How We Discuss — Lab Handbook

This is the single thing I consider most important for our whole team’s growth.

Feedback is there to make us stronger, not to prove who’s right

We encourage direct, honest questioning of each other’s ideas, methods, and results — research needs exactly that kind of collision. But what we question is the content, not the person.

So in discussion we keep the focus on:

Where might this idea or method have a problem?
Under what conditions does this conclusion hold?
Which piece of evidence or reasoning is still missing?

Rather than:

How could you think that?
Did you get this wrong?
You’re not being careful enough.

We want everyone to see feedback as a form of collaboration: we’re not grading each other, we’re trying to see the problem clearly together.

It’s fine not to know; bluffing is what doesn’t help

When you’re asked about something you’re not sure of, it’s completely fine to say “I’m not sure, but I’ll look into it” — and it’s usually more valuable than bluffing.

If you can go one step further and propose the next action, even better. For example:

Let me go build a toy example and see.
I need to check the related literature again.
This direction may be wrong; I’d like to try another angle.
I think there’s a counterexample here — I want to confirm it first.

For me, the best state isn’t “being right every time.” It’s being able to point out honestly where you’re still unsure, and then knowing what to do next.

Likewise, if you notice I’ve said something wrong, say so directly. The earlier you point it out, the more time it saves everyone.

Understand before judging; ask before concluding

When you receive feedback, you don’t have to accept it immediately — but you usually should first work out what the other person actually means.

A lot of feedback falls into different categories:

Some is recognition — letting you know what you did right.
Some is guidance — telling you how to do it better.
Some is evaluation — telling you roughly where you currently stand.

In discussion, we want to keep these three as separate as possible. Especially when an opinion or piece of feedback sounds heavy, don’t rush to read it as a rejection of you as a whole. First try to tell them apart: is this about the work, the method, the progress — or is it really about you as a person?

If a piece of feedback leaves you stuck, the best move usually isn’t to push back immediately, but to ask the question clearly first. For example:

Do you mean the math here isn’t complete enough, or that the experiments aren’t strong enough?
What matters most to you — novelty, clarity, or scope of applicability?

That usually helps everyone move forward more than arguing directly.

The same principle applies when you’re the one giving feedback.

Don’t rush to state your objection or rebuttal. First lay out your current understanding — for example, “If I understand correctly, you mean…” — to confirm you’re both discussing the same thing, and then raise your question or suggestion. Many apparent disagreements are just differences in understanding.

Also, try to avoid jumping straight to a verdict like “this has no novelty” or “this doesn’t make sense.” Instead, be specific about what gave you that impression. For example: “I don’t yet understand the difference between this and existing methods,” or “does this assumption need more evidence to support it?” Feedback like that is usually much easier to act on.

Finally, good feedback should focus on the work, the ideas, and the argument — not on the person. The goal is to help each other do better research, not to prove who’s right.

Newcomers, please speak up

In discussions, we don’t want people who’ve just joined to only ever listen. Even if your idea isn’t fully formed, you can still say what you understand, what you’re unsure about, or where something feels off to you.

Being wrong is fine; staying silent too long is the bigger waste — because many gaps in understanding only surface once they’re said out loud.

I care a lot about everyone practicing independent thinking and being able to state their ideas clearly. This doesn’t mean saying it perfectly from the start; it means gradually building the habit, in discussion, of daring to ask, daring to guess, and daring to correct yourself.

Harsher feedback in private; in public, mostly questions, recognition, and suggestions

I hope this situation comes up very rarely, but when feedback needs to be more direct and harder to hear, I’ll usually choose to talk one-on-one.

In public settings, I’d rather feedback took mainly these forms:

Questions: helping the other person articulate their idea clearly
Recognition: pointing out what was done well
Gentle suggestions: noting how something might be improved

This isn’t because criticism is off-limits. It’s because I want feedback to actually be absorbed, rather than putting someone into a defensive state first. Very often, the most valuable feedback isn’t the heaviest or sharpest remark — it’s the one that best helps the other person see the problem and find the next step.