Information Dissemination and Human Factors Databases

At the Cognitive Systems Lab, we are always looking to share knowledge with the broader Human Factors community. As part of this effort, we are building research article databases on special human factors topics through Mendeley, a free research management tool. These databases are constantly evolving, and we hope to establish a symbiotic association with interested researchers in building the resources.
Check out some of the article collections -
Driver Interaction with Vehicle Automation
Trust in Technology

Human-Automation Cooperation

Sponsored by National Science Foundation Graduate Research Fellowship Program

Student Researcher: Erin Chiou

microworld interruption

Increasingly capable technology fundamentally changes human-machine relationships. Rather than reliance or compliance with more or less reliable automation, we investigate cooperation with more or less cooperative autonomous agents. A joint-task microworld scenario was developed to explore the effects of different levels of agent cooperation on participant cooperation. This approach complements the supervisory control perspective by considering interdependence and cooperation to support system resilience.
Related Papers:
1. Chiou, E.K., & Lee, J.D. (2016). Human-automation cooperation to support resilience: A microworld study. Human Factors: The Journal of the Human Factors and Ergonomics Society (submitted).
2. Chiou, E. K., & Lee, J. D. (2015). Beyond reliance and compliance: Human-automation coordination and cooperation. In 59th Annual Meeting of the Human Factors and Ergonomics Society (Vol. 59, pp. 195–199).

Developing Driver Distraction Interface Evaluation Tool

Sponsored by NHTSA

Student Researcher: Joonbum Lee
Collaborators: Prof. Dario Salvucci, Drexel University


The main purpose of this project is to produce a tool that predicts distraction potential by using DVI parameters and characteristics. Based on ACT-R that simulates human cognition, we develop a computational model to prototype and evaluate DVI. The proposed model is expected to have an ability to examine the effect of coordinating salient visual features and drivers’ expectations, and in using both together, generate robust predictions of performance.
Related Papers:
1. Evaluating the Distraction Potential of Connected Vehicles.
2. A Saliency-based Search Model: Application of the Saliency Map for Driver-vehicle Interfaces.

Driive Project

Student Researcher: Tony McDonald

Above figure: A confusion matrix displaying the performance of a steering based algorithm for drowsiness detection

A substantial amount of research has been dedicated to the development of algorithms for detecting drowsy, drunk, and distracted drivers individually. The Driive project seeks to integrate these algorithms and design new algorithms capable of identifying and differentiating between the three major types of impairment. The project is a joint work with the University of Iowa and the National Advanced Driving Simulator, and utilizes a combination of simulator and naturalistic driving data.

Driver Response and Crash Avoidance Behavior Modeling

Sponsored by Toyota CSRC North America

Student Researcher: Vindhya Venkatraman
Collaborators: National Advanced Driving Simulator, University of Iowa

As use of vehicle automation increases, the phenomenon of driver distraction due to engagement with non-driving tasks becomes prevalent. Not surprisingly, the onus of recovery from safety-critical situations falls on the driver-assistance systems in the vehicle. While such systems of the past could get away with crude models of driver responses, automation designed to help mitigate crashes in the future will need to implement dynamically informed models to situations of varying criticality. In this study, the team works towards understanding perception and behavioral responses of US drivers to safety critical situations, such as lane departures. The conceptual models built through this understanding would be developed into working computational models for advanced driving assistance systems (ADAS).

Impact of Health Information Technology (HIT) on Informal Caregiver Burden

Student Researchers: Rashmi P Payyanadan, Brian Sullivan (CS Department), & Liang Wang (CS Department)
PI: Human-Computer Interaction Lab, Prof. Bilge Mutlu

Use and implementation of health information technologies (HIT) have been slow due to a myriad of issues such as: a) perception of HIT devices by older adults to be intrusive and invasive, b) use of HIT devices are foreign to both the older adult generations as well as the caregivers and thus reduces use and acceptance rates of such technologies, and c) HIT devices require constant monitoring and thus can be time consuming for the caregivers and doctors, which can further increase their burden. Literature review has shown a serious gap in addressing how HIT devices might affect caregiver burden. Hence the purpose of this project is to investigate the impact of a HIT device on informal caregiver burden and to address whether HIT devices would reduce some of the key challenges informal caregiver face such as burden, stress, isolation, loneliness and isolation.

In-Vehicle Voice Control Interface Performance Evaluation

Sponsored by NHTSA

Student Researchers: Vindhya Venkatraman and Joonbum Lee
Collaborators: Human Factors and Statistical Modeling Lab, Prof. Linda Boyle

Driver distraction with use of in-vehicle devices is a rapidly increasing problem in today's driving context. Nomadic devices such as mobile phones, embedded systems such as radios and text messaging services, driver information systems such as GPSs are commonly used within the vehicle. The automobile of today also acts as a mobile communication center for its occupants, and inefficient design of such technology could lead to more ill effects than benefits. The aim of this study is inform design guidelines through development of assessment tests and performance thresholds for voice-controlled systems (systems that operate through user's speech commands) within the vehicle.

