Developing an Accessible Keyboard

As a child, Sid would often come to listen to our music, looking up at the clouds through our sunroom. We’ve always wondered what he was pondering over. Sid has Down Syndrome and is on the nonverbal spectrum as well, but growing up, he wouldn’t let that be his defining characteristic. Now twelve years old, those who know him will comment on his remarkable wit and intelligence.

Between those early days to this year, with the incredible support of his family, he has overcome many obstacles and learned to utilize his right pointer finger and sometimes a pencil eraser to spell out words on his letter board.

Sid communicates with RPM with the help of his mother.

Sid communicates with RPM with the help of his mother.

One day, Sid’s mother came to us with the question: can we improve his method of communication? We welcomed the possibility by observing at first. We examined his non-verbal mode of communication using a letter board and the Rapid Prompting Method (RPM) technique. We watched him spell out words and eloquent sentences, albeit slowly and by trial and error. Often, his mother would switch between different letter board sizes to confirm his letter selection, re-center the board within his range of motion, pull his arm back, or momentarily block his eyes to recalibrate. Many times, his limited fine motor skill impeded his ability to communicate efficiently, leading to frustration and exhaustion.

We noticed Sid would often indirectly look at the letter board using what we believed was his peripheral vision. Our hypothesis was that his visibility of the board, and thus ability to discern letters from one another would improve if environmental lighting conditions were optimally controlled. Our eyes collect most of the ambient light at the center of our vision, while peripheral vision collects very little. We speculated that under dimmer conditions, Sid could improve his visual and pointing accuracy. We set out to test this theory by controlling the brightness and color temperature of the ambient light. Sid’s feedback immediately revealed two things: he preferred dimmer and cooler temperature light. He also preferred diffused over focused lighting. While this reduced environmental distractions, it did not have a dramatic effect on his pointing accuracy. Something more had to be done.

Our next experiment consisted of increasing the letter sizes and improving the visual separation between letters. With the flat letter board, he would often slide to neighboring letters. We increased the separation distance between letters and added depth to each letter “key” by using small plastic boxes as buttons. We arranged these boxes with letters printed on its back face in a rectangular grid, similar to his letter board. His accuracy improved. However, the box size was rather large to practically fit the entire alphabet within his range of motion. It also became clear that Sid wasn’t receiving sufficient tactile (and perhaps haptic) feedback from touching the letters.

As far as the letter board method of communication has served him and others with his abilities alike, there is one crucial limitation: using the letter board relies on having an intermediary, in Sid’s case, his mother, to decode his input. As a result of this limitation, there have been a few mobile apps with features such as predictive typing. However, for Sid, these apps, often costing hundreds of dollars, just don’t work. The three main issues are that touchscreens have trouble detecting his input, the apps lack customizability, and the predictive input suggestions often limit the elaborate and long sentences he likes to make.

As is often the case with this era of technology, it is tempting to think that the solution lies in using more complex machinery such as artificial intelligence, when much simpler, yet just as effective solutions are possible. Our joint vision was to give the power to the individual, not the machine. The machine was to serve as a messenger for the human brain behind it, without distractions.

At Kapshur Labs, we aimed to create a device that satisfies three main goals: ease of use, independence, and customizability. Using our favorite platform, Legos, we crafted iterative prototypes, frequently seeking Sid’s input to inform our design choices. We looked at three main input button types: push, toggle, and lever. Conferring with Sid revealed the lever mechanism worked best. We optimized a few parameters, such as push force, key travel, and lever angle for ergonomics.

Sid tests our first keyboard prototype.

Sid tests our first keyboard prototype.

In recognition of the years of meticulous training required to use the letter board, we tailored our design to minimize retraining efforts to use the new keyboard device. We did so by closely replicating letter size, spacing, and the A-Z letter configuration. After two months of work brainstorming, developing, revising, and programming the keyboard, Sid is currently testing our first full scale prototype. In the coming years, we hope to gather further input, refine the design, and make the a potential platform available to others with similar needs.

Our work constantly evolves as we learn more about the needs of individuals facing similar challenges.

Written by Kapila & Sushruta Chandramouli, the lead investigators and engineers on this ongoing project.


Leave a comment

Comments must be approved before they are published.