I built Mind to explore the everyday patterns behind how we describe ourselves. Then a few people tried it, and their answers made me rethink the questions.

Have you ever read a sentence in a personality-style quiz and thought, “Yes, that sounds like me”?

Easy enough. You choose the highest option and move on.

But wait. Does recognising yourself in a sentence mean it describes you almost every time? Or did it remind you of one moment, one interest, or the person you would like to be?

That question became surprisingly important while I was building Mind, or Zihin Atlası.

Almost everything was lighting up

The first version had 45 questions, with five answers ranging from “doesn't fit me” to “fits me completely.” I asked a few people around me to try it.

Something kept happening: if a statement felt familiar, some people went straight to the highest option. Before long, their maps were spread across almost every area.

A broad profile can be perfectly reasonable. Someone may genuinely have many interests. Still, I wanted to understand what those answers were telling us.

Calling this dishonesty would be a leap. Perhaps the wording was inviting people to describe an identity when I really wanted them to remember a behaviour.

“I like solving problems” leaves a lot of room. I might enjoy the idea of being analytical. I might also have solved one difficult problem last Tuesday. Neither tells you how often I actually stop, compare explanations, and check my reasoning.

What did you actually do?

So I changed the conversation.

Mind now asks about the last four weeks, and about what happened when a suitable opportunity came up. For example:

Before accepting a claim, did you compare information supporting it with information against it?

You answer with a frequency: never, rarely, sometimes, often, or almost every time. There is also a separate option for having no opportunity or being unsure.

If you remember doing something in roughly half the relevant situations, “sometimes” makes sense. The highest option asks you to think about almost all those situations.

Will clearer wording eliminate every response bias? I wouldn't assume that. It gives us a more specific conversation to work with, and that is something we can examine and improve.

Your answer opens a different path

This is where the software became more interesting.

Every area begins with three everyday behaviour questions. From there, your answers determine whether another example would help.

  • No example comes to mind? A more ordinary situation opens up. If you still have no example, the gap stays visible.

  • All three behaviours happen often? Mind asks for a more concrete situation. Your response can lead to another context or a more detailed example.

  • The frequencies differ substantially? Follow-up questions explore the circumstances. Acting differently at work and at home can make sense.

  • No follow-up is needed? The journey moves to the next area.

The current journey takes between 30 and 57 stages. There is a limit of three follow-ups per area, because reflection can also become tiring.

Three visual choice stops break up the reading. You choose an activity you might like to try. These choices shape the suggestions at the end; they do not secretly add points to an ability score.

You can inspect the branching map inside Mind. The rules run in the browser. The current version needs no AI service to choose its next step.

Where the science enters the conversation

The vocabulary behind the project comes from Howard Gardner's multiple intelligences framework. Mind explores linguistic, logical, spatial, musical, bodily, naturalistic, interpersonal and intrapersonal areas. It also includes existential reflection, clearly marked as a proposed additional area.

Gardner's overview of the framework is useful background. His explanation of why multiple intelligences and learning styles are different ideas also matters. Enjoying diagrams doesn't establish that you can only learn visually.

There are real scientific questions around the framework. A 2021 review by Ferrero, Vadillo and León found methodological problems in studies of multiple-intelligences-based classroom interventions that prevented a sound evaluation of their effectiveness.

That review didn't test Mind. I take it as a reason to be careful about what research can support.

The same care applies to adapting questions. A study by Haley and colleagues found advantages for adaptive testing using a calibrated physical-functioning question bank. That setting and measurement model are very different from this project.

Mind's questions and branching rules are my own design. They have not yet been validated as a psychological measurement instrument. A working decision tree doesn't establish scientific measurement accuracy.

A map with room for missing pieces

What happens if you've had no chance to try a particular behaviour?

Mind leaves that answer out of the numerical calculation. Each area has three behaviour headings, given equal weight. If a heading still has no usable example after the follow-ups, that area's score is left blank.

Extra questions therefore don't give one heading more weight just because it received more attention. Context differences are shown, and a genuinely broad profile is allowed to remain broad.

The 0–100 values summarise your reported behaviour frequencies. They don't tell you your IQ, how rare you are, or which career you should choose.

To make stronger claims, I would need actual work on validity, reliability and how questions behave across different groups. The AERA, APA and NCME testing standards are part of that background.

For now, I want the result to help you ask a useful next question: “Where does this show up in my life, and what would I like to explore?”

Try it with one ordinary month in mind

The live application is currently in Turkish. You can open Mind here, follow your own path, and inspect the method from the scientific-framework link in the footer.

Your answers and profile stay in your browser's local storage. You can return to the journey, export a PNG, print a PDF, or clear the local records from the privacy screen.

When you try it, think about ordinary situations. A message you rewrote. A claim you checked. A rhythm you noticed. A conversation where you paused to understand someone.

I am interested in how those small examples connect. And if the result feels incomplete, that is useful feedback for the project too.

What would we notice about ourselves if we gave the question a little more attention before choosing the answer?

Follow the sources