What project AA4 buys
This project focuses on research that improves Soldier-system performance in future force environments by looking at key phenomena underlying Soldier integration with intelligent technologies and autonomous agents. This project researches optimal methods for information exchange between Soldiers and intelligent technologies including 1) human performance in automated, mixed-initiative (human control-machine control) environments; 2) visual scanning and target detection; 3) performance-related Soldier state changes; 4) integration across multiple sensory modalities; and 5) collaborative (team) and independent multi-task, multi-modal, multi-echelon Soldier-system performance - all cast against the influx of emerging intelligent technologies and autonomous systems. Technical solutions are being pursued in the areas of data generation and algorithm development in these emerging environments in order to update and improve our understanding of performance boundaries and requirements. These solutions include multi-disciplinary partnerships, metrics, simulation capabilities, and modeling tools for characterizing Soldier-system performance, and provide a shared conceptual and operational framework for militarily relevant research on critical aspects of human-agent teaming. In the area of translational neuroscience, research is carried out to examine leading edge methodologies and technologies to improve the measurement and classification of neural states and behavior in operationally-relevant environments; to examine the potential for application of neuroscience theories to autonomous systems to improve Soldier-system interactions; to model the relationship between brain structure and cognitive performance for understanding individual differences and injury; and to assess how neural pathways implicated in functional processing can be enhanced through dynamic system interface technologies for improving in-theatre performance and training. In the area of cybernetics, which is a scientific discipline that bridges the fields of control theory and communication theory for the study and modeling of behavior in complex systems, research is carried out to examine the complex human-system-environment relationships that define, constrain, and influence the interactions between Soldier and system.
Project AA4 funding, FY2025–FY2031
Prior years are actuals, the budget year is the request, and the outyears are the FYDP plan. Estimate types are colored and never summed into one figure. Projects carry the full five-year plan; the activities inside them stop at the budget year.
| Fiscal Year | Estimate Type | Amount ($M) |
|---|---|---|
| FY2025 | Actual | 19.6 |
| FY2026 | Enacted | 13.6 |
| FY2027 | Request | 10.7 |
| FY2028 | Outyear | 11.1 |
| FY2029 | Outyear | 11.3 |
| FY2030 | Outyear | 11.5 |
| FY2031 | Outyear | 11.7 |
10 accomplishments / planned programs
The R-2A exhibit — the only level of the budget that describes work that has not happened yet. Activities carry the prior, current and budget year only, no five-year plan. Coverage is partial across the corpus, so count activities, never total them.
FY2027 planned work Will expand neuro-inspired networks to improve performance of multiple coordinated systems performing spatial reasoning; investigate algorithms for multi-timescale mathematical relationships to understand performance of large groups in tasks that require adaptation; use neuro-inspired designs to improve, explain, and develop agentic architectures that expand system capabilities; use agentic systems to improve cognitive performance and overall decision making in dyads; investigate process efficiencies created through integrating foundational model-based agentic systems and multiple human interaction modalities for improving joint decision making.
FY2026 to FY2027 change This is not a new start effort. FY 2027 funding increase reflects the consolidation of other ongoing efforts within this project from Translational Neuroscience and Novel Forms of Joint Human-Intelligent Agent Decision Making to support the creation of Adaptive Soldier-Intelligent System Teaming for Enhanced Decision-Making (ASIST) in FY 2027. Funding increase reflects additional research in foundational model-based agentic systems.
FY2027 planned work Will conduct basic research to advance the science of assessment to develop whole-individual profiles that capture and incorporate within-person variability to better predict real-world success; explore team mechanisms to understand how much individuals drive overall teamwork by linking readiness states, behavioral processes during action, and team performance; further understanding of why and how leaders change and develop over time and develop interventions to prevent adverse leadership trajectories.
FY2026 to FY2027 change This is not a new start. FY 2027 funding reflects realignment within this project of Science of Measurement of Individuals and Collectives, Understanding Multilevel and Organizational Dynamics, and Formal and Informal Learning and Development to create Foundational Research in Personnel Science. Increase reflects economic adjustment.
