The Four Intelligences in Practice: Questions to Ask This Week
A framework is only useful once it turns into something you actually say out loud at dinner. Here is what that looks like across four ordinary weeks.
A framework can sit on a page indefinitely without changing anything about a Tuesday evening. The four intelligences, rational, embodied, relational, and existential, are useful as a map of what AI cannot replicate, but a map only matters once someone actually walks the route it describes. This is the practical half of that idea: what it looks like to ask about each intelligence in an ordinary week, not as a program, just as a habit of attention.
AI is strong in exactly one of the four. Most households, without deciding to, spend most of their attention developing that same one, because it maps most directly onto school performance and is the easiest to measure. The other three do not announce themselves the way a test score does, which means they need to be asked about on purpose or they quietly go unattended.
Rational: the one AI is already good at
Rational intelligence, thinking, problem-solving, strategic reasoning, is where AI performs strongest, which is exactly why it cannot be the only intelligence a child relies on. The goal here is not to abandon rational thinking. It is to make sure a child practices it themselves rather than outsourcing it by default, and to build the specific habit of treating an AI answer as a draft rather than a verdict.
A useful weekly question: what is one thing today that did not go the way you expected? It trains the habit of noticing the gap between expectation and reality, which is the raw material of actual reasoning, not just retrieval. A version for slightly older children or a classroom: what is one answer AI gave you today that you did not check before using it? That question does not assume AI is wrong. It assumes checking is a habit worth having regardless.
Embodied: the wisdom no algorithm has access to
Embodied intelligence is physical awareness, sensory experience, the accumulated wisdom of a body that has adapted to its actual environment. It has no digital equivalent, and it atrophies specifically through disuse, not through any dramatic event, just fewer physical challenges attempted because a screen offered an easier alternative.
A simple weekly question: where in your day did you feel relaxed, and where did you feel tense? This trains the habit of locating feeling in the body rather than only naming it abstractly, a foundational skill for everything from stress management to physical safety. A classroom or older-child version: what did you work out today before checking AI or a search engine? It rewards the attempt, not just the correct outcome.
Relational: what a real explanation gives that an AI answer cannot
Relational intelligence is social connection, emotional depth, the ethical reasoning that comes from actually being in relationship with other people rather than processing information about them. It is built through practiced attention to other people's experience, which is a habit, not a personality trait, and it is one of the easiest habits to let slide in a busy week.
Ask: did you notice anyone who needed a hand today? The question trains attention outward and, as a side effect, tells you something real about your child's world that a school report never will. For a classroom setting: who explained something to you today in a way AI could not have? It surfaces what a human explanation actually offers, patience calibrated to a specific person, context an algorithm was never given, care that shows in the explaining itself.
Existential: the questions worth sitting with, not resolving
Existential intelligence covers meaning-making, the sense of time and consequence, the questions a child needs practice sitting with rather than having answered for them. This is the intelligence most likely to be skipped entirely in a results-oriented week, because it produces nothing measurable and its value only becomes visible much later.
Ask: if you could change one thing about today, what would it be, and why? It trains the habit of naming what actually matters, not just recounting what happened. A classroom version: if AI could do this assignment perfectly, what would still be worth doing yourself? That question does not need an answer handed back immediately. Sitting with it is the point.
Two things that matter more than any single question
Let your child sit with difficulty before you rescue them. And let them watch you question AI's answers instead of trusting them by default, because a child absorbs the habit of checking from watching it modeled far more reliably than from being told to do it.
- Rational: What is one thing today that did not go the way you expected?
- Embodied: Where in your day did you feel relaxed, and where did you feel tense?
- Relational: Did you notice anyone who needed a hand today?
- Existential: If you could change one thing about today, what would it be, and why?
None of these questions need a program around them. They need to be asked, genuinely, often enough to become expected rather than novel, and answered honestly, including by the adult asking them. That repetition, not any single conversation, is what actually builds the four intelligences that AI has no way of replacing.
What to do when a question falls flat
Some evenings, a question just does not land. A tired child answers with a shrug, or gives the same flat response three nights in a row. That is not a sign the question is wrong, it is closer to normal, and the temptation to abandon the habit after a few quiet nights is exactly the moment it is worth pushing through instead, because consistency is what turns a question from a novelty into an expectation a child eventually starts answering more honestly, once the pattern has proven itself safe and unchanging.
It also helps to answer the questions yourself, out loud, unprompted, before or after asking them. A child who hears an adult name a moment of feeling tense, or admit to not questioning something AI told them, learns far more from that example than from any explanation of why questioning matters. The habit spreads through being modeled, not through being explained.
None of these four questions are exclusive to any one age. A five-year-old can answer where in your day did you feel tense in their own limited vocabulary, and a sixteen-year-old can answer the same question with real depth. What changes with age is not whether to ask, but how much silence to leave after asking, since older children generally need more room to arrive at an honest answer rather than a fast, easy one designed to end the conversation.