Senior Research Scientist
Obelisk lab, Astera Institute
2022 - Present
Research Goals
My long-term interest is in contributing to research on how the brain undertakes metacognition and self-reflection. Research on mindfulness, cognitive behavioral therapy, and other self-reflective states of consciousness has shown that metacognition has a large influence on mental health and learning. It should be possible to simulate metacognition and self-reflection from first-principles in artificial neural networks, and doing so would allow us to better understand both the higher cognitive capabilities of humans as well as mental illnesses.
Current Research Focus
At Astera, I started a project called Fluid to develop a unique and complex AI agent architecture featuring short-term memory and reasoning via cellular automata, hierarchically clustered episodic memory, meta-learning based exploration policy, internal feedback control, and dynamic, selective attention to inputs. My research spans neuroscience-inspired algorithms, reservoir computing, cellular automata for neural algorithmic reasoning, emergent multi-agent communication, and hierarchical memory structures. So far, Fluid has served as an integrative framework for many useful problem areas and an experimental tool for cognitive neuroscience, allowing simulation of complex cognitive mechanisms in problem solving.
Links
OSF for supplemental materials of published projects
Recent Preprints (2024-2025)
Miconi, T., McKee, K., Zheng, Y., & McCaleb, J. (2025). Thinking agents for zero-shot generalization to qualitatively novel tasks. arXiv preprint arXiv:2503.19815.
Zheng, Y., Wolf, N., Ranganath, C., O’Reilly, R. C., & McKee, K. L. (2025). Flexible prefrontal control over hippocampal episodic memory for goal-directed generalization. arXiv preprint arXiv:2503.02303.
McKee, K. L. (2025). Meta-Learning to Explore via Memory Density Feedback. arXiv preprint arXiv:2503.02831.
McKee, K. (2025). A Method of Selective Attention for Reservoir Based Agents. arXiv preprint arXiv:2502.21229.
McKee, K. (2024). Reservoir computing for fast, simplified reinforcement learning on memory tasks. arXiv preprint arXiv:2412.13093.
McKee, K., Crandell, I., Chaudhuri, R., & O’Reilly, R. (2022). Adaptive Synaptic Failure Enables Sampling from Posterior Predictive Distributions in the Brain. arXiv preprint arXiv:2210.01691.
McKee, K. L., Crandell, I. C., Chaudhuri, R., & O’Reilly, R. C. (2021). Locally learned synaptic dropout for complete bayesian inference. arXiv preprint arXiv:2111.09780.
Peer-Reviewed Publications
McKee, K.L. (2021). Hierarchical Biometrical Genetic Analysis of Longitudinal Dynamics. Behavior Genetics.
https://doi.org/10.1007/s10519-021-10060-0
Kaplan, B. A., Franck, C. T., McKee, K. L., Gilroy, S. P., Koffarnus, M. N. (2021). Applying Mixed-Effects Modeling to Behavioral Economic Demand: An Introduction. Perspectives on Behavior Science.
Hunter, M. D., McKee, K. L., Turkheimer, E. (2021). Simulated Nonlinear Genetic and Environmental Dynamics of Complex Traits. Development and Psychopathology.
McKee, K. L., Crandell, I. C., Hanlon, A. L. (2020). US County-Level Social Distancing and Policy Impact: A Dynamical Systems Model. Journal of Medical Internet Research.
https://doi.org/10.2196/23902
McKee, K. L., Russell, M., Mennis, J., Mason, M., & Neale, M. C. (2019). Emotion Regulation Dynamics Predict Substance Use in High-Risk Adolescents. Addictive Behaviors.
https://doi.org/10.1016/j.addbeh.2020.106374
McKee, K. L., & Neale, M. C. (2019). Direct estimation of the parameters of a delayed, intermittent activation feedback model of postural sway during quiet standing. PLoS one, 14(9).
https://doi.org/10.1371/journal.pone.0222664
McKee, K. L., Hunter, M. D., & Neale, M. C. (2019). A Method of Correcting Estimation Failure in Latent Differential Equations with Comparisons to Kalman Filtering. Multivariate Behavioral Research, 1-20.
https://doi.org/10.1080/00273171.2019.1642730
McKee, K. L., Rappaport, L. M., Boker, S. M., Moskowitz, D. S., & Neale, M. C. (2018). Adaptive Equilibrium Regulation: Modeling Individual Dynamics on Multiple Timescales. Structural Equation Modeling: A Multidisciplinary Journal, 1-18.
https://doi.org/10.1080/10705511.2018.1442224
Moscati, A., Verhulst, B., McKee, K. L., Silberg, J., & Eaves, L. (2018). Cross-Lagged Analysis of Interplay Between Differential Traits in Sibling Pairs: Validation and Application to Parenting Behavior and ADHD Symptomatology. Behavior Genetics, 48(1), 22-33.