Measuring Driver Attention Using Electroencephalography (EEG)

Student Researcher: Rashmi Payyanadan
Collaborators: Human-Computer Interaction Lab, Prof. Bilge Mutlu

According to analyses of traffic accident data, 90% of traffic accidents are caused by driving failures due to impaired driver mental state. With the development of optimal Advanced Driver Assistance Systems (ADAS), researchers require a profound understanding of the causes and factors that lead to driver inattention such as fatigue, mental overload, distraction etc. The aim of this study is to use the brain computer interface headset from Neurosky to monitor driver attention levels, and thereby determine ways to provide visual or auditory feedback to drivers during periods of inattention.

Naturalistic Driving and Obstructive Sleep Apnea

Student Researcher: Tony McDonald
Collaborators: Prof. Matthew Rizzo, University of Iowa

Sleep Apnea
Above figure: A demonstration of a technique to reduce continuous driving data to symbols

This project explores the complex relationship between obstructive sleep apnea and driving through the use of innovative data reduction and analysis methods. A primary focus of the project is the concept of developing algorithms based on naturalistic driving data that are capable of detecting sleep quality and sleep intervention use amongst drivers with sleep apnea.

Privacy Management in Consumer Health Informatics

Student Researcher: Dee Miller
Collaborators: Human-Computer Interaction Lab, Prof. Bilge Mutlu

This project explores the role of context and privacy management in current artifact usage for logging and tracking personal health information among older adults. The purpose of the project is to understand current artifact usage and identify privacy management strategies and rules that older adults current use in logging, tracking, and sharing personal health information to inform the design of future consumer health information technology applications used to support care coordination activities.

Promoting Strategic Change in Poultry Growing Through Checklist Design

Student Researcher: Pat Ferrara

In industrially developing regions, globalization of industry has brought about increased skill and management requirements. The strategies that workers develop over time govern their adoption of technology and artifacts. A CSL study of checklist design and function for promoting strategy change in poultry growing examines their use and effects on diagnostic and assimilatory skills and chicken house management.

Social Influence and its Effect on Teenage Driving Behavior

Student Researcher: Shannon Roberts

Unsafe driving behavior, like texting and driving, combined with teenagers' inexperience attributes to the fact that traffic related crashes are the number one cause of death for teenagers. A solution to curb the number of teenage traffic fatalities combines driving feedback with social influence to promote safe driving behavior. As such, a new study at the CSL has been designed with the goal of examining the effect of social influence on distracted driving behavior within a teenage social network and examining the effect of a feedback system on teenage driving behavior.
Related Paper:
1. Using Agent-Based Modeling to Predict the Diffusion of Safe Teenage Driving Behavior Through an Online Social Network.

Understanding the Mobility and Independence Needs of Older Drivers

Student Researchers: Rashmi Payyanadan and Mahtab Ghazizadeh
Collaborators: Center for Health Enhancement Systems Studies, Prof. David Gustafson

Mobility is a key indicator of independence in American culture, and driving is one of the primary indicators of continued independence for people over 65. CSL and CHESS are collaborating on a 5 year project with the purpose of creating innovate technology and mobility solutions that support older drivers. Resulting findings and/or technologies will be integrated into Elder Tree (ECHESS)—a study-wide resource that will collect relevant data and transmit advice to the elder drivers.

Completed Projects

Research Environment for Supervising Control of Heterogeneous Unmanned Vehicles

Student Researchers: Mai Lee Chang and Erin Chiou
Collaborators: MIT Humans and Automation Lab


Appropriate reliance on technology represents a central challenge in managing unmanned vehicle (UV) teams. Managing a team of UVs requires complex automation that integrates information from multiple platforms. This research assesses the impact of automation transparency, capability, and resource on appropriate trust and reliance to enhance human-automation performance.

Text Reading and Text Input Assessment Using Visual Occlusion

Sponsored by NHTSA

Student Researchers: Mahtab Ghazizadeh and Vindhya Venkatraman
Collaborators: Human Factors and Statistical Modeling Lab, Prof. Linda Boyle


The goal of this study is to estimate the time drivers spend completing text reading and text entry tasks of varying difficulty levels on a touchscreen display. The visual occlusion technique is used to mimic the timesharing between driving and performing the text manipulation task. The results of this study and a related study conducted at the University of Washington (using a driving simulator) will help determine the amount of text that can be read or entered while driving, without hindering road safety.
Related Papers:
1. Assessing text reading and text entry while driving using the visual occlusion technique
2. Factors affecting glance behavior when interacting with in-vehicle devices: Implications from a simulator study
3. Text Readability and Driver's Reading Time: Insights from the Visual Occlusion Method. (Best Paper Award: HFES 57th Annual Meeting, San Diego).