Identity Crisis

A game of personalities and how we construct them

Individual Project
Research, UX, Interaction Design


How do we define our personalities? This was the starting question that brought me into this project. There are no shortage of options out there to put us in nice little boxes, but I’ve often felt unsatisfied with the assessments from these evaluations. If I’m one thing, what does that mean in opposition to the others? How do all these questions add together to define me in this way? What does this look like for other people? How can we explore these ideas together, without making people feel insecure about their choices?

All this brought me to the idea of a competitive card game, exploring aspects of our personalities without having to define yourself, but giving players the tools to do so if they choose.

How might we learn about ourselves in a casual and engaging way?


Identity Crisis is a game of constructing personalities. Players use models based on psychological assessments to compete and be the first to construct 1 of 4 personality groupings. By seeing how our personalities are constructed, players get an engaging look behind the curtain of how a person's thoughts and feelings contribute to who they are.


The first player to build a personality wins. When no cards remain, the set closest to a personality is the winner.

There are 5 types of cards.

Each card has one defined state. Positive (+) or Negative (-).

There are 50 cards in total. 10 of each type, 5 of each state.

Each player is dealt 5 random cards. You can have a maximum of 5 cards in your hand.

Your goal is to build one of the following personalities.

On the playing field, each player has their own space to build a personality.

On your turn, you have three options.

1) Lay a card in your pile to build your own personality, then draw a card from the deck.

2) Lay a card on top of an opponent’s card, within their pile, to disrupt them from building their personality. Then draw a card from the deck.

3) Draw a card from the deck while discarding a card from your hand to the bottom of the deck.

Primary Research

I started by doing a few casual interviews with friends and peers, to gain further insight on how people think and feel about personality tests. There were a lot of thoughts! I found several key insights for developing my approach.

1) People don't often feel they learned anything new after taking one

2) The results are either too obvious to be useful or not clear enough to make an impression

3) People aren't sure what to do with the information

Secondary Research

First, I need to understand how our personalities can be split into separate measurable traits. My research landed on the system used by IPIP, the International Personality Item Pool. It is a scientific collaboratory for the development of advanced measures of personality and other individual differences. All information is in the public domain, free to be used for any purpose.

They have a variety of testing items and scales to use here, with recommendations on how to administer and construct your own test. For my purposes, I used a test that would align best with a standard deck of cards, so that I could reference similar uses for how I would construct this game. I chose to adapt the 50-item Sample Questionnaire.

This test is based on “The Big Five” which constructs a personality based on 5 categories.

In a 50-item test, the recipient is given statements that are keyed as positively (+) or negatively (-) weighted toward one of these traits. For example, one statement would be the following.

Which the test writer would grade with one of five options.

This is completed for all 50 items. Then collectively, we can see how a test writer performed in each category, and how that compares to other people that have taken the test. This is how “which Game of Thrones character are you?” and other tests are often constructed. People imagine how these characters might answer the questions, and use the results to generate a personality metric.


With this initial framework in mind, I started designing how this system could be represented with cards. I decided on putting each statement on a card, and using each of the “The Big Five” categories as a suit, the way Hearts or Spades are used. I also settled on simplifying the weighting system, so that each statement is fully weighted to its category, meaning they’re defined with a response of “Very Accurate.” I wanted to reduce the number of systems at play initially, and could later revisit this element of the system.

Concept Testing

With this representation, I started playing with how these cards could function together as a deck. Continuing on the standard playing card model, I experimented with different common game systems. How might they function in a game like poker? How about a game like Go Fish? A number of play styles could serve as the basis, but in adapting these models, it is essential to consider whether they function to educate the player on personality traits or obscure the knowledge.

I experimented with the idea of players using the whole deck to perform a personality test, but this did not educate the player any differently than taking the original IPIP test. Introducing multiple players only made this function worse, as it introduced a social pressure to pursue an ideal personality.

The most natural system came from randomly dispersing the cards to several players, then seeing how these could stack together. But at this phasethere was no obvious win scenario to this system. If players were not defining their own traits, what was the motivation to move through these personality traits? There had to be a variety of card configurations to pursue. This would require further research.


