Rethinking Intelligence in Academia: Are Our Metrics Outdated?

As an extracellular vesicle (EV) researcher, it’s literally my job – and, I’ll admit, a serotonin boost – to keep learning and reading widely. Over the years, I’ve realised that how I think about life often mirrors how I think about my work. Studying EVs as potential biomarkers, a field both innovative and full of caveats, has made me reflect on the idea of “markers” themselves – especially the ones we use to define intelligence.

We like to believe that academia is where intelligence is most rigorously defined and fairly evaluated. Yet if we look closely, many of the traits we celebrate as “markers of intelligence” are not only narrow—they’re increasingly obsolete.

For decades, certain signals have been consistently rewarded: speaking fluently in jargon, responding quickly under pressure, publishing frequently, affiliating with prestigious institutions, confidently defending established ideas, or asking endless questions in a seminar. These have become shorthand for brilliance – but the problem is, they are often just that: shorthand.

Take fluency. The ability to speak smoothly and authoritatively is often mistaken for depth of understanding. In reality, it can reflect familiarity, rehearsal, or simple confidence. Some of the most original thinkers – especially early in their careers – struggle to package ideas neatly. Their thinking is nonlinear, evolving, and hard to compress into polished soundbites. Yet academia tends to reward those who sound certain, not those still exploring.

Speed is another deeply ingrained metric. The person who answers first in a seminar or fires back instantly in discussion is often perceived as sharp. But thoughtful science is rarely fast. Good ideas take time—to wrestle with discomfort, to challenge assumptions, to mature. Does responding first truly mean deeper understanding, or just greater confidence? Are we mistaking quickness for insight?

Then there’s productivity—the sacred triad of papers, grants, and citations. While output matters, equating volume with contribution is a dangerous simplification. It rewards incremental work and predictable outcomes. High-risk, high-reward ideas—the kind that actually move fields forward–take longer, fail more often, and rarely fit neatly into annual performance metrics. So we must ask: does publishing more really mean being smarter, or simply more adept at playing the game?

Prestige also casts a long shadow. Association with renowned institutions, labs, or supervisors often shapes perceptions of intelligence before one’s work is even read. It’s far from neutral—it amplifies certain voices while quietly filtering out others.

And confidence: academia loves it. Those who assert ideas boldly are often seen as more capable. Yet confidence is not always correctness – it's temperament, training, or privilege. Meanwhile, those who hesitate, who admit uncertainty, or who say “I don’t know yet,” may be engaging in the most honest form of thinking–and still be overlooked.

I’ll admit something personal: I’ve felt imposter syndrome for more than a decade – and I suspect many readers have too. But perhaps the problem isn’t that we fall short of some objective standard. Perhaps it’s that the standards themselves are designed more to preserve an aura of exclusivity than to assess real intellectual contribution. Are these metrics genuine measures of intellect, or just relics sustaining the idea of the “academic elite”?

Consider a more playful personal example. I once prided myself on recalling every step of the Krebs cycle or solving problems instantly on the board. Memorising details felt like mastery; scoring high felt like belonging to the “academic elite". But looking back, I wonder: did knowing the exact NADH count make me a better thinker, or just better at following rules?

All these markers share a common feature: they’re easy to observe and measure. And that convenience is exactly why they persist.

The contrast becomes even starker when we look beyond academia. Technology evolves at breakneck speed; artificial intelligence reshapes knowledge itself. Information once requiring years to master can now be accessed in seconds. Yet our evaluation systems remain anchored to pre-digital paradigms. Much like the BMI index – a crude, context-blind measure of health – our current metrics of intelligence ignore creativity, nuance, and adaptability.

Consider AI: it can process vast information instantly, yet the quality of its output depends entirely on human guidance—on how we prompt, interpret, and connect ideas. Two people using the same model can produce entirely different results. Isn’t that what real intelligence looks like? Not memory, but synthesis; not recall, but creativity.

Biological and technological evolution alike reward adaptability, not rigidity. Yet academia often prizes consistency over curiosity, conformity over imagination. Those may ensure survival within the system but not progress.

True intelligence is harder to measure. It’s visible in how someone navigates uncertainty, learns from failure, connects distant ideas, and persists with curiosity. It’s nonlinear, messy, unquantifiable – and precisely what drives scientific innovation.

This disconnect isn’t abstract – it has human costs. Who feels welcome in academia? Whose ideas are nurtured, and whose are dismissed? How many breakthroughs never happen because the system rewards speed and confidence over reflection and originality?

In EV research, we rely on intensive screening, characterisation, and validation before confidently calling anything a biomarker. Even then, we know our capture methods—whether antibody-based or bead-based—never recover every vesicle population. The beads bind what they can, but some subtypes slip through, hidden from detection. That doesn’t make them any less real; it just means our tools are imperfect.

And isn’t that the same with intelligence? We keep trying to capture it through narrow, convenient markers—fluency, speed, confidence, productivity – but inevitably, we miss entire “subpopulations” of brilliance that our metrics aren’t built to detect. Just because something isn’t captured doesn’t mean it doesn’t exist.

The challenge, then, is simple but profound: are we still judging intelligence with tools built for a world that no longer exists? Because, as evolution, technology, and history have shown, progress depends not on predictability but on courage, creativity, and the willingness to rethink what we think we know.

So here’s the question I leave you with: if our measures no longer capture what truly matters, how much brilliance are we failing to see?