I entered the fourth week of the program with a lot of motivation. This week’s main focus was to develop the Study Design plan and how we want to approach our research questions in terms of a study. I’ve met with my teammate at the beginning of the week to discuss our current study plan and start listing down our initial steps to conduct our study. I showed my teammate what software I use to modify captions files such as .srt and .ass files and how to burn them on videos using the command line. We also discussed reforming our research questions. To state them again, our research questions are: the first one being; how do we incorporate paralinguistic and emotive information in captions while retaining speaker identification? And our second research question; will incorporating paralinguistic and emotive information through the use of speaker identification help improve compression and retention for deaf and hard-of-hearing?

We have met with our mentors on how to improve our research questions and we decided to keep the same questions but remove speaker identification and only focus on how to incorporate emotive and paralinguistic information in captions. This way we will have fewer variables to work with and it can be more controlled in terms of developing stimuli. With that approach in mind. We want to make our study a 2 by 2 study testing 2 conditions. Condition 1 will be testing Memoji for displaying emotive information. Condition 2 will be using text description to display paralinguistic information. So to approach these two conditions we want to make 5 short video clips for each condition and this will be about 30 seconds each. For each video clip, there will be 2 versions, one with original captions with no modification and the other version will have captions that are modified to match the condition. So in all, we will develop two surveys with each survey having one condition. On each survey, there will be a total of 10 video clips. To develop these video clips, we have decided to develop scripts and film them ourselves as finding videos online through youtube is possible but not ideal as it will be harder to edit the captions. By developing the videos ourselves we are able to focus on the elements that we want to test in our research.

So, we made a document to develop our scripts and for each script, we have a targeted emotion or paralinguistic information that will be displayed in the video. We developed these scripts in mind of all of them having three speakers in the video. One of the five scripts in each condition will also have one speaker that will speak off-screen just to see if that has any effects on our conditions. With all the progress we had so far with developing our idea of how we want to conduct our study, I have translated that into a document discussing the step-by-step of our design that way it will remind us and help us stay on track. This way it also told us what area needs to be worked on. As this happened, more work has been conducted on developing the consent form, demographic questions, interview questions that include questions to ask when viewing videos, and follow-up questions to ask. Overall we want this study to be around 15 to 30 minutes long that way we can recruit many participants. Towards the end of the week after all our progress, we have discussed goals to achieve for next week and that includes making the videos, captions the videos, and start developing the methodology section.