Class projects - HCI/Research

To evaluate and redesign an existing system
Class: Engineering Psychology
Team members: Elsa Eiriksdottir, CJ Hutto, Jessica McMillen, Yelena Nakhimovsky, and Allison Sall

To dazzle Professor Corso with our creativity, we decided to redesign a karaoke system. We chose the International Bowl and Arcade in Doraville, GA. The manager graciously let us use the room one night for free since we flashed our student ids and Oliver Twist expressions. This particular venue featured private rooms for groups with a video display.

After coming down from the momentary high from our creative choice in systems, we began to fear karaoke wasn't the best choice afterall. An entertainment system? Where were our heads? We thought that no one really cared about the efficiency of an entertainment system since they are there with time on their hands and fun to be had.

Then, it hit us. People want to have fun at the karaoke bar, they don't want to spend 30 minutes trying to figure out how to work the system. That was all the inspiration we needed to pour ourselves into this project and devise an awesome plan for redesign.

Our final report included the following:
Mission Statement, Mission Scenario, Environmental Analysis, User Analysis, System Requirements, Function Analysis, Function Allocation Assessment, Functional Flow Diagrams (22 in total), Task Analysis, Workload Analysis, Subjective Workload Assessment (Modified Cooper-Harper), Interface Analysis, Alternative Analysis, Improvement Recommendations, New Mission Scenario, New Interface Design

Our recommendation was that a multi station touch screen table (seen left) should be installed in place of the coffee table. This table acts as both an input device as well as a place to rest glasses, food trays, beer pitchers, etc.

Open the gigantic final document in ZIP format



To design, prototype, and test an interface
Class: Introduction to Human-Computer Interaction
Team members: Leland Beauchamp, CJ Hutto, Courtney Lessl, Jessica McMillen

Given the theme of voting, my team settled on an in-car radio system that allowed the user to vote "Yea" or "Nay" on songs that played over the radio with the goal of creating the perfect playlist. We liked this idea because we were challenged to consider the divided attention of the user while driving, we understood the user group, and we felt passionate about the subject matter. Our assignment was to create the interface for the system, so we wizard-of-ozzed the functional details.

We began by analyzing the tasks and goals of the user as well as the environment of the car while interacting with the system. In doing so, we developed a user analysis, environment analysis, analysis of existing systems, task analysis, functional and non-functional requirements, and a scenario of use. After brainstorming implementations of the system and getting feedback from classmates on three divergent design possibilities through a poster session, we decided to go with steering wheel mounted controls and display. We proceeded with creating low-fi prototypes which informed the design of our high-fi prototype and physical mock-up.

Our final design
Working Computer Simulation Demo
The only hotspots on the demo are the places where the users should click according to the script we read to them. We did not develop the prototype to support incorrect paths because of limited time and resources. The script is here. The prototype is here.

Along with testing users with the computer prototype, we tested them in the car as well with a steering wheel overlay and a divided attention task. The divided attention task was a PowerPoint slideshow of photographs upon which an icon would randomly appear. When the icon was on screen, the user would have to say it out loud. We recorded how many times the user missed the icon. The slideshow played on a laptop that sat on the hood of the car facing the driver's seat.

We encouraged a think aloud protocol. Our jobs were as follows: CJ stood by the driver's side window to change out the displays, Courtney sat in the passengers seat and read the testing script and manipulated the car cd player to play the songs the script called for, Leland recorded what the user said and did, and I recorded the results of the divided attention task.

I learned so much from this project. Unlike my psych-major teammates, this was my first user study, and I learned how helpful and difficult they are. It is really fascinating to watch other use your design while they tell you what is good and bad about it.

Open the final document


Observe the use of technology in a public place
Class: Qualitative Methods
Team members: Leland Beauchamp, Jessica McMillen

Our original research question was "How do people use cell phones in public settings?" Leland and I chose to observe two sites in a local mall with different feels to them, the food court (loud, busy) and a lounge area outside one of the department stores (quiet, empty), because we were interested in discovering if different noise and crowd levels affect a person's interaction with his cell phone. After nine hours of observation on three separate occassions and many hours of painful attempts at applying grounded theory, we found our data did show not an interesting difference between patterns of behavior in the two areas. Instead, we found ourselves led to another research question, "How are interactions between individuals in a group impacted by a group member's interaction with a cell phone at the mall?" Our analysis revealed people try to obtain a certain level of privacy when using a cell phone around others, generally by adjusting their body position or looking off in another direction.

Having never engaged in qualitative research before, I learned a lot from this project.

Open the categories, category profiles, and code book.
Open the analysis/reflections document - it's really interesting!

Class projects - Design/Programming

Design and prototype of the Yea-Nay Radio web companion
Class: Intro to HCI

This is the web companion to the Yea-Nay Radio system described above. The Yea-Nay Web Companion site has three primary purposes. The first is to provide the new user with a way to jump start the development of his personalized playlist by giving the Yea-Nay playlist generation engine clues to his listening preferences. This would include ruling out (voting “nay” to) particular genres or artists he is not interested in or making sure some of their favorites are included (by voting “yea”). The second is to provide the frequent user a way to view and edit his listening profile so that he can change the weighting of particular genres or artists. The listening profile will be displayed on the screen as a collection of generalities the engine assumes about the user’s listening preferences. For example, if he has voted nay to enough Pat Benatar songs, then the display would reflect that in its weighting of Pat Benatar. The third purpose is to display a user’s listening history in order to correct voting mistakes that are made in the car. If a user mistakenly voted nay to “Take It Easy” by The Eagles the mistake could be undone in this area of the site.

The secondary purposes of this site center around providing information about Yea-Nay Radio for interested or accidental visitors. First, there will be the sales pitch to non-subscribed visitors explaining how Yea-Nay Radio can enhance radio listening and how the system works. Then, there will also be a way for visitors to subscribe to the Yea-Nay Radio service. Finally, there will be an area where users can suggest artists or specific songs that are not yet available on Yea-Nay Radio.

To the right, there are pictures of the paper prototypes. The HTML prototype is available here. There is only one path through the HTML prototype. The task prototyped is to jump start the profile by selecting various genres and artists liked or disliked. The steps to complete that task are as follows