In the pre-Internet world, content knowledge was king, i.e., students with the strongest memory stood out the most (think Jeopardy!). The internet changed that. While memory is still important, critical thinking, or the ability to make connections and solve problems, determines who rises to the top. In an AI-driven world, we are likely to see yet another shift. Reasoning and recall still matter, but it is flexible thinking, or the ability to take feedback, adapt, and see beyond what AI chatbots like Claude can provide, that will determine exceptionality. And it’s not coincidental that this is exactly where Claude could arguably use some support.
We evaluated Claude through the lens of MindPrint, a cognitive assessment normally used to map students’ learning strengths across memory, reasoning, attention, and more. The results show where student thinking still outperforms AI, and which skills educators should prioritize to help students thrive in an AI-driven world.
Claude’s MindPrint Profile
Strengths:
- Memory: No surprise here. Claude showed an exceptional ability to remember and retrieve information on any topic or format. No human can come close. Since foundational knowledge is key to Reasoning, this is a key strength for Claude.
- Complex Reasoning: Claude did very well on all three skills, but with notable variability. Claude was exceptional at manipulating and solving Visual-Spatial problems beyond most human capabilities. Claude’s Verbal Reasoning was also very strong – with a well-structured prompt, Claude can answer almost any question correctly. But Claude sometimes falters on more nuanced challenges, suggesting that Claude has some ways to go in Abstract Reasoning.
- Working Memory: With the free version, Claude started to forget details we agreed upon after a few steps. Claude’s paid version did better but wasn’t perfect. In short, Claude’s short-term memory is very good, but human checks of Claude’s work on multi-step projects remain essential.
- Processing Speed: Undoubtedly, Claude is faster than any human at solving complex problems, but that doesn’t mean that Claude is always most efficient. For straightforward questions, a quick search and maybe your own brain are still faster and better.
Skills to Support:
Like any student, Claude showed variability in skills, giving a sense of where and when Claude is no match for a gifted student
- Attention: Claude was surprisingly sloppy at keeping track of details in a multi-step project. While Claude apologized for making mistakes, Claude’s lack of sustained focus could be a source of frustration and errors if you aren’t careful.
- Flexible Thinking: While Claude’s Flexible Thinking was good, it was undoubtedly Claude’s lowest skill. Claude was unable to pivot without explicit direction, suggesting clear limitations on creative problem-solving without a flexible human to guide him.
So, what does this all mean?
Claude’s results, like a student’s MindPrint profile, offer a map of strengths and areas to support, not a single score. Claude is a powerful assistant for building content knowledge and even tackling complex reasoning. But it still needs a human in the loop, especially when it comes to staying on track over multiple steps or adapting when the first approach doesn’t quite work. Those are exactly the skills we should be helping students develop; the ones no chatbot can replace.