FY2026 to FY2027 change Funding decrease reflects realignment to Adaptive Soldier-Intelligent System Teaming for Enhanced Decision-Making (ASIST) within this project.
FY2026 plans — current year Will expand neuro-inspired neuronal networks to perform better than deep networks on a spatial reasoning; create first of their kind topology informed neuronal networks to understand mixed formation performance; expand algorithms for multi-timescale mathematical relationships to include multiple humans and machines; develop simulations of spatial reasoning brain systems to expanding cognitive representations of spatial knowledge.
FY2025 accomplishments Will expand simulation models to generate novel abstract mapping relationships that go beyond what has been observed in mammalian brain activity; expand the capabilities of brain inspired spatial reasoning neuronal networks to include tasks that require flexibility and adaptation; explore the translation of breakthroughs in understanding multi-timescale and time-invariant mathematical relationships in the brain to represent human technology coordination.
FY2025 accomplishments Will investigate extending single agent human-guided machine learning techniques to multi-agent reinforcement learning settings; explore novel approaches to integrate generative language models and human feedback to speed up learning; create algorithms to incorporate ranking-based feedback from small groups of humans for the adaptation of multi-agent systems; explore ensemble-based techniques to improve uncertainty-based reasoning for human-guided machine learning.
FY2025 accomplishments Will explore initial ideas for the application of theory-driven approaches and methods to the analysis of very large datasets to identify the potential for generalizability of approaches across a wide range of human-centric data sets; assess computational/statistical models consistent with a theory-driven Big Data framework to establish initial empirical baselines.
FY2026 to FY2027 change Funding decrease reflects realignment to Adaptive Soldier-Intelligent System Teaming for Enhanced Decision-Making (ASIST) within this project.
FY2026 plans — current year Will investigate potential vulnerable vectors in information processing and subsequently decision making in human-intelligent agent collectives, where aggregation of informational elements is fundamental to the decision.
FY2025 accomplishments Will investigate distributed forms of information processing where joint human-intelligent agent decision making is performed while aggregating informational elements from many human and non-human sources.
FY2026 to FY2027 change Funding decrease reflects improvements garnered from achievements of previous years investigations and accomplishments.
FY2026 plans — current year Will investigate algorithms that leverage crowd-sourced human feedback to refine and improve multi-agent machine learning systems; investigate approaches to organize hybrid human-machine thinking based on artificial intelligence (AI) inferred human knowledge, skills, and abilities; investigate hybrid human-AI approaches to harness collective insights for dynamic adaptation in rapidly evolving contexts.
FY2025 accomplishments Will investigate large-scale, multi-human, multi-agent complex decisions that require many diverse and complex subtasks; perform experiments that target surveying a large decision space and rapidly settle on creative solutions in a hybrid human-technology complex scenario; investigate avenues of decision correction with rapidly evolving contextual and/or environmental changes across a hybrid human-technology team composed of many humans and intelligent entities.
FY2026 to FY2027 change Decrease reflects realignment of this effort within this project to support creation of Foundational Research in Personnel Science.
FY2026 plans — current year Will advance psychometric theory and methods to measure more complex types of individual and collective behavior and performance data within dynamic environments.
FY2025 accomplishments Will conduct research on novel approaches to assess multiple cognitive (e.g., ability to learn new information) and non-cognitive (e.g., personality) constructs; will conduct research to improve prediction of individual and team performance.
FY2026 to FY2027 change Decrease reflects realignment of this effort within this project to support creation of Foundational Research in Personnel Science.
FY2026 plans — current year Will conduct research on emerging trends in career decision making and its impact on organizational systems.
FY2025 accomplishments Will conduct research to improve scientific models of organizational functioning (e.g., team and multi-team performance and organizational effectiveness).
FY2026 to FY2027 change Decrease reflects realignment of this effort within this project to support creation of Foundational Research in Personnel Science.
FY2026 plans — current year Will develop and update theories and models of individual and collective learning to fuel individual, team, and organization learning outcomes.
FY2025 accomplishments Will conduct research to optimize learning and development across the lifecycle of a Soldier's career.