Internship Reflections – AMNH

In a nutshell, what I did in American Museum of Natural History during the 2017 summer internship:

  1. Evaluated the user experience and learning outcomes of the Galactic Golf (a newly-developed educational augmented reality experience using Hololens): developed the evaluation instruments, facilitated the floor-based evaluations, process the data, and work with partners to interpret the findings.
  2. Designed handbooks of all the digital and physical interactives (161) in the halls of American Museum of Natural History: conducted survey over all the interactives in the permanent exhibits, designed layouts and icons of the formed snapshots, and optimized workflow of content creation by merging data.

The following is my reflection of the internship on a weekly basis:

Week 1

On May 24th, I started my internship. On my first day, Mr. Barry Joseph, the Associate Director for Digital Learning, gave me a to-do list for learning how my work will be like:

  1. ✅ Review 1st Day for Interns with Barry
  2. ✅ Read Welcome from past interns: Great rules of thumb of how to conduct evaluations. Cannot wait to put into practice. It seems that the previous interns really learned a lot.
  3. ✅ Lessons Learned from Prototyping Digital Specimen Interactives in FY17: It is thrilling to see the insights of observing and interviewing plenty of participants. The lesson learned is also the cornerstone of greater iterations.
  4. ✅ Review Google docs to understand how evaluations are being done: Evaluation is important in the design process. With well-defined purposes and well-designed instruments and implementations, evaluators are conscious with the participants’ authentic feedback and capable of summarizing patterns to answer questions from data collected.
  5. ✅ Review Hall Snapshot books: Snapshot is like a condensed ethnographic field note especially made for the museum, which really grasps moments from experiences of visitors. And it is beneficial for discover problems in the field and therefore to enhance the better learning experience.
  6. ✅ Visit some halls: this tour basically happened when I was trying to find my way out of the museum. Although I have been here several times. 😂

The rest of two weeks I was doing a brand new Hall Snapshot books: I was trying to find out all the interactive devices, no matter physical or electrical, and documented the content, mechanic, year of installation and marked their locations on the floor plans. The purpose for this marking is for designers to get some inspirations and iterate their design of a global climate change data visualization, which will be placed in the Gottesman Hall of Planet Earth.


Week 2

This week I am assigned to a new project: designing and implementing the instruments of evaluating Martian Golf.

Visitor is swinging the virtual golf club

Martian Golf is a blended reality experience played on Microsoft Hololens. The photos of Martian surface taken by Curiosity Rover is the most scientific feature of it. Barry first defined the research questions, and I drafted the questions in both the observation and interview forms. To come up with the questions, I reviewed keynotes from the Research of Game and Simulation class on my way to work every day and tried hard to align them with the research questions. Then Barry revises them according to his experiences. I really learned a lot from those revisions. For example, taking the where you conduct the evaluation into consideration. I first draft the question as “how many times did the visitor swing until they successfully strike the golf ball into the hole?”, then I deleted it for I thought there should be some user logs for us to check by mirroring the Hololens onto a laptop. Barry then added this question and explain why we should directly count those swings or ask the visitors: In the Hall of Universe, people will gather around or line up when conducting playtesting. Therefore putting a mirrored screen at a side will distract the other visitors, and also make the observer even harder to concentrate on keeping tracks. When I was in the field, I realized that was so true: primary school kids rushed into the hall. When one of them was participating in the playtesting, the others are asking questions, waving hands in front of the lens, and even try to read the reflections from outside of the lens. If there will be one mirrored screen in the hall, they will definitely discuss it extreme-actively. Both the participant and observers will be greatly distracted.

During the evaluation, I was still anxious and asked leading questions “do you like this game?” once, and this answer makes the observation invalid somehow. For the participant will generally give positive answers for being nice. I realized sometimes it is challenging to ask questions because the participants and I are both use English as the foreign language. In all, I think that keep on practicing will gain myself more experiences of doing such job.

Week 3

This week I continued to work on Martian Golf and Hall Snapshots of Interactives.

Martian Golf

A 3D printed prize for the playtesting participants: the crater where Curiosity Rover landed

After bugs were fixed, we took the Hololens to the floor again and did playtest. In the after the interview, we found that two visitors drew the conclusion as “golfing on Mars requires more force”, which is the opposite of the learning objective. After consolidating all the data from interview and observation, and also the feedback from the developer, we infer the reason was that mismatches of what the narrator told the players to do and what the player thought they are supposed to do. When the narration was “useless force”, the players often used more force on the clicker, hence they clicked longer than before. However, the player needs to click shorter to achieve the fewer force effects. As the conclusion, the confusions came in. The prototype was going to be iterated for eliminating this confusion, namely, redo some of the narration. I felt the experience of witnessing a misconception happened was really cool.

