The Need for Machine Fitness Assessment: Enabling Joint Human-Machine Performance in Consumer Health Technologies

Abstract

We introduce the concept of machine fitness assessment, which is the process of correctly determining the degree of fit between a machine’s inferences on a specific world and the world itself. We describe its importance in complex, high-stakes worlds, including healthcare, and how it will be critically important to realize the potential of consumer health technologies that promise institutional-quality health diagnosis and planning in decidedly non-institutional settings (e.g., our homes, offices, or anywhere else).

Towards Joint Activity Design Heuristics: Essentials for Human-Machine Teaming

“…Therefore, machines will be unreliable at communicating their own limits and need help from people. At minimum, this requires some way for people to simultaneously understand the world, how the machine is "seeing" the world through its model, and how well the machine's view of the world is aligned to the world itself (Rayo et al, 2020).…”

Joint Activity Testing: Towards a Multi-Dimensional, High-Resolution Evaluation Method for Human-Machine Teaming

“…When estimates from baseline instead conflict with real-time measurements, this violation of expectations may function as the most informative cue of impending surprise (see "anomaly response" in Woods & Hollnagel, 2006). This two-pronged approach can serve as both an evaluation of the system in real-time and an evaluation of the evaluation process, similar to how machine fitness assessment functions in HMTs (Rayo et al, 2020). Anomalous mismatches between measures from the lab and operations point to regions of performance where the system is not well-understood and needs further investigation.…”

Where the Cognitive Engineering Rubber Meets the Systems Development Road

“…This pattern is advantageous both for opaque (Rayo, in press) as well as observable (Morey, 2021) automation. They all rely on the assumption that operators must be able to assess the fitness of their machine counterparts as those machines carry out their tasks, and respond accordingly (Rayo, 2020). JAD creates an additional set of requirements in the analysis phase and design seeds in the design phase of SD related to attributes that AIinfused software solutions must have in order to contribute to Joint Human-machine activities.…”