A Structural Model of Attentional Effort Dynamics: Evidence from a Naturalistic Discrimination Task

Objective. To propose a model of how attentional effort varies over time in a vigilancetask and how this effort relates to subjectively inferred context. To propose an estimationmethodology and test the empirical validity of the proposed model in a naturalistic dataset.Background. Attentional effort in a task can vary based on how an individualsubjectively perceives the task context. However, both attention exertion and subjectivecontext perception are not directly observable. We present a methodology for estimating astructural model that explicitly incorporates subjective models of context perception andattention allocation policies. To our knowledge, this is the first methodology to estimate astructural model of attentional effort dynamics.Method. A Bayesian model of attentional allocation that integrates subjectiveperceptions of task-relevant context is developed. An estimation methodology based uponexpectation-maximization algorithm is proposed to uncover how the allocation ofattentional effort is adapted to subjectively perceived context.Results. The methodology is applied to a naturalistic dataset of Major League Baseballumpire decisions, revealing context perception (i.e. how umpires infer game situations) andattention allocation policy (i.e. how umpires adjust attentional effort). Model reveals thatumpires adjust attentional effort based on inferred game criticality and status bias.Conclusion. This work advances understanding of vigilance failure by providing astructural account for contextual inference determines attentional effort. The estimatedmodel closely track empirically observed decision accuracy patterns in a naturalisticdataset.Application. The proposed model enables counterfactual predictions, allowingexploration of hypothetical interventions to improve decision accuracy in environmentsthat require sustained attention.