Maurice Lamb

I am a postdoctoral researcher in the Department of Psychology at the University of Cincinnati. My research focuses on developing cooperative dynamical algorithms to facilitate human-artificial agent interactions. The aim is to use nonlinear dynamical models of human behaviors and interactions as a basis for algorithms guiding the behavior of artificial agents.

Extended Abstracts


Learn More

Project Overviews

My research is at the intersection of Psychology, Philosophy, and Engineering.


Project Media and Links


Learn More



Model design and verification

Dynamical models of human individual and cooperative behaviors are well established. These models are the basis of novel algorithms that improve the robustness of artificial agents (virtual and robotic) engaged in cooperative tasks with humans. In turn, to the extent that these artificial agents successfully coordinate with humans, they serve as a form of model verification.

New experimental methods/tools

Dynamical models implemented in artificial agents introduces new possibilities for experimental design in complex systems research of human behavior and coordination. 

Towards more approachable machines

The Turing test is a hypothetical measure of how human-like an artificial agent’s intelligence is. Algorithms based on human coordination dynamics are tested to see how well they can pass a behavioral form of a simplified Turing test . Using complex systems analysis tools, quantitative verification of the algorithm’s successful coordination coordinate with humans can be provided.


Emergence in artificial agents

Dynamical systems approaches assume that many of the lower level details of a system are irrelevant to its higher level behaviors. This assumption is justified in the context of complex systems theory in physics and mathematics, but its ontological and epistemological implications need to be more clearly worked out and demonstrated in larger scale sciences. 

Structure of explanation in science

When individual systems become highly coordinated, their behavior can be modeled in terms of their coordination dynamics. The resulting model treats smaller than systems scale variations as mostly irrelevant to the coordination phenomena. As a result, it may be that so called higher level scientific explanations are not in some way less explanatory. In fact, measuring and identifying coordination phenomena may provide quantitative means for demonstrating that, in some cases, lower level explanations do not explain better and may explain less.


Evidence based algorithm design

Using ongoing research in psychology, evidence based algorithm design starts with models of human coordination and action.  This research is used to develop artificial systems that can robustly and seamlessly coordinate with human agents when needed. 

Robust Coordination

When humans coordinate, there doesn’t appear to be a clear well-defined script governing their behaviors. However, most artificial agent design begins by programming or training a machine to work with as many behavioral variations as possible. This approach can be time consuming and computationally burdensome. By using established and verified models of human coordination, artificial agents can achieve robust coordination behaviors without need for prior training or anticipation of every contingency. 

Flexible decision making

Decision making can be modeled by treating potential goals as points in a multi-dimensional attractor space. Artificial agents act within that space according to their current contexts and task constraints. These models can be implemented in artificial agents that must engage in task selecting and switching activities. 

Practical Applications

Technical and Manufacturing Applications

Dynamical approaches to human-artificial agent coordination has implications for manufacturing, remote medicine, and  long-range and extreme environment remote robotics contexts where humans and artificial agents must coordinate and switch between goals and control tasks with minimal communication. 

Diagnostic and Therapeutic

Dynamical models have already shown promise in diagnostic and therapeutic contexts. The current research further develops tools and methods for creating new implementations of these models in technology for these applications. 

Research design and public science

The current research builds on increases in cheap computational power along with increased accessibility of technology resources with the aim of  developing open source solutions that enable researchers to create novel and special purpose research tools and platforms.