Predator Exhibit at the
Georgia Aquarium


As part of the Expansion 2020 project, the Georgia Aquarium aims to introduce the Predator exhibit shark tank. Contrary to the exhibit's name, the aim of the project is to use technology to educate and flip the perception of sharks, from fear to fascination.


We implemented User-Centered Research methods to inform the design for an interactive screen exhibit with a mobile-based interface, that caters to multiple family members. The solution educates about the role of sharks in the food chain and demystifies myths that make them scary.


The Aquarium will use our research inferences to inform the design decisions of the upcoming Predator Exhibit, especially in terms of guest management. They will also incorporate features from our design into interactive and educative solutions that are feasible to incorporate.


August 2019 - December 2019


Aditya Kundu
Jae Hyuk Kim
Hyun Tae Park
Ruchita Parmar
Arpit Mathur


Conducted Research and Analysis in a team of five. Led brainstorming and design sessions. Prototyped the interactive screen solution, and tested with users and experts.



Literature Review
Field Observations
Survey Design
User Interviews
Data-Elicitation Methods
Affinity Mapping
Journey Mapping


Feedback Session 1
Feedback Session 2
Final Concept
Feedback Session 3
Design Iterations


High-fidelity Prototyping
Shark-Head poster
Mobile Web Application
Interactive Screen Game
Input panel


Stakeholder Feedback
Cognitive Walkthrough
Task-based Usability Testing
Next Steps


Problem Statement

To kick off the project, we had a detailed meeting with our stakeholders to understand their expectations for the project. The result of the meeting was a triad of goals, as follows:

Provide a technological and interactive solution that helps alter the perception of sharks in the minds of the aquarium visitors from “fear” to “fascination”.

Provide a solution that provides an entertaining and immersive experience to the aquarium guests.

Provide a solution that ensures the safety and comfort of the marine animals, while also keeping them at the center of the visitor’s focus.

User Research

To begin, we conducted research with visitors at the aquarium, to understand the shortcomings of the current exhibits. The proposed design of the predator project can be seen on the right, and we decided to conduct research in current exhibits with similar structures.

We employed the following research methods in order to gather data:


WHAT did we want to know?

- Challenges in designing solutions for exhibits
- Existing exhibits that educate viewers

WHY this method?

- To leverage existing work
- Give us directions to pursue
- Avoid making mistakes

HOW did we do it?

- We went through various publications of studies on exhibit design
- We read about Interactive installations in various contexts (for instance, museums)


WHAT did we want to know?

- Understanding of exhibits at the aquarium
- Behaviour of the visitors at the aquarium
- What works and what doesn’t

WHY this method?

- Captures the context of the aquarium
- Unobtrusive, does not interfere with the users

HOW did we do it?

- Spent time silently observing the users from a distance, and taking notes
- Categorized observations using the AEIOU framework
- Conducted the observations for 5 exhibits similar in various ways to the Predator exhibit


WHAT did we want to know?

- Generalizable and quantifiable data about the likes, dislikes and expectations
- Identify the target user groups for our designs

WHY this method?

- Easy to get data from a large numner of people
- Low effort, cost and time

HOW did we do it?

- We designed a survey on qualtrics informed by our observations
- We distributed the survey online, and also went to the aquarium for on-site surveys
- Received a total of 202 responses


WHAT did we want to know?

- Qualitative Data pertaining to the experience of the visitors, as well as group dynamics

WHY this method?

- Gives us the opportunity to explore themes through the follow-up questions
- Gives us rich amounts of non-verbal data

HOW did we do it?

- Identified themes informed by the observations and survey results
- Used the themes to guide an interview session with 10 users
- Asked follow up questions to explore interesting avenues

Data Elicitation

WHAT did we want to know?

- Validate the perception of sharks amongst people
- Identify what characteristics of a marine animal cause positive or negative perceptions

WHY this method?

- Helps capture subconscious behaviours of users
- Avoid putting bias on users

HOW did we do it?

- Had 21 participants sketch sharks and provided rationale behind specific features of sharks
- We had them make personal meaning maps to identify keywords associated with sharks

Data Analysis

The research methods provided us with a lot of raw data. We categorized the data into three different categories and employed different methods to make sense of them. The goal of the analysis was to convert raw data into formats that inform brainstorming.


We collected a lot of quantitative data, primarily through surveys as well as through field observations. In order to identify trends and opportunities for designs, we visualized data in the form of bar graphs and pie charts. The trends helped result in inferences and personae.


We had a lot of qualitative data from all of the research methods that we conducted. To organize the data, we built a bottom-up affinity diagram. We started with individual user quotes and observations, and after four layers of grouping, identified themes.


