Abstracts for Poster Presentations (On display from 1:00 to 2:30 pm)

Muhammad F. Walji (UT Health Science Center at Houston)
Many informatics interventions are aimed at improving health outcomes by influencing clinician or patient behavior through tools such as decision support, computerized physician order entry, patient reminder tools amongst others. These interventions share the need to appropriately deliver a message to a user to influence decisions making. For example computerized drug allergy or interaction alerts are often presented to physicians who prescribe and pharmacists who dispense medications. However, these alerts are frequently overridden or ignored, and in some cases lead to adverse events. Clinical practice guideline reminders or suggestions are often presented to physicians at the point of writing orders. Patients are also provided with timely prompts alerting them when its time to take their medications or sign up for a recommended screenings. Many of these initiates have reported varying degrees of success and have attributed non adherence to usability issues, irrelevant messages, or justifiable reasons to override or ignore.

However, there is little understanding on how the message itself can play a role in influencing a decision. To date little research exists to explain how users (patients or clinicians) process such messages. A better understanding of message processing, and the factors that effect acceptance of such requests may help improve adherence of many informatics initiatives.

In this research we develop a comprehensive model to explain how users process persuasive reminders. We will focus on creating persuasive email reminders to help diabetics adhere to their clinic visits. We prototype various persuasiveness messages and assess impact on adherence by patients.

Zhihua Tang, Todd R. Johnson, Jiajie Zhang, & Juliana Brixey (UT Health Science Center at Houston)
The safe operation of a medical device requires that the device has good usability. Due to many reasons, however, it remains a challenge to take device usability into full consideration during a typical device procurement process. The current project aims to develop a set of usability guidelines that can be practically incorporated into the device purchasing process. The guidelines include two components. The first is a usability/safety matrix that captures the full range of medical device use while emphasizing the compatibility between users’ information processing and the device interface design. The matrix categorizes usability/safety issues into four areas, including device set-up and configuration, user-device interaction, instructions and documentation, and training. In particular, it stresses users’ information processing characteristics at various stages of user-device interaction, such as perception, information interpretation, user control, system feedback, and error correction, and illustrates how the user interface should be designed to accommodate these characteristics. The second component of the guidelines is a cluster of usability evaluation methods suited for healthcare environments. In general, device usability information can be obtained through user testing, heuristic evaluation, or during clinical trials. Such information may also be gleaned from device manuals, device error reports, published device reviews, and manufacturers’ human factors engineering plan and results. These methods have a varied degree of validity and also differ widely in time, cost, and other resource requirements. It is recommended that they be used judicially to meet the needs of a particular medical device purchasing project. By defining device usability in such terms that are familiar to healthcare professionals and proposing an array of practical methods for device usability evaluation, the current guidelines provide a useful tool to in corporate usability considerations in medical device purchase.

Kritina L. Holden, Ph.D., Danielle Smith, Li Hua (Lockheed Martin Space Operations, Houston, TX), & Mihriban Whitmore, Ph.D. (NASA Johnson Space Center, Houston, TX)

Wireless Crew Communication for the International Space Station (ISS) and Beyond

Unlike Star Trek, where crewmembers merely tap an insignia to communicate with other crewmembers, the International Space Station (ISS) astronauts must stop their current activity, travel to a communications panel, (Audio Terminal Unit-ATU), and speak into a wall-mounted microphone, or use a handheld microphone that is cabled to a panel. This unnecessarily increases task times, cable management issues, and crew frustration. A wireless communication system would improve onboard efficiency, and ultimately safety, since time to communicate in an emergency situation would be greatly improved.

The Habitability and Human Factors Office at the NASA Johnson Space Center has been working with the Electronic Design Branch at NASA on near- and long-term solutions for an improved communications system. The near-term goal is to provide a wireless solution that can be quickly tested and implemented on ISS. It involves supplementing the existing ISS technology base (ATU) with a well human-factored wireless communicator. Due to cost and schedule constraints, the communicator will be limited by the design of the existing ATU system; in other words, virtually all reconfiguration and channel selection will have to be done at the ATU. Once configured, the crewmember will be able to move about freely using the wireless communicator. A prototype is in development and planned for testing onboard ISS at the end of the project.
The longer term goal of this effort is to develop human factors requirements and design concepts for a wireless communication system for future mission applications.The current ISS ATU system is a powerful, but complex communications system with many capabilities and options. But, as is often the case with complex systems, the usability of the system presents some challenges. It is not possible to solve these issues for the current system, but they are serving as a starting point for specifying a wireless communication system for future missions.
The electronics group has been working on “back-end” solutions, while human factors personnel and habitability designers have been working to identify the human factors requirements, and begin the conceptual design work for the near- and long-term communications systems. The effort began with the collection of information about the benefits and limitations of the current system, and a needs analysis for future systems. This information was gathered via interviews with long duration crewmembers, and a focus group session with stakeholders. The poster will describe the basic human factors challenges, the activities completed to date, and some examples of near-term and future wireless communications design solutions.

