Confirmation Bias in Adversarial Reasoning: Insights from the ASCEND Framework for Adaptive Cyber Defense
Authors:
- Grit DenkerGDGrit Denker
- Laura TinnelLTLaura Tinnel
- Paul BurkanderPBPaul Burkander
- Joshua LawsonJLJoshua Lawson
- Brodie MatherBMBrodie Mather
- Virginia VaughanVVVirginia Vaughan
- Joshua PattersonJPJoshua Patterson
- Michael McAnallyMMMichael McAnally
- Alex FafardAFAlex Fafard
- Nathan BuntingNBNathan Bunting
- Gerald MatthewsGMGerald Matthews
- Richard D. RobertsRRRichard D. Roberts
Open Access
- 15 September 2026
- journal article
- research article
- Published by Springer Nature in Computational Brain & Behavior
- p. 1-21
Abstract
Cyberspace represents a dynamic cognitive battlespace where adversaries continually adapt to evade defenses. Traditional cybersecurity approaches emphasize technological countermeasures while overlooking the cognitive and behavioral processes shaping attacker decision-making. The Adaptive Security through Cognitive Exploitation for Defense (ASCEND) project advances a cognitive cybersecurity framework that identifies and strategically leverages cognitive vulnerabilities (CogVulns)—psychological biases that influence adversarial behavior—to enhance defensive effectiveness. ASCEND integrates behavioral science and cybersecurity research through two experimental paradigms. A Multivariate, Multi-Country Heuristic (MCH) study assessed CogVulns, personality, and risk-taking across 2,121 participants from 91 countries. Complementary Capture-the-Flag (CTF) experiments simulated adversarial decision-making in realistic environments using Cognitive Vulnerability Triggers engineered to engage targeted biases. Behavioral, psychometric, and cyber-interaction data were jointly analyzed to examine cognitive and individual-difference factors affecting attacker susceptibility. Psychometric analyses revealed variability in cognitive vulnerability profiles linked to individual traits. In CTF experiments, targeted triggers consistently had a negative impact on adversarial decisions. Identified cyber data surrogates serve as CogVuln Sensors, providing real-time proxies to infer susceptibility to CogVulns. ASCEND demonstrates the feasibility of embedding cognitive principles into cyber defense. By exploiting predictable human biases, cognitive defenses can actively shape adversarial behavior. These findings mark a step toward adaptive, psychologically informed security architectures that transform human vulnerabilities from liabilities into strategic assets.Keywords
Funding Information
- SRI International
This publication has 51 references indexed in Scilit:
- Human factors in cybersecurity: an interdisciplinary review and framework proposalInternational Journal of Information Security, 2025
- Proceedings of 4th International Conference on Frontiers in Computing and Systems2024
- Cybersecurity knowledge and skills taught in capture the flag challengesComputers & Security, 2021
- Knowledge as a formative construct: A good alpha is not always betterNew Ideas in Psychology, 2021
- Breaking Bad in Cyberspace: Understanding why and how Black Hat Hackers Manage their Nerves to Commit their Virtual CrimesInformation Systems Frontiers, 2019
- Confirmation Bias through Selective Overweighting of Choice-Consistent EvidenceCurrent Biology, 2018
- Development and validation of the General Risk Propensity Scale (GRiPS)Journal of Behavioral Decision Making, 2018
- No one is immune to contextual bias—Not even forensic pathologists.Journal of Applied Research in Memory and Cognition, 2018
- Short and extra-short forms of the Big Five Inventory–2: The BFI-2-S and BFI-2-XSJournal of Research in Personality, 2017
- Anxiety and selective attention to threat in tactical decision-makingPersonality and Individual Differences, 2011