The data from the data-elicitation methods consisted of sketches and personal meaning maps. We created word-cloud visualizations to illustrate the perception of sharks in the minds of the general public. This was validation for the assumption that sharks get a bad reputation.

Research Outcomes

User Research provided us data. However, data is not usable to the designer in raw format. After the analysis, we ended up with three formats that would help us brainstorm design solutions, and also ensure our designs are user-centered.


We identified some notable findings through our research, and discussed the implications these findings might have on our proposed solutions. We identified 18 such findings and their implications. Some of the findings relevant to our final design our listed here.

FINDING: Most visitors to the aquarium travel in families, or groups of friends.

IMPLICATION: The solution should afford interaction with, and facilitate interaction amongst, all members of the group/family.

FINDING: People are interested in learning facts about the marine animals, but the information dissemination mechanisms are ineffective.

IMPLICATION: Employ better methods of information dissemination that are attractive, as well as helpful in retaining information.

FINDING: Parents often have to compromise their own experience at the aquarium in order to provide a better experience for their children.

IMPLICATION: Consider solutions that enable different family members to play different roles while providing a pleasant experience to all.

FINDING: The visitors enjoy experiences where the aquarium anthromorphises the marine animals, by attributing human-like characteristics such as name, personality etc.

IMPLICATION: The solution should develop empathy for the animals by helping the user identify human-like traits in the animals


In order to ensure that our designs remain user-centered, we took the data gathered from our User Research and designed the following Personae that encapsulated the characteristics of our target users. We went from designing for visitors to designing for the Richards family.


To design a positive experience for the Richard family at the aquarium, it was important for us to identify what works and what doesn’t. We created Journey maps of three different Georgia Aquarium exhibits to capture the high and low points of the existing visitor experience.



Now, we wanted to use the research outcomes to brainstorm a large number of ideas. We put up the design implications, personae and journey maps on the board, and spent time thinking of ideas individually. We then had a discussion and came up with more ideas to pursue.

We evaluated our ideas on the basis of feasibility, effectiveness as well as creativity. We shortlisted 4 design ideas, taking different approaches to tackle the problem statement. We sketched the essential features and working of the ideas.


Through a mobile app, you can scan a shark at the aquarium, and find out about the rescue story of that particular shark.


A touch based interface, where you can interact with digital representations of the sharks, to get information about them.


You can view the point of view of a shark through your mobile screen, and identify whether different objects are food or not.


An interactive projection on the walls of the tank, visualizing a dirty and overpopulated ocean without of sharks.

Feedback on Ideas

We needed an evaluation of our ideas to identify characteristics that are desirable. This evaluation needed input from the stakeholders, users as well as the designers. To gather this feedback, we conducted the following three types of informal evaluations:


WHAT did we want to know?

- Feedback in terms of feasability, and efficiency of solutions
- Get an idea of what the stakeholders consider a good solution

WHY this method?

- We conducted an informal meeting with the stakeholders
- We presented sketches and storyboarded the ideas
- The discussion also explored potential ideas similar to the sketches

Focus Group
with Users

WHAT did we want to know?

- Feedback on the user's impressions of the ideas
- Feedback on the how well the solutions are as an experience for visitors

WHY this method?

- We recruited 4 participants and conducted a focus group
- We presented the sketches and facilitated a discussion amongst the participants
- At the end of the session, we had the participants rank the ideas


WHAT did we want to know?

- How well do our proposed solutions align with our research
- Do the solutions cater to our personae's needs

WHY this method?

- We created four different matrices for each idea
- On one axes we mapped our design implications, and on another we mapped the jobs to be done
- We highlighted the cells that each solution catered to

Based on the feedback, it was clear that while each design idea had unique positives, they also had several issues pertaining to usability, feasibility, and safety of marine animals. There was no clear favorite among the four ideas for both stakeholders and study participants.

Converging the Ideas

On the basis of our feedback, we took the positive components to form a coherent experience for the visitors. The final solution would consist of interactive QR code posters, a mobile web-based interface, and an interactive screen. The flow of the proposed solution is as follows:

We then went on to develop low-fidelity prototypes of each component of the solution. We kept the prototypes low fidelity, in order to gain conceptual feedback, rather than feedback about the details of the design. The prototypes consisted of wireframes and paper prototypes.


The posters will attract attention from the visitors, who can then scan a QR code on the poster to get more information about that particular shark.


