Publication

A conceptual framework and taxonomy for understanding and categorizing driver inattention

This documents reports on the results of the Inattention Taxonomy project, which was carried out by the Driver Distraction & Human Machine Interaction (DD & HMI) Working Group, under the framework of the United States and European Union Bilateral Intelligent Transportation Systems Technical Task Force (US-EU Bilateral ITS TF), with the main objective being to define a conceptual framework and taxonomy of driver inattention. While driver inattention is known as one of the main factors contributing to road crashes, commonly agreed definitions of key concepts are still lacking. This makes it difficult to estimate the true magnitude of road-safety problems associated with driver inattention and makes comparison of results across studies problematic. Thus a common taxonomy of inattention-related concepts such as driver distraction is strongly needed, both from scientific and applied standpoints. The taxonomy developed in the project is mainly intended to be applied in the context of driver behaviour and incident/accident analysis, the design and interpretation of experimental studies and in the design and evaluation of vehicle systems. The term taxonomy here refers to the definition of key theoretical concepts and their mutual relationships. This should be distinguished from a coding scheme, which refers to detailed operational definitions of phenomena observable in the available data. A key starting point for the project was that, in order to create a taxonomy of driver inattention, one first needs to be clear about what is meant by driver attention. To this end, a conceptual framework for driver attention was formulated in terms of a set of key principles. This framework proposes an action-oriented view of attention, where driver attention is generally defined as the allocation of resources to a set of (driving- or non-driving related) activities. The distribution of resources to activities depends on two main aspects. The first relates to activation (how much of one or more resource is allocated) while the second relates to selectivity (how resources are distributed between activities). Driver inattention was then conceptualised in terms of mismatches between the driver’s current resource allocation and that demanded by activities critical for safe driving, rather than in terms of attentional failures of the driver. This systemic perspective helps to circumvent conceptual problems associated with the notion of driver error related to hindsight bias and the attribution of blame. Based on this conceptual framework, a general taxonomy of driver inattention was developed. Driver inattention was broadly divided into two general categories: (1) insufficient attention and (2) misdirected attention, relating to the activation and selective aspects of attention respectively. For each of these categories, a set of sub-processes giving rising to them was defined. The report ends with a discussion of some key implications of the proposed conceptual framework and inattention taxonomy, and how the taxonomy can be used for its intended applications.

Author(s)
Johan Engström, Christopher A. Monk, Richard J. Hanowski, William J. Horrey, John D. Lee, Daniel V. McGehee, Michael Regan, Alan Stevens, Eric Traube, Marko Tuukkanen, Trent Victor, C. Y. David Yang
Research area
Safety performance evaluation
Publication type
Project report
Project
Global Inattention Taxonomy Definition (C20)
Year of publication
2013