Second-generation UMTRI coding scheme for classifying driver tasks in distraction studies and application to the ACAS FOT video clips.

Yee, Serge; Green, Paul; Nguyen, Lan T.; Schweitzer, Jason; Oberholtzer, Jessica · 2006 · ROSA P / University of Michigan. Transportation Research Institute

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This report details the development and application of the second-generation UMTRI coding scheme, designed to classify driver tasks and subtasks in distraction studies. The research was motivated by the need for a standardized, objective method to quantify secondary tasks that overload drivers, supporting the broader SAVE-IT project’s goal of developing a workload manager for in-vehicle systems. The authors identified limitations in prior schemes, such as the Stutts et al. and first-generation UMTRI schemes, which suffered from subjective classifications (e.g., "high involvement"), ambiguous task endpoints, and missing codes for common activities like chewing gum. To address these issues, the new scheme was designed to be descriptive, consistent with human factors engineering practices, and practical for use with video data. The methodology involved creating a comprehensive coding structure comprising 12 primary task categories (including using a cell phone, eating/drinking, smoking, grooming, and conversing) plus a code for drowsiness. Each task was broken down into 3 to 17 specific subtasks, such as "reach and get phone" or "groom using tool," with clearly defined start and end points to eliminate subjectivity. The scheme also incorporated codes for driver state (gaze direction, head position, hand location) and driving context (weather, road surface). This scheme was applied to video clips from the Advanced Collision Avoidance System (ACAS) Field Operational Test, which included data from over 100 drivers. The coding process occurred in two passes: the first pass coded 2,914 four-second clips for task presence, drowsiness, and environmental conditions; the second pass involved frame-by-frame analysis of 819 clips (403 distracted, 416 nondistracted) to record specific subtasks, gaze, and head direction. The findings demonstrate that the second-generation scheme successfully provides higher resolution and objectivity than previous methods. By specifying distinct endpoints for subtasks and removing subjective involvement ratings, the scheme allows for precise measurement of task duration and frequency. The application to the ACAS dataset confirmed the scheme's utility in capturing detailed behavioral data, including activities previously uncoded, such as chewing gum and tobacco. The structured breakdown of tasks into preparation, execution, and completion phases enabled researchers to link specific subtasks to visual, auditory, cognitive, and psychomotor demands. The significance of this work lies in its contribution to the field of human factors and automotive safety. The refined coding scheme provides a robust tool for analyzing naturalistic driving data, enabling more accurate assessments of how secondary tasks interfere with driving performance. This data is critical for the development of workload managers that can dynamically adjust in-vehicle information presentation based on driver demand. By establishing a replicable and extensible standard for task classification, the report supports future research into driver distraction and the design of safer, more usable vehicle interfaces.

Key finding

The second-generation UMTRI coding scheme provides a more objective and granular classification of driver activities by defining distinct endpoints for subtasks and including previously omitted activities like chewing gum.

Methodology

dataset

Sample size: 2914

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• distraction detection algorithms
• temporal
• visual
• visual manual
• manual
• external distraction

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Methodological Resource: measurement protocol, tool software
• Theoretical Contribution: conceptual framework