Amber B. Raley (Rice University), Jennifer L. Lucas (Agnes Scott College), Christi Washington (University of Houston), & Melissa A. Blazek (University of Tennessee at Chattanooga)
Despite the ubiquity of products designed with the assistance of engineering psychology, ergonomics, and human factors psychology, many in the general public are unaware of the three fields (Stone & Moroney, 1998). In addition to the awareness activities put forth by the HFES Bulletin (Cuevas, Shapiro, & Young) and the teaching suggestions provided by Carkenord (1994) and Stone and Moroney (1998), the current study recommends that more attention be devoted to the selection of introductory psychology textbooks containing engineering psychology, ergonomics, and human factors psychology content as a way to inform students about the three fields.

Each year, more than one million North American students take a course in introductory psychology (Miller & Gentile, 1998). Introductory psychology is the only psychology course taken by the majority of students in those courses (Buskist, Miller, Ecott, & Critchfield, 1999). Therefore, introductory psychology is often the only chance for students to be exposed to engineering psychology, ergonomics, or human factors psychology in an academic setting.

The current study examines the representation of engineering psychology, ergonomics, and human factors psychology in introductory textbooks. Of the 62 textbooks evaluated, the average number of pages containing content was 0.35 (SD=0.75) for engineering psychology, 0.23 (SD=0.93) for ergonomics, and 0.81 (SD=1.46) for human factors psychology. The average percentage of content pages per textbook was 0.06%, 0.02%, and 0.13% for engineering psychology, ergonomics, and human factors psychology respectively. Over half of the textbooks analyzed had no content pages of engineering psychology (47, 76%), ergonomics (57, 92%), or human factors (37, 60%). Of the 59 textbooks with glossaries, most did not list engineering psychology (51, 86%), ergonomics (55, 93%), or human factors psychology (52, 88%) in the glossary. Engineering psychology, ergonomics, and human factors psychology are all grossly underrepresented in introductory psychology textbooks. Increased representation would enhance the visibility of the three fields to introductory psychology students, an important audience, considering many of them will enter a psychology profession.

Frank P. Tamborello (Rice University)
Visual Displays: The (Modeled) Highlighting Paradox Download Poster
Making certain items in a visual display salient, such as highlighting by color, can aid human visual search. Fisher and Tan (1989) conducted two experiments examining the effects of highlighting on a visual search task. Subjects were told to search an array of five numerical digits for a target number, and that sometimes one of the digits (not necessarily the target) would be highlighted. They found that manipulating the proportion of trials in which the highlighting was a valid predictor of an item’s status as target or distracter (highlighting validity) impacted performance on the visual search task. Subjects were 90 ms faster on trials with valid highlighting in the 100% valid condition than they were on trials with valid highlighting in the 50% valid condition. Furthermore, even at 50% validity, there is still a greater chance that the highlighted item is the target compared to the 12.5% chance that the target is one of the four other items on the display. And so there is still some incentive for subjects to check the highlighted item first. Consequently, within the 50% validity condition, the valid trials are still faster than the control trials. The authors speculated that the difference in performance on valid trials between the two validity conditions occurred because subjects did not always attend initially to the highlighted digit in the 50% validity condition, but always attended initially to the highlighted digit in the 100% validity condition when they saw that the highlighting was more predictive of the items’ status as target or distracter.

In this paradigm, subjects have two basic strategies they can adopt to complete each highlighted trial as quickly as possible: they can look at the highlighted object first, or they can ignore it. Subjects should always look at the highlighted object first in both conditions, but that might not be such an obvious choice in the 50% validity condition. Do people learn to pick a strategy based purely on the overall time cost of doing trials one way or the other, or are they somehow sensitive to the probability of the highlighted object being the target? The study described in this report attempted to come at this question by constructing functional models of a person performing Fisher and Tan’s task.

Two models were built using the ACT-R/PM cognitive architecture. The models ran on a simplified version of the Fisher and Tan paradigm. It was found that in order to simulate the difference in response times Fisher and Tan found between the two validity conditions, the model needed to be sensitive to the relative frequencies of validly highlighted and invalidly highlighted trials. The results of the modeling experiment presented in this study imply that people seem to be capable of making assessments about the relative frequencies of valid and invalid visual cues, and that these assessments are necessary to determine how much those cues should be relied upon in a visual search task.

