Toward a Logic of Generalization about Visualization as a Decision Aid

TL;DR

This paper develops a logical framework for understanding how visualization research generalizes across different decision-making contexts, highlighting the importance of utility and decision theory in this process.

Contribution

It introduces a decision-theoretic approach to analyze the generalization of visualization effectiveness across varied decision scenarios.

Findings

Utility is a key but under-explored concept in visualization decision-making.

Decision theory provides a useful lens for understanding context variation.

The framework helps identify when visualization tools are likely to generalize effectively.

Abstract

Visualization as a discipline often grapples with generalization by reasoning about how study results on the efficacy of a tool in one context might apply to another context. This work offers an account of the logic of generalization in visualization research and argues that it struggles in particular with applications of visualization as a decision aid. We use decision theory to define the dimensions on which decision problems can vary, and we present an analysis of heterogeneity in scenarios where visualization supports decision-making. Our findings identify utility as a focal and under-examined concept in visualization research on decision-making, demonstrating how the visualization community's logic of generalization might benefit from using decision theory as a lens for understanding context variation.