Players could use these cards to define a personality, but without a framework to pursue, it was unclear what it really meant and why they should move through the system. What do these personalities mean? My research led me to a recent study that answered exactly these questions.

“A new study has sifted through some of the largest online data sets of personality quizzes and identified four distinct “types” therein. The new methodology used for this study is rigorous and replicable, which could help move personality typing analysis out of the dubious self-help section in your local bookstore and into serious scientific journals.”

Our personalities can be loosely fitted to one of several main groups.

These configurations showed the value of the personality test responses and gave my system the win scenario it needed. The next thing I would need to solve is how to properly integrate these groups.


My first phase of testing and ideation taught me how I could adapt IPIP test questions into playing cards. With further research, I found the basis for a system that was a better fit for a social game. I nowneeded to adapt each of the 4 personalities to my 50 card system. How many cards would it take to create any of the above personalities?

With each of these, I also considered several modes of play tested in a 2 player scenario. These were:

1) Players working together toward constructing one personality

2) Players constructing their own personalities competitively

3) Players constructing their one personality competitively.

User Testing

While testing modes of play, I used varying numbers of cards to define each personality. The results quickly showed me that the fewer cards I needed per category, the better. Requiring 5 categories to be established to win already took a lot of playtime, so requiring a larger number of states per category lengthened playtime beyond what users felt was an enjoyable round. To best reflect the relationship between each category and shorten playtime, I settled on 3 states.

Each attribute has a status of either positive (+), neutral (0), or negative (-).

To achieve the Average personality, 1 Extraversion + card, 1 Agreeableness + card, 1 Concientousness + card, and 1 Emotional Stability + card, would all work for their categories. However, you would require a neutral state for Intellect/Imagination. There were no statements that were balanced neutrally, so the requirement for a neutral was decided as one positive and one negative card, as seen below.

User Testing

First Mode of Play

In testing my modes of play, I did not find an interesting structure for my first model, direct collaboration between players. Players could work together to lay out the cards necessary for a personality, but then what? There were no barriers, and it made the journey through the cards too simplistic and not particularly engaging.

Second Mode of Play

For the second mode of play, players were given 5 random statement cards each, then randomly assigned a personality to construct. Whenever they put down a card to towards constructing their given personality, they could withdraw another statement card from the deck. This is would complete their turn, alternating to the next player. Whichever player constructed their personality first would be the winner.

Third Mode of Play

This mode showed some similar issues to the first, so after initial testing I added the ability for players to lay cards on top of the personality their opponent is building. This way, players could disrupt the opponents progress as they attempt to build their assigned personality. This made a significant difference, and meant the second mode of play was far more successful then the first. There was more interaction between players and there was now room for players to strategize how they use their cards.

It also became evident that players could get stuck when they did not want to lay any cards. For this, a rule was created allowing players to discard a card to the bottom of the deck, and take one card from the top. This would complete their turn.

For the third mode of play, the rules were similar to the second except that players used the same playing field to construct their personality, as shown below

There were two large benefits to consolidating the playing field. First, it meant players had less to keep track of, which helped them figure out the game much quicker. Second, there was a much greater need to strategize, since every move you make affected the outcome for the other player.

The issue that arose during this mode of play came with randomly assigning a personality to each player. Scenarios would arise where a players hand and their assigned personality was far more difficult then their competitor. In these scenarios, players could only try to mess things up for the other player, and the game would reach a stalemate.

User Testing Results

The results showed that the second and third models were the most engaging for the player, though some adjustment would need to be made. The added element of strategy in the third model was favoured over the second, so development of this mode of play was continued, with another round of ideation to solve the shortcomings that arose during testing.


There needed to be more flexibility in how players could win, so the idea of randomly assigning these personalities at the start of a game was removed. A new method was tested with players able to win by constructing any of the personalities. The idea was derived from Poker, in that every player knows what hands they can construct, and will often change approach mid game.

This changed the dynamic significantly, making the game function more like Tic Tac Toe or Connect Four. Players would have to change their approach often, having to counter their opponent while also searching for a way to land the card that would give them the winning turn.


Work in Progress! I have a number of things to consider here. This will be updated soon.

Feel free to contact me by email or LinkedIn. I'm always open to new opportunities and I'd love to hear what you're doing.