I spent the summer after my freshman year of college working in my school’s Human-Computer Interaction (HCI) lab. My work was funded by WPI’s EREE program (Early Research Experience in E-Term, basically a summer research program designed to get you the research experience you need to get hired elsewhere), which only had a 13% acceptance rate that year.

Although HCI is not my primary discipline, it was a wonderful experience I am extremely grateful for. Working under Dr. Erin Solovey and her PhD students gave me invaluable insight into what academic research really looks like, and set me on a path to potentially pursue a career in research myself. Even if I go a different direction, most tech projects can greatly benefit from an HCI outlook, since it is a data-driven approach to considering the needs and wants of the end user, as well as measuring the project’s effectiveness after prototype production.

The project I worked on focused on a capacitive fabric prototype from Drexel University. The fabric uses a single carbon-coated conductive nylon thread to add the touch-sensitive functionality, with a mapping algorithm to tell at exactly (x,y) position it was touched. The rest of the fabric was made with green and purple non-conductive polyester yarn.

The spring before, the PhD candidate I was primarily working under, (now) Dr. Shruti Mahajan, ran a study exploring gesture design. Using a paper rectangle and a Mickey Mouse cut-out, she led participants through a series of referents (for example, Mickey moving from one part of the screen to another) and asked them which gesture they would do to cause that. My favorite example, having grown up with early-2000s Disney Channel, is that to make Mickey appear, one participant drew the Mickey Mouse ears with her finger. Shruti would go through all of the referents with the fabric on the lap and table, alternating which one was shown first to each participant. The neat part: the fabric acts as more than just a touch-screen. You can also fold it, stretch it, twist it—you name it! Future prototypes will be able to accept all sorts of inputs, and we wanted to figure out which ones users want and when.

My job was then to analyze all the data. But first, I had to collect it. I was handed a video recording for each participant, then I used ATLAS.ti to code (sticky note) timestamps in the video with the referent, gesture, and body part(s) used. Many, many hours later, I transferred that data into Excel and went to work. I employed inductive thematic analysis, which, in simple terms, just means you look at the data and then look for themes in it, as opposed to starting with boxes you want the data to fit into and then sorting it into them. This is a more data-driven approach, which is perfect for a study like this. Please refer to my poster below for some of the results—they’re quite fascinating!

As a side project in the lab, I worked on retooling a previous senior capstone project that used augmented reality (AR) to create an interactive memory quilt on the fabric. The exposure it gave me to the HoloLens led me to be hired on another project for the upcoming semester, where I handled the deployment and recording features (including writing my first C# script) for an AR application for electricians. The project partnered with Schneider Electric, who wanted their employees to be able to see the readings (voltage, current, etc.) of what they were working on in real time while being completely hands-free. I did quite a bit of prep-work for an eventual study before I left the project, as well as provided my input as the only member of the team with an engineering background (which is the nice way of putting, “the only person who even knew what units you use for the metrics, let alone what they mean”). This project really pushed me beyond my comfort zone, and helped me improve at teaching myself new systems. I definitely didn’t become an expert in anything, certainly not when I was using such in-depth platforms such as Unity, but in that discomfort I became a much stronger and more versatile programmer.