Hall Snapshots

After we charted all the interactives on the floor plan, we try to visualize it to another level. There is a method for the original designers to classify all the interactives: by labeling one item with the following three tags all together:

  1. The name of exhibit designer
  2. The working title of that item
  3. A descriptive title of that item

But we abandoned this coding principle as our goal was to make the snapshots more sense-making to the designers. So we did some card sorting based on the mechanics: what visitors do to the interactive; and what results they will have. Here are some icons to indicate the mechanics:

Icons for card sorting of interactives

VR Dino

This week also marks the initiation of HTC Vive VR experience. On Friday’s staff meeting, two dinosaur fossil experts participated in the brainstorming. It is exciting to see how scientists contributed in creating a both entertaining and educational experience. Here are also two trivia from the meeting:

  1. About 80% of the dino fossils in AMNH are real;
  2. Dino did not extinct, their descents are still an important part of the current ecosystem.

Week 4

  • This week Tianjiao, another evaluation intern, came in; so we collaborated in the Galactic Golf evaluation.  The first step was to learn how to manipulate Hololens by using gestures and voices. This week we conducted two rounds of field observations and interviews altogether. We will also focus on whether we can extend the digital experience to a self-guided hall experience (that extends the science learning from the digital experience).
Tianjiao and I are learning Hololens gestures

  • There was a science visualization lab breakfast symposium held on Friday this week. Three speakers gave speeches: Matthew Faerber, the Technology Lab Director from North Carolina Museum of Natural Sciences; Mark SubbaRao, Director of Visualization at Adler Planetarium; and Shari Werb, the Director of Education and Outreach, Smithsonian National Museum Natural History. From the presentation, we were informed that the speakers themselves have various backgrounds: ecologist, engineer, and anthropologist. All of them devoted themselves into presenting sense-making scientific findings to the general public. I am also aware of what kind of staff are needed for a future educational place in the museum: a person can use plain languages(e.g. sci-fi metaphor) to talk about “how scientists know what they know”. The skills of programming and troubleshooting and playing around with hardware are also essential in the learning environment. I found the requirement of skill sets would have guiding significance for whom want to work in museums or maker space.


Week5 & 6

AMNH Birds of North America by DK, now in Clearance only $10

These two weeks Tianjiao and I continued working on the Galactic golf evaluation. Another volunteer Grace and photographer intern Nick came to help. We were acknowledged that we should be approved by the parents or guardians if we want to use the photos of children in publications. In this stage, The numbers of participants are also growing from 10 to 20 in every session. Our evaluations are becoming more detailed and presenting more statistical results.

For the interactive book, the data merge function of InDesign was introduced for improving the workflow. By importing the spread sheet in CSV file, InDesign can generate multiple pages once in a time. I began to find out data merge because I am told that the data could be renewed, and it is very difficult to change my previous manual arranged layout one by one. Lucky me found data merge, one template could be used for every observation. It is just similar to making trade cards. Besides, I learned from the Encyclopedia Birds of North America that how to use color coding for the reader to find the information they need more handily, how to make the contrast between fonts: the sans serif and the serif, etc. Although people rely on the internet to do research, books are still inspiring in terms of designing layout and readability.

There are plenty of tutorials on youtube teaching how to do data merge. I followed this one, which is pretty straightforward with useful tips:

Week 7 &8

Last two weeks, the evaluation came to an end, all the findings were shared with developers and designers. I also got to know the prior design process and informed the specific pipeline of creating this experience.

For the interactive survey, we could show the completed volume of the lower level and 1st floor. Feedback from interactive designers is collected for updating the snapshots in the future. With the data merging workflow, any information changes would remain handy. Here is a video that flipping through the lower level volume:

Finally, I become an alumna from the Science Visualization Group(what Science bulletin changed its name into). I also left some welcome messages for the future intern: The educator guides are great resources for you to become well familiar with the learning environment within the museum. They have explicit introductions of the exhibits and detailed guidelines that will give you insights of how education could be achieved. You can access to them via the link:

Again, I want to say thank you to everyone on the board, especially Mr. Barry Joseph: I deeply appreciate your mentorship. Thanks for opening my eyes to what design thinking is, and how to be educational while stay entertaining!

My Polaroid moved from this board to the alumni board on 14 July 2017. 🤓

Share Post :

More Posts