Our Voodoo Bear's beady eyes from Colin Narver on Vimeo. We initially programming the eyes to flash in a consistent pattern of high and low, but ended up later having them only on high while talking to the user. Otherwise, the eyes were on low.
Photo of the point where our back flex sensor snapped. We learned that we had to reinforce our flex sensors with additional rubber tubing as well as a piece of structural cork support in order for it to be strong enough to withstand the hugging and punching that the participant delivers.
Photo of us testing the Emic 2 test-to-speech module hooked up to two buttons. One button triggered a random phrase from the positive reaction array strings and the other a random phrase from the negative reaction array string.
Documentation of our arrays of strings that included the positive and negative phrases that the user would receive based upon their chosen interaction. At one point we were going to use a hair brush with magnets that would trigger hall sensors within the bear in order to elicit a positive response. The hall sensors were unfortunately very fickle and hard to calibrate. As a result of our testing, we determined that using them would be and unreliable means of making interaction. As a result, we scrapped the idea.
The day of our midterm presentation. We bought several speakers that we intended to insert within our voodoo bear, but were unfortunately far too quiet. As a result, we used two external battery powered speakers connected together to boost the volume. During the presentation, our bear had a hard time differentiating a punch from a hug. I think we punched him a few times in a row and he kept saying "that feels great," which was a piece of unintended humor. We realized that in order for our flex sensors to work in the way that we intended them to, we needed to achieve a level of resolution from the sensors that was good enough and reliable enough for us to get the results we hoped to achieve (either a clearly negative or positive response). This led us to an investigation of new, conductive materials for the final where we could return unequivocal results garnered by on our desired input.
Here is a link to a video of us playtesting the bear at the midterm: http://tinyurl.com/dx8c8os