They were three small words, but their meaning was huge.

Sitting in a lab at the University of Pittsburgh last week to help test the BrainPort vision device, Jose Neto read out the words projected onto a wall in front of him.

CAT. HAT. HOT.

For a person who is totally blind — as Neto has been for over two-and-a-half years since he was shot in the face by a bullet in Chinatown — identifying words on a wall is revolutionary. And so is the study taking place here to test the prototype of a device that aims to introduce a new type of vision for those who can't see.

Neto's flown from Calgary to participate in the study and says his first days of training have proven successful.

"It's great. You would never know you could do it again," he says.

"It's kind of weird because even though we don't see anything, we feel it in our tongue and the information goes to your brain. When I'm telling you about it right now, I can visualize — actually visualize — the things I did as I was seeing them."

Standing with Neto was his friend Kevin Taylor, who travelled to Pittsburgh with him to help.

"I'm almost in tears," says Taylor.

"I was there when Neto read the first three letters and I was like, 'Wow.' We put three letters on the wall and you read all three of them."

"The first time he did it, he got it all on his first try. I was amazed."

The BrainPort vision device is undergoing medical testing and is not yet available on the market.

It consists of a camera, mounted on glasses that transmit information via a hand-held computer to a mouthpiece that sits on the tongue. The tongue pad emits small electric pulses to produce a pixelated, real-time image of what the camera is capturing. It allows users to decipher the images and, with practice, visualize what is in front of them.

As one of 30 people participating in the study, Neto's portion started last week with baseline testing to establish his level of blindness. With two prosthetic eyes, it's no surprise he is completely blind.

Then, came a few challenges while wearing the BrainPort.

A cup was placed in front of him with sugar packs and stir sticks scattered around the table.

Without feeling the tabletop with his hand, as he normally would to find the location of something he can't locate, Neto's task was to decipher the information on the tongue, reach out, grab the sugar pack and place it in the cup.

"With the BrainPort it was really, really easy. At first you wouldn't understand the information, but as long as you learn to translate information, you get it really well.

"We would identify where the sugar packs were and go pick them up right off the bat," says Neto.

Later, came the letters.

"There was an 'A', 'R', which is kind of hard, and 'C'. And then they put the W and I got kind of confused. I said, M, though."

To identify letters, the trick is to group letters into shapes, he says, either triangles, circles or squares.

"The A would be a triangle one. The top is small and goes open. Then you feel the bar in the middle," says Neto.

"For me, it was five minutes training with that and I got it right. You can imagine if you have six months training."