Real-time monitoring of individual and team states and processes.
The first critical capability is the ability to assess and deliver technologies to enhance human-agent team performance to the right individuals at the right time, to enhance the performance of the group. This builds on current capabilities focused on continuous monitoring of individual states to optimally tailor augmentation technologies to the individual in order to enhance individual performance. Instead, however, this capability requires (1) aggregation of multiple data-types to provide valid measures of team states and processes, and (2) a systems-based perspective, continually monitoring all nodes within the human-agent team to determine the optimal intervention in real-time. This holistic assessment is then used to tailor delivery of technologies to enhance team states, processes, and performance in the most effective ways. Importantly, this could include tradeoffs in individual performance for the benefit of the collective, or varying interventions for particular individuals or subgroups in order to achieve desired states of the collective.
Adaptation during complex events in a dynamic environment.
In addition to understanding the individual and team dynamics in order to deliver at the point of need, effective implementation of these future technologies requires understanding of and adaptation to the external dynamics critical to the context of the team and mission. Examples of these external dynamics that must be monitored and incorporated into implementation of future technologies include environmental factors, sociocultural influences and shifts, changes in mission goals, and perhaps changing goals of higher-level and adjacent teams. When considering technologies to enhance the effectiveness of a human-agent team, a systems-based approach to affecting this performance must include adaptation from the environment during all phases of planning and action.
Synergizing cognitive, affective, and behavioral processes.
Another critical capability in regards to implementation of individualized, adaptive technologies to enhance human-agent teaming includes the capability to synergize between different types of processes and states within the team. Team states and processes are correlated, so understanding, measuring, and accounting for the dynamic interplay between different types of emergent group properties is important. For example, one can imagine times when providing more information (enhancing shared situational awareness) could result in delays in collective action, as human team members take time to process new information. Considering the dyadic pairs and subgroups of the team, and understanding critical elements of emergence (e.g., distribution of team situation awareness, or coordination among pairs versus entire group), approaches to optimally balance across subgroups may be beneficial as well. For example, when information and action demands are high, “cognitive” processing may be offloaded to a subgroup of intelligent agents for a period of time to allow for greater concentration on physical movement in a subgroup of human teammates. Thinking over slightly longer timescales, there might be times when teams should focus efforts on enhancing team affective states, such as cohesion, rather than jumping into specific cognitive and behavioral processes, such as decision making. Future technologies must consider tradeoffs amongst these states and processes in order to achieve optimal performance over the life-cycle of the team.
Coordination of Individualized, Adaptive Agents and Humans.
The final capability needed to implement individualized, adaptive technologies to enhance human agent teaming is the ability to dynamically adapt control and decision authority between humans, autonomous systems, and consensus protocols in the face of dynamic team states, goals, and environmental context. Of particular importance is the fact that external factors, such as military doctrine, rules of engagement, or political implications, may often dictate when intelligent agents are allowed to make certain decisions. In situations where the intelligent agent is not permitted to make decisions, the team must be capable of shifting the balance of control for that decision to the appropriate human team member with minimal disruption to the remaining team functions. Enabling these shifts in control could involve tradeoffs where the immediate team performance is sacrificed in order to satisfy top level constraints, such as military doctrine and enable long-term mission success. As complex coordination mechanisms develop, considering transformation of such top level constraints will be critical, as well, particularly with emerging advances in consensus decision making in heterogeneous teams.
Call for Comments:
- What additional capabilities are necessary to implement individualizable and adaptive team-enhancement technologies?
- Additional related comments
Comments
Another capability that would be useful for the human-agent teams is language and cultural resources and adaptability that goes beyond local population familiarization with seeing Americans with agents/tech. There’s two aspects to this: 1) the tech could supplement/enhance human language, cultural, and regional proficiencies, but also 2) may need to be adaptable for different countries, settings (urban/rural), cultures, and languages in terms of the usefulness of the tech to the team and threat/non-threateningness of the tech in theater.