Distributed Control Over Conditions and Transitions: How Different Participants Can Control Different Parts of the Same Process
Year: 2026
Type: article
Abstract: This conceptual paper introduces Distributed Control Over Conditions and Transitions, an account of how practical control over different parts of a process can lie with different participants. It maps each relevant condition or transition to whoever can admit, block, redirect, delay, execute, or revise it, identifying the available intervention and its scope. The account assesses a claim of control through a specified intervention capability and its observed or supported effect, separately from legitimate authority. A participant can hold authority without the means to act, or possess the means without authority. A possible control point does not establish that anyone has the information, timing, or capability to use it. Control may be partial, shared, or absent, and participation alone does not establish control. Where the evidence does not identify who could intervene, that allocation remains unknown. The discussion sets the mechanism beside Hutchins’s account of distributed cognition and examines two documented cases. During Apollo 11’s lunar descent in July 1969, the computer, ground controllers, and crew acted on different transitions, including responses to program alarms and selection of the landing site. The Formula One case draws on Mercedes’s description of its pit crew and Scarborough’s analysis of pit-stop systems to examine wheel changes, jack operation, completion signals, and release decisions. These case readings distinguish the distribution of information from the distribution of intervention capability, including situations in which authority to act does not supply the information needed to act reliably. Part of the Relational Conditions and Dynamics series within On Human Modeling, the paper examines who can change each transition, who can interrupt or override another’s control, and whether those affected can challenge its use. Potential applications in human–AI research include delegated workflows, human oversight, and multi-agent coordination, where formal authority and nominal control points must be distinguished from evidenced capabilities to intervene.
Author: Cary, Alexandra
Language: English
Subfield: Human Factors and Ergonomics
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7235379375
