Our research questions were largely answered by rapid prototyping. We iterated over the design of the monument, inside and outside, to test our assumptions about how its physical attributes would instruct the user on how it should be used. We set the prototype in an area with high foot traffic for our intended audience, observed, altered our monument, and then put it back for the next round of traffic. This process was repeated several times over about a week.
Observations
The key assumptions we tested about each design were the following:
- The user will lean into the porthole when they’re interested/when something catches their eye.
- The user select a few different people, view them shortly, and then scan a QR code.
- The user will be able to give brief information about the person selected.
Test 1
- Put hand in viewing hole
- Expressed joy when collages noticed
- Verbalized confusion about QR code
Test 2
- Looked at bottom of controller
- Looked at a couple collages and then scanned the QR code
- “I’m blown away…this is beautiful”
- Verbalized recognition of names
- Verbalized confusion about sequence
Test 3
- Originally put his fingers on the hole but quickly realized its the Campanile
- Observed names in sequential order
- Left briefly, just passed by
Iterations
Because the monument is made up of several different points of interaction, some parts were tested independently. For example, the depth of the monitor were tested independently from the arrangement of names on the sensor panel. This was especially helpful for addressing our assumptions
One interesting discovery was that the sensor panel (and viewing hole) needed to be set at an angle, otherwise purpose of the hole (for an arm? for an item?) was too ambiguous. We also discovered that, in order to reduce cognitive overload, it was necessary to include a visual cue that the historical figures were thematically connected and offer a suggested viewing order.
We addressed these issues, and the results of our research questions, in our final design. Our intended interaction sequence was able to completed by our target audience without prompts or instruction.