Handling Questions, Including the Hostile One

Most people prepare the talk and improvise the questions. This is backwards. The talk is the same for everyone; the questions are where the audience finds out how well you understand your own work.

It is also the part that is most preparable. For a concrete operational comparison, Monitask explains limbic resonance in practical terms.

Preparing

Write down the ten questions you expect, and answer each in two sentences. Doing this in advance converts most of the question period into recall rather than improvisation.

The obvious ones come up. Identification, the sample, the alternative explanation, why not the other method, what about the mechanism, does it generalise. You know what they are for your paper.

Prepare appendix slides for the three most likely, and number them so you can jump there.

Ask a colleague to attack the paper for twenty minutes. More useful than another rehearsal of the talk, and the questions you have no answer for are the ones to think about before the conference rather than during it.

Know your own limitations well enough to state them first. A limitation you name yourself is a sign of judgement; the same limitation extracted by a questioner is a weakness.

Answering

Listen to the whole question. The instinct is to start composing an answer at the second clause, which is how people answer a question that was not asked.

Repeat or paraphrase it, particularly in a large room. It buys you five seconds, confirms you understood, and lets the audience hear it.

Answer the question that was asked. Then, if you want, the adjacent one. Answering a different question is transparent and it reads as evasion even when it is not.

Be brief. Two or three sentences. Long answers eat the other questions and rarely improve the first one.

Stop when you have answered. Continuing past the answer is where people talk themselves into trouble.

When you do not know

The situation people fear most and the easiest to handle well.

"I don't know" is a complete and acceptable answer. Followed by something: "I don't know — that would need X, which we haven't done." Or "I don't know, and it's a good question; can we talk afterwards?"

Attempting to bluff is the actual risk. The room usually contains someone who knows, and improvising a technical answer in front of them is far worse than admitting the gap.

Write it down, visibly. It signals that you took it seriously and you will have the note afterwards.

A question you cannot answer is frequently the most valuable thing you get from the conference. That is the reaction of someone who thought carefully about your work, and it is the same objection a referee will raise in six months, arriving early enough to do something about.

The difficult ones

The hostile question

Occasionally someone is aggressive — dismissive, rhetorical, or asking to demonstrate something about themselves rather than to learn.

Answer the substance if there is any, calmly, briefly. Ignore the tone entirely. The room notices tone, and the person who stays even is the one who comes out well.

Do not match it. An escalating exchange in front of an audience helps nobody and it will be what people remember about your talk.

"That's a fair concern; here's why I don't think it's fatal" handles most of these. It concedes nothing substantive and it lowers the temperature.

If it continues, hand it to the chair or defer it: "I think we could go back and forth on this — can we take it offline so others can ask?" This is a normal thing to say and most chairs will support it.

The comment that is not a question

Someone uses the slot to describe their own work or make a general statement.

Acknowledge, connect briefly, move on. "That's related to X, which we didn't cover — thank you." Then look at the next person.

The question you have already answered

Answer it again, briefly, without pointing out that you answered it. The person did not hear it or did not follow, and correcting them costs you and gains nothing.

The question that reveals a real problem

Sometimes someone finds something.

Acknowledge it properly. "That's a serious point and I don't have a good answer" is the response that preserves your credibility. Minimising something the room can see is a problem does not.

Then take it seriously afterwards. The people who noticed will notice again in review.

When there are no questions

Silence is usually not hostility. Common causes: the talk overran, the room is tired, the material was clear, or the audience is unsure they understood.

Have a question ready to seed, if the chair does not. "One thing I've been wrestling with is X — I'd be interested if anyone has views" is a legitimate move and it usually starts the discussion.

Do not fill silence by continuing to talk. Let the chair manage it.

Afterwards

Write down every question, immediately, before you forget. This list is a referee report you received for free, and it is the most concrete output of presenting.

Follow