Christy Avera Harper & Arthur Hart (Hewlett Packard)
Linking Human Factors and Reliability to Create Remote Control Button Specifications
Download Poster

In a human factors test designed to analyze perception of button feedback on remote control devices, participants described some buttons as having no click and/or being either too sensitive or too hard to push. This is good feedback but does not provide a definitive measurement. Likewise, quality and reliability testing of the same remotes established the peak to trough ratios of the buttons, showing the ratio between the force applied to engage and the force felt during verification. This provides a quantitative measure, but what does it mean to an average customer? To provide meaning to these data, we compared the results of analytical machine measurements with subjective human analyses. To provide the link between machine measurements and human perceptions, four remote controls were investigated for button characteristics independently during a human factors test and standard reliability tests. Correlating these qualitative human impressions with the machine characterization resulted in the following specifications for remote control buttons:

Force to initiate the remote button operation should be between 0.15Kg and 0.2Kg (could be thought of as ‘Sensitivity’)
Ratio between remote button peak and trough forces should be at least 1:0.8 with a higher ratio being better (could be thought of as ‘Verification’)
Distance traveled between the peak and trough should be less than .25 mm (could be thought of as ‘Crispness’)

Comparing human feedback and machine measurements provided the link between qualitative and quantitative measurement that was needed to specify values for remote control buttons. Additionally, these specifications can be reused for other products requiring a remote control device.


Patrick Caldwell, Veronica Acevedo, Robin Clark, Jewel Darby, Alyssa Fox, & Meredith Tabor (NetIQ Corporation)

Usability Evangelism – Shining New Light on the Design Process

A software development company seeking to improve profitability, NetIQ Corporation set up a task force to figure out how to reduce costs and improve sales. The task force proposed the following two initiatives to meet the objectives:

1. Revise the product development and approval process
2. Promote usability awareness among all team members

Changes to the product development process elevated product feasibility analysis and usability assessment from after thoughts to business drivers. The new process required usability studies in the initial phases, resulting in a more focused product plan and more specific feature proposals. Every team member could start the project with more focused tasks and objectives.

The task force also developed the Fab Four training initiative to accomplish the following objectives:
1. Simplify product evaluation (trial-ability) to improve sales conversion rates
2. Make it easy for customers to “try before you buy” (install-ability) to shorten sales cycles
3. Make products user friendly (usability), to reduce technical support calls
4. Immediately expose a clear value proposition to speed product acceptance

The Fab Four, Trial-ability, Install-ability, and Usability while providing a Clear Value Proposition to the customer, were not overnight superstars. The training demonstrated why usability could benefit each team member. It also showed how building in usability at every stage of product development could make each team member’s job easier and move the company towards greater profitability, improving job security.

Many team members were resistant and did not see the value of these changes. Some managers thought usability testing a waste of time and refused to plan time in the development schedule. Some teams simply disagreed that usability is a business driver and continued to ignore it.

For every manager and team member who shunned the usability movement, many more embraced it and continue to see repeated value. Information Developers are often among first to convert and spread the word since usable products are easier to document. This presentation defines the Fab Four spirit and method, and shows how one product team at NetIQ Corporation embraced the Fab Four to turn a dying product into the most profitable product in its division today with a substantial market share.

The presentation describes the Fab Four in use by the entire product development team including Product Management, Development, Information Development, Quality Assurance, and Technical Support. The session also highlights key points of the product design and development process, integrating heightened usability awareness in every phase.

We show specific examples of how any product team can use the Fab Four to improve profitability, and include user interface samples before and after applying the Fab Four criteria. By incorporating simple prototyping techniques along with practical user testing, many people at NetIQ Corporation have changed from usability snobs to usability evangelists.

 

Mike A. Ferrer, Marcos Jaramillo, Richard Morency, Luis Velasquez,L. Javier Gonzalez, Sudhakar Rajulu
A Process for Space Suit Sizing Using a Vitronix 3D-Body Scanner and Pedus 3D-Foot Scanner

The objective of this study was to establish a process to collect anthropometric data with the use of scanner technology that can be used in the sizing of Extravehicular Activity (EVA) space suits for astronauts. A process including the use of a Vitronix 3D whole body scanner, Pedus 3D foot scanner along with ScanWorx and Innovmetric’s Polyworks software was developed to obtain the data. Results show that a total of fourteen poses are needed for collecting 118 required measurements of the suit sizing protocol. A validation study included using three subjects and three measurers. The data collection and extraction was organized in such a way to minimize the time required to collect the data. This process concludes by providing the least amount of poses and the appropriate tools needed for collecting and processing the anthropometric data efficiently. Further studies on new technologies must be done to provide greater efficiency in collecting the data and the measurement process.