed earlier (although there were many changes in tactical operations centre personnel in the six months that had elapsed since the earlier observations.) I joined the unit during the twelfth day of a fourteen day period of intense battle simulations. When I joined them, the unit had just finished a battle simulation and was beginning to prepare for the next battle. The opposition in these simulations is played by U.S. military personnel judged to be the best in their areas of expertise. These
simulations involve other units from other branches of the military, and are the largest and most complex simulations performed in the U.S. military. I observed the unit's preparation for the battle, participation in the battle and after action review. This involved living with the unit for two days and nights under field conditions in the desert. As before, I observed activities in and surrounding the tactical operations centre. The peripheral membership role (Adler & Adler 1987) was chosen when
observing to minimize the potential of the study to influence the participants' behaviour. Thus, I introduced the purpose of the study to the participants and myself, but did not perform any tasks or offer advice to the participants during the exercises. Interactions among group members, interactions between group members and higher and lower echelons, and interactions among other groups were observed.
In the ethnographic tradition (Lofland et al. 2005), note taking was used
extensively to record data. Notes, video recordings and photographs were made while observing events during the exercise. Later, away from the setting, the notes were augmented with sketches of areas where the exercise took place and additional details about events and interactions, using the field notes as prompts; and summaries of overall impressions about events which occurred during the simulation.
The observational data were not from an actual battlefield situation per se because it is impractical to observe an actual battlefield situation. However, the high degree of cognitive and emotional involvement of participants in simulations and the similarity of their behaviour, to behaviour in actual situations, has been observed in other studies. Interviews
To augment the observation data, formal and impromptu interviews with experienced military personnel who participated in the simulations were conducted. Each interview participant had between 6 and 40 years of military experience.
Each formal interview consisted of an initial introduction which described the purpose and nature of the study, the confidentiality of the participant's responses, and the participant's right to request clarification or to not answer questions. This introduction was followed by a series of open-ended and critical incident questions (Flanagan 1954.)
The open-ended questions focused on organizational structure, activities and roles in command and control. The critical incident questions asked participants to discuss their most satisfying command and control
experience and their most dissatisfying command and control experience. This technique allowed me to learn about conflicts, successes and failures in command and control. Each interview ranged from 1 to 2 hours in length. Audio-recording was used during the interviews, and recordings were transcribed.
Impromptu interviews were conducted before and after the simulation exercises. During these interviews participants clarified and shared their perceptions regarding command and control activities. Note-taking was used sometimes during but primarily after the interviews to capture these discussions.
The data were analysed to discover examples of information sharing that were perceived as failures by participants. Failures were defined by participants as those incidents that directly led to actual deaths of friendly forces or there was a strong likelihood that deaths would have occurred if the situation occurred in real life as opposed to occurring during a battlefield simulation.
From the data analysis, several types of challenges to effectively sharing information emerged. Each challenge is discussed below.
Recognizing differences in the underlying meanings of shared symbols
Different social groups, organizations and disciplines may develop unique meanings for symbols. For example, the non-flashing yellow light in a traffic light has only one meaning in the U.S. (begin to stop your vehicle, red light is imminent). Yet the same symbol has two meanings in Sweden and other European countries (begin to stop your vehicle, red light is eminent - or - get ready to enter the intersection because a green light is eminent.) There can be clues, such as others' behaviour and others' reactions to our own behaviour, to help us detect and resolve such differences in the meaning of symbols. Working together over time, establishing a common ground, is suggested as necessary to identify and resolve such differences in meaning (Clark 1996.) However, as the
following example illustrates this is not always the case.
During battlefield simulations it is typical for those assuming the responsibility of command and control to develop a map that tracks their unit's locations and enemy locations. In classroom settings push pins (i.e., a pin with a thick plastic cover at one end that allows you to easily push the pin into a bulletin board) are often used to indicate these locations. Different colours are used to indicate the locations of one's own forces and enemy forces.
Figure 1: The US Army Field Artillery flag
In one observed simulation, participants from different branches of the military (and in some cases from different countries) had been training together during the previous six months and had jointly participated in two previous simulation exercises. The participants assuming the responsibility for command and control during their third joint simulation exercise were members of the US Army Field Artillery. The Field Artillery colours are red and gold, as found in the Field Artillery flag (Figure 1.) These colours are used in many ways to represent Field Artillery. For example, the 'The Artillery Man' Website has a similar red
background with gold text. Thus, the field artillery personnel chose red pushpins to represent friendly forces on the map. When selecting this colour participants commented that field artillery is always represented by the colour red. The enemy in this simulation was called the 'Red Armed Forces' and the field artillery personnel used silver pushpins to represent enemy locations.
During the simulation exercise other participants who were not from Field Artillery and who needed to use the map assumed that the red push pins represented the Red Armed Forces. They made decisions regarding battlefield actions based on this assumption. They did not question the meaning of the push pins because, as they later explained, from their perspective the push pins were being used appropriately.
The meaning of the red pins was not questioned until near the end of the battle when action was slowing down. At that point personnel coming to look at the map had received additional information about enemy locations that was in obvious conflict with the red pins on the map. When this issue was brought to light, the field artillery personnel justified their actions saying: 'But Field Artillery is always red'.
They expressed surprised that the discrepancy in meaning had not been detected earlier. The successful common understanding of push pins in previous simulations and the quickly-changing troop locations were mentioned to help explain why others had not realized earlier that the red pins represented friendly forces.
As this situation illustrates, a symbol may have a commonly understood, or shared, function (representation of troop location) but an uncommon or specialized implementation (representation of friendly troop location) only known to some but not all group members. The differences in implementation may not be recognized either by the persons using the symbol to convey information or by the persons viewing the symbol to gain information. In the situation described, both the presentation and acceptance phases surrounding the symbol were consistent with participants' expectations.
The problem is made worse when the information the symbol is intended to convey is dynamic, i.e., changing quickly. In such situations, if participants suspect differences in their common ground, they may assume the differences are only due to changes in specific information content and not to differences in meaning. The differences between intended and perceived meanings may only be recognized when other trusted information and/or outcomes indicate large discrepancies between the intended and perceived meanings.
Sharing implications of information
When a young child goes to put their hand on a hot burner of a stove, a nearby adult will quickly tell the young child not to do that 'because the burner is hot'. The adult will typically further explain that the hot