The user is brought to a mobile web based interface that allows them to get information about the particular species of shark, and also personalize their own "pet" shark


We simulated the interactive game on the screen through paper prototypes. The screen would have an interactive panel, through which the user can add multiple animals, and find out if they are food for sharks or not.

Testing Low Fidelity Prototypes

We wanted to test the concept of the entire flow of our solution. To achieve this goal, we recruited 4 participants and had them role-play members of two different families. They then interacted with the prototypes and provided us with feedback regarding their interactions.

The session provided us with multiple insights. The participants responded positively to the proposed solution. We got feedback on the design of each component of the solution and altered the designs to reflect them. We implemented these changes in the high-fidelity prototypes.


High Fidelity Prototyping

Each of us spent time developing high-fidelity prototypes of the different components of the solution. We 3D printed the poster, designed the web-based interface on sketch and principle, and developed the screen-based game using p5.js and


Created for each shark species, various posters will be situated next to multiple viewing glasses. There is the lifelike shark that first draws the attention of visitors. On the bottom, there is a QR code, scanning which redirects the user to a mobile-based web experience.

The following figure represents how the testing influenced the high-fidelity prototypes:


The QR code leads to a web-page, through which the user can access information about the species they have scanned. The user can personalize a digital shark of the same species. Finally, the user is given instructions to proceed with the final aspect, the screen-based interface.

The following is an end to end flow of a user interacting with the high-fidelity prototype of the mobile-based webpage:


The personalized shark is dropped into an interactive screen interface through NFC technology. There is also a panel of buttons displaying different species of marine animals. Pressing a button drops a specimen into the ocean, and interacts with the sharks to give information.

Inserting shark in dormant screen
The user can enter the shark by placing their phone on top of the NFC reader. This enters the shark in one of the columns on the screen. A dormant screen visualizes a dirty and overpopulated ocean, which symbolizes an ocean without any shark in it.

shark interacting with food
The user can then interact with the button panel, and enter different marine animals into the interactive screen. If the animal is food for sharks, the user's shark approaches the animal and eats it up. A popup gives data about the role of sharks eating that particular species.

shark interacting with human
If the user enters an animal that their particular species of shark does not eat, the animal approaches the animal but does not eat. A popup gives data about why the shark does not eat that particular animal. This can help demystify myths about how sharks eat humans.


Summative Research

With our final designs ready, we moved onto conducting a thorough and detailed evaluation. We wanted feedback on specific design details of our high-fidelity prototypes, and the experience our solutions provided to our users. We wanted feedback from end-users, as well as experts.


WHAT did we want to know?

- Feedback regarding learnability of solution
- Uncover critical issues with the solution

WHY this method?

- Utilizes the experience of people with expertise in particular field
- Provides rich amount of qualitative data
- Helps identify problems at relatively low cost

HOW did we do it?

- Conducted the session with 6 experts: 4 marine biologists and 2 exhibit designers
- Asked them to interact with the solution by giving them tasks, and interviewed them

Task Based
User Testing

WHAT did we want to know?

- Perception of the user regarding the solution
- Feedback on Usability of the solution
- Evaluating the enjoyment of the experience

WHY this method?

- Rich qualitative data in the form of follow up questions
- Ability to observe interctions of users

HOW did we do it?

- We recruited 8 participants, and had two sessions with 4 participants each
- We had the participants play different roles of family members, and had them interact with our solution through the Wizard of Oz technique
- We interviewed the participants about their likes and dislikes

Inferences from testing

The solution was received well and provided an overall pleasant and enjoyable experience to the visitors. The testing also suggested that there was scope for improvement, and certain aspects of the solution required more thought. These are shown as follows:


Visitors might not be willing to scan QR codes without knowing what is in store for them. They cannot expect what are the outcomes of scanning QR codes, so they are apprehensive of doing so. It will be better to get a preview of what comes next after scanning QR codes.


Some users did not understand what NFC scanning is and how it works. There is too much technical jargon used that may make users uneasy. We can replace jargon with friendly language that everyone understands and make it accessible to users with varying levels of technical skills.


The message behind the dead and barren ocean in the absence of sharks was misunderstood by most people. We can animate the interactive screen to visualize the percentage of sharks in the ocean, developing a stronger correlation between the ocean and the population of sharks.

Lessons Learned

Working with industry partners

I learned that it is important to keep the stakeholders in the loop, and their goals can be in unison with the users'.

Involving users in iterative design

I learned the value of taking feedback for every iteration of the solution, and how it helps the designer look beyond their biases.

Context is important

Richer data can be gathered on the field, as compared to recreating the setting artificially in a lab. Moreover, the fidelity of the design effects the quality of feedback.