CoChat
Sign Up
Cited by 33Human-Automation Interaction and SafetyTeam Dynamics and PerformanceMilitary Strategy and Technology

Toward a science of human–AI teaming for decision making: a complementarity framework

Open access
Gonzalez, Cleotilde, Donahue, Kate, Goldstein, Daniel G, Heidari, Hoda, Jalali, Mohammad S, Schelble, Beau, Singh, Aarti, Woolley, Anita Williams·2026·PNAS Nexus
DOI 10.1093/pnasnexus/pgag030PMC PMC12983458
Open this paper in CoChat — free account

Abstract

As artificial intelligence (AI) becomes embedded in critical decisions involving health, safety, finance, and governance, the key challenge is no longer whether humans and AI will collaborate, but rather how to structure this collaboration to achieve true complementarity. Human-AI complementarity refers to the conditions under which human-AI teams outperform either humans alone or AI systems alone. This paper advances the science of human-AI teaming for decision making by integrating insights from cognitive science, AI, human factors, organizational behavior, and ethics. We propose a framework grounded in collective intelligence and anchored in the foundational cognitive processes-reasoning, memory, and attention-to understand and engineer effective human-AI teams. We examine the sociotechnical factors that shape team effectiveness, including team composition, trust calibration, shared mental models, training, and task structure. We then outline design principles for achieving complementarity: defining goals and constraints, partitioning roles, orchestrating attention and interrogation, building knowledge infrastructures, and establishing continuous training and evaluation. We conclude with theoretical, practical, and policy implications, emphasizing alignment with human values, accountability, and equity. Together, these insights offer a roadmap for building human-AI teams that are not only high-performing and adaptive, but also transparent, trustworthy, and fundamentally human-centered.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/pnasnexus/pgag030

Cites

92 of 92 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • AI should be understood not as a substitute for human judgment, but as a cognitive amplifier that supports human decision-making under conditions of uncertainty, stress, time pressure, and information overload.

    “AI should be understood not as a substitute for human judgment, but as a cognitive amplifier that supports human decision-making under conditions of uncertainty, stress, time pressure, and information overload (Palermo, 2020a, 2020b), consistent with human-centered AI and human-AI teaming approaches (Shneiderman, 2020; Gonzalez et al., 2026; Kargarnovin et al., 2026).”

    From Enhancement of modern defense capabilities by convergence of cognitive neuroscience and artificial intelligence (2026) · cites this paper as Gonzalez et al., 2026

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 19

  1. [5]Edmondson AC. 2013. Teaming is a verb. Reflect SoL J. 13(1): 43– 47.
  2. [17]Du Y, Rajivan P, Gonzalez C. Large language models for collective problem-solving: insights into group consensus decision-making. In: Samuelson LK, Frank SL, Toneva M, Mackey A, Hazeltine E, editors. Proceedings of the Annual Meeting of the Cognitive Science Society. Cognitive Science Society, 2024. p. 3011–3018.
  3. [33]Wilson K, Caliskan A. 2024. Gender, race, and intersectional bias in resume screening via language model retrieval. Proc AAAI ACM Conf AI Ethics Soc. 7(1): 1578– 1590.
  4. [40]Bansal G, et al 2019. Beyond accuracy: the role of mental models in human-AI team performance. Proc AAAI Conf Hum Comput Crowdsourc. 7: 2– 11.
  5. [41]D’Amour A, et al 2022. Underspecification presents challenges for credibility in modern machine learning. J Mach Learn Res. 23(226): 1– 61.
  6. [55]Hannebauer C, Patalas M, Stünkel S, Gruhn V. Automatically recommending code reviewers based on their expertise: an empirical comparison. ASE’16: ACM/IEEE International Conference on Automated Software Engineering, Singapore; September 3–7, 2016.
  7. [71]Zercher D, Jussupow E, Heinzl A. When AI joins the team: a literature review on intragroup processes and their effect on team performance in team-AI collaboration. Paper presented at: ECIS 2023: European Conference on Information Systems; Kristiansand, Norway; June 11–16, 2023.
  8. [73]Klein GA. Sources of power: how people make decisions. MIT Press, 2017.
  9. [75]Seeber I, et al 2020. Machines as teammates: a research agenda on AI in team collaboration. Inf Manage. 57(2): 103174.
  10. [76]Malone TW, Bernstein M. Handbook of collective intelligence. MIT Press, 2022.
  11. [81]Schelble BG, Flathmann C, McNeese NJ, Freeman G, Mallick R. 2022. Let's think together! Assessing shared mental models, performance, and trust in human-agent teams. Proc ACM Hum Comput Interact. 6: 1– 29.37360538
  12. [89]Greenwood S, Levy K, Barocas S, Kleinberg J, Heidari H. Designing algorithmic delegates. Poster presented at: NeurIPS 2024: Workshop on Behavioral Machine Learning; Vancouver, Canada; December 14, 2024.
  13. [90]Peng K, Garg N, Kleinberg J. 2025. A no free lunch theorem for human-AI collaboration. Proc AAAI Conf Artif Intell. 39(13): 14369– 14376.
  14. [98]Okolo CT, Kamath S, Dell N, Vashistha A. “It cannot do all of my work”: community health worker perceptions of AI-enabled Mobile health applications in rural India. Paper presented at: CHI ‘21: CHI Conference on Human Factors in Computing Systems; Yokohama, Japan; May 8–13, 2021.
  15. [99]Chen V, Liao QV, Wortman Vaughan J, Bansal G. 2023. Understanding the role of human intuition on reliance in human-AI decision-making with explanations. Proc ACM Hum Comput Interact. 7(CSCW2): 1– 32.
  16. [100]Yin M, Wortman Vaughan J, Wallach H. Understanding the effect of accuracy on trust in machine learning models. Paper presented at: CHI ‘19: CHI Conference on Human Factors in Computing Systems; Glasgow, Scotland; May 4–9, 2019.
  17. [113]Schmager S, Pappas I, Vassilakopoulou P. Defining human-centered AI: a comprehensive review of HCAI literature. In: MCIS 2023 Proceedings. Midwest Instruction and Computing Symposium. University of Northern Iowa, Cedar Falls, Iowa, 2023.
  18. [115]Bansal G, et al Does the whole exceed its parts? The effect of AI explanations on complementary team performance. Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems. CHI ‘21: CHI Conference on Human Factors in Computing Systems, Yokohama Japan; 2021.
  19. [118]Rastogi C, Leqi L, Holstein K, Heidari H. 2023. A taxonomy of human and ML strengths in decision-making to investigate human-ML complementarity. Proc AAAI Conf Hum Comput Crowdsourc. 11(1): 127– 139.

Cited by

Sorted by citations

Related

Authors
Gonzalez, C., Donahue, K., Goldstein, D. G., Heidari, H., Jalali, M. S., Schelble, B., Singh, A., & Woolley, A. W.
Affiliations
  • Social and Decision Sciences Department, Carnegie Mellon University , 5000 Forbes Avenue, Pittsburgh, PA 15213 ,
  • University of Illinois Urbana-Champaign (US)
  • Microsoft (United States) (US)
  • Carnegie Mellon University (US)
  • Harvard University (US)
  • University of Tennessee at Knoxville (US)
ORCID
Venue
PNAS Nexus
Publisher
Oxford University Press (OUP)
Published
2026-02-19
Volume
5
Issue
3
Pages
pgag030-pgag030
ISSN
2752-6542 (ISSN-L)
DOI
10.1093/pnasnexus/pgag030
PMC
PMC12983458
OpenAlex
W7130511309
Cited by
33
References
89
Citations by year
2026: 33
FWCI
266.34
Type
journal article
Language
en
Open access
Licence
Locations
  • PNAS Nexus (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Funders
  • AI Research Institutes Program
  • National Science Foundation
  • AI Institute for Societal Decision Making (2229881)
Topics
  • Social Sciences › Psychology › Social Psychology › Human-Automation Interaction and Safety
  • Social Sciences › Psychology › Social Psychology › Team Dynamics and Performance
  • Physical Sciences › Engineering › Control and Systems Engineering › Military Strategy and Technology
Keywords
human-AI teaming, human-AI complementarity, decision-making, collective intelligence, team effectiveness, trust calibration, shared mental models, team composition, task structure, knowledge infrastructure, AI ethics, accountability, equity, human-centered AI

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review