Writing Slowly, Reading Fast
What I’ve learned from comments on this blog
After 6 months of writing this blog, I’ve learned something mildly uncomfortable: when I comment on other people’s work, I don’t always read as carefully as I should. Because I feel so time pressured, I tend to skim. If the argument is going the way I like, confirmation bias kicks in and I use it to support my preconceived notions. If not, I often dismiss it. That is normal and human. But often, if I notice a phrase that connects to a debate I already know well, I anticipate where the author is going and begin composing my response before I even reach the end. More than once, I have argued with a position that was present more in my head than on the page. That recognition has changed how I think about comments on my writing.
When I write an essay, I move slowly. I revise sentences, qualify claims, and try to bound my statements so they are proportionate to the evidence. Above all, I try to be accurate and painfully precise. For medical topics, I often stress test my arguments by getting feedback from a friend or even by using Open Evidence before publishing.
I assume that people will read my essay in the same deliberate way that I constructed it. Sometimes they do, but sometimes they don’t. That mismatch is not about intelligence or good faith. It reflects the way most of us actually read when we are busy. We tend to read quickly because we don’t have time or energy to really focus, so we scan for signals. Certain words carry weight long before we complete and process the paragraph around them. Terms like calcium score, keto, statin, CCTA, screening, or the name of a well-known podcaster function as cognitive shortcuts. They activate stored positions, prior debates, and often strong emotions. Once that activation occurs, the brain supplies a familiar script.
The result is that we sometimes substitute the debate we already know for the point that actually is being made. A contextual reference to a research trial is interpreted as advocacy. An illustrative example is heard as a policy recommendation. A discussion about clinical reasoning gets reframed as a contest between two procedures. The shift is efficient and can be invisible to the reader.
Professional identity adds another layer. In medicine, frameworks are not abstract ideas. They are tools that have guided real decisions and, in many cases, saved lives. When a framework is questioned or reframed, it can feel less like an academic disagreement and more like a challenge to years of training and practice. That reaction is normal and I recognize it in myself as well.
At the same time, writers overestimate how carefully they will be read. I assume that the architecture of my essay will be followed from beginning to end. In reality, most readers encounter only fragments of it. Some arrive through a shared excerpt. A carefully bounded claim can lose its guardrails when encountered in isolation, and sometimes I may not have been as clear as I could have been.
This dynamic extends beyond comment sections on social media. In clinical medicine we rely heavily on pattern recognition, which allows us to move efficiently through complex information. But the same efficiency can lead us to answer a question adjacent to the one being asked. We fill in gaps with what is most familiar. The habit is adaptive, but it clearly is fallible.
I comment much less on any social media than I ever used to, but when I feel the urge to respond quickly, I pause and ask myself whether I can restate the author’s central claim in neutral terms. If I cannot, I read it again or I just move on. As a writer, when someone interprets my work differently than I intended, I try to clarify the original claim rather than escalate the debate that has been triggered. Sometimes the reader misinterpreted my thesis. Sometimes the clarification reveals that I was unclear. Sometimes it reveals a genuine disagreement. Clarification and genuine disagreement are preferable to arguing past one another.
None of us can eliminate our biases or our prior assumptions. They are part of our expertise and our experiences. What we can do is notice when a familiar word has taken control of the conversation before we have understood the paragraph in front of us. Writing and reading never will align perfectly, so all I can do is try to be as clear as possible about what I am writing and answering. Having said all that, I prefer Substack to any other social media platform that I have been part of, including Facebook, Twitter, Bluesky, Threads, Mastodon, and others. The comments here tend to be more on target, less reflexive, and more genuinely inquisitive. I appreciate that.



Read twice, comment once. Not the other way around.
I look forward to your posts each week and appreciate the time you take to share your thoughts and knowledge about your field. Thank you!