A single brick, no matter how well-made, isn't a building.
In an age when a single headline can reshape habits and a cousin's forwarded link can feel like settled truth, science educator Jeffrey Lee offers a quieter reminder: knowledge is not a moment but a process. A single study, however peer-reviewed and official in appearance, is one brick in a structure that takes years and many hands to build. The question is not whether to trust science, but whether we have learned to read its pace — incremental, self-correcting, and far more humble than the headlines suggest.
- Every week, new studies land in newsfeeds with the force of revelation, and people remake their diets, habits, and beliefs on the strength of a single finding.
- The machinery of peer review slows misinformation but does not stop it — predatory journals, flawed designs, and the sheer noise of publication make discernment harder than ever.
- Scientists and educators are pushing back with a call for evidence literacy: understanding not just what a study says, but how many studies exist, how consistent they are, and who funded them.
- Meta-analyses and systematic reviews are emerging as the real arbiters of scientific confidence, synthesizing dozens of studies to reveal where the weight of evidence actually falls.
- The path forward runs through trusted intermediaries — trained clinicians, rigorous science journalists, and frameworks like the Trust Project — who can place new findings in the context of everything already known.
When a cousin calls to report that eggs are dangerous because he read a study, the question of whether to change your breakfast routine opens into something far larger: how do we actually know what we know?
Jeffrey Lee, a geographer and science educator, has spent decades watching people treat a single published finding as settled fact. The confusion is understandable. Research gets published, headlines appear, social media amplifies, and a result begins to feel like truth. But science builds the way a structure does — one brick at a time, each piece checked against all the others. A single brick, however well-made, is not a building.
Peer review raises the odds that published work is worth reading, but it guarantees nothing. Honest errors slip through, experimental designs carry blind spots, and some journals will publish anything for a fee. More importantly, no single paper resolves a question permanently. What it does is add one piece to a larger puzzle. Confidence grows only when many independent teams, using different methods, arrive at the same conclusion. When findings contradict each other, confidence should weaken. Systematic reviews and meta-analyses exist precisely to take stock of this accumulated evidence — examining data from dozens of studies at once to reveal where the weight of understanding actually rests.
The volume of research matters too. Thousands of studies examine the causes of lung cancer; perhaps one or two touch on how specific genes affect hair loss. Scientific certainty about the former dwarfs certainty about the latter, and recognizing that difference is as important as understanding any individual result.
This is where expertise becomes indispensable, even as the concept has lost standing in some quarters. An expert is not someone who spent an afternoon online — it is someone immersed in a field for years, trained to spot methodological weakness, and aware of their own biases. Most people lack direct access to such specialists, but healthcare professionals and science journalists who understand how evidence accumulates can serve as reliable guides. Red flags are worth watching for: findings that sound too good to be true usually are, and a study praising candy funded by a candy company deserves immediate skepticism.
Lee is not asking everyone to conduct a literature review before dinner. He is asking for something more modest — a little humility about the limits of individual knowledge, and a realistic sense of what the scientific community, across years and many voices, has actually come to understand.
Your cousin calls with news from the internet: eggs are bad for you, he read a study. Should you listen? Should you change what you eat? The question seems simple until you realize it points to something much larger—how we actually know what we know, and why a single piece of research, no matter how official it looks, almost never tells the whole story.
A geographer and longtime science educator named Jeffrey Lee has spent decades watching people mistake one study for settled fact. The problem isn't hard to understand. Research gets published, headlines appear, social media amplifies, and suddenly a finding feels like truth. But science doesn't work that way. It works the way a building gets constructed—one brick at a time, checked and rechecked, fitted against all the other bricks that came before. A single brick, no matter how well-made, isn't a building.
When researchers publish a study, they've typically done one of two things: filled a gap in what we know, or tested whether an existing theory holds up under scrutiny. They design an experiment or collect field data, then submit their work to a peer-reviewed journal. Other experts read it, poke holes in it, and decide whether it's solid enough to publish. This process isn't perfect. Peer review doesn't guarantee a conclusion is right—honest mistakes happen, experimental designs have blind spots, and rarely, fraud occurs. Some journals are essentially paper mills, publishing anything if the author pays a fee. But peer review does raise the odds that what gets published is worth reading.
Here's the crucial part: no single paper solves anything permanently. It doesn't erase previous research, and it doesn't settle the question. What it does is add one piece to a much larger puzzle. The real work happens next, when scientists ask whether this new finding lines up with other research on the same topic. If many peer-reviewed studies, conducted by different teams using different methods, all point toward the same conclusion, confidence grows. If they contradict each other, confidence should weaken. Sometimes researchers compile these comparisons into systematic reviews or use statistical techniques called meta-analysis to examine data from dozens of studies at once. The more good evidence testing an idea, the stronger the foundation.
The number of studies matters too. There are thousands examining what causes lung cancer. There might be one or two on how specific genes affect hair loss. Scientists' certainty about lung cancer is vastly greater than their certainty about those genes. Understanding the strength of the evidence—how much research exists, how consistent the findings are, how rigorous the methods were—matters as much as understanding the evidence itself.
This is where expertise becomes essential, even though the very idea of expertise has fallen out of favor in some circles. An expert isn't someone who spent an afternoon browsing the internet. It's someone immersed in a topic for years, trained to spot weaknesses in studies, ideally someone who has conducted research themselves. They know which findings are likely to be wrong. They also know their own biases—the human tendency to accept what you like and reject what you don't. Most people don't have direct access to such experts, but healthcare professionals with years of training and a requirement to stay current with their field's literature can help. So can journalists who understand how science actually works, who can report accurately on new research and place it in context with what else is known. The Trust Project offers guidance on finding trustworthy reporting. But watch for red flags: if someone is selling you something that sounds too good to be true, it probably is. If a study praising candy was funded by a candy company, that's a major warning sign.
Lee isn't suggesting everyone needs to conduct a full literature review before discussing science or deciding what to eat. But he's asking for something more modest: a little humility about the limits of your own knowledge and a realistic sense of what the scientific community actually understands. Don't remake your life based on one study, even if your cousin insists. Science builds slowly, across years and many voices, each study a small contribution to something larger than itself.
Notable Quotes
Scientific knowledge is constantly being refined as new information comes to light.— Jeffrey A. Lee, geographer and science educator
Don't make life-altering decisions based on an article describing one scientific study.— Jeffrey A. Lee
The Hearth Conversation Another angle on the story
Why does one study feel so convincing when we read about it, even though you're saying it's not the whole picture?
Because it's presented as news—as something new and definitive. A headline says "Study Shows X" and our brains treat that like a fact. But a study is really just one person or team asking one question in one way. It's a data point, not a conclusion.
So when should someone actually change their behavior based on research?
When the evidence has accumulated across many studies, when experts in that field agree, and when the findings have held up over time. That takes years, sometimes decades. It's not dramatic, but it's how we know what we actually know.
What about the peer review process? Doesn't that guarantee a study is correct?
It raises the odds, but no. Peer review catches obvious problems and filters out the worst work. But honest mistakes slip through. Experimental designs have blind spots. And peer review is only as good as the journal doing it—some journals will publish almost anything if you pay them.
How do you know which experts to trust?
Look for people who've spent years in the field, who update their views as evidence changes, and who don't have a financial stake in the answer. Be suspicious of anyone selling you something or pushing one particular conclusion too hard.
Is there a shortcut for people who don't have time to understand all this?
Find a good journalist or healthcare provider who understands how science works and can put new findings in context. But even that requires some judgment on your part about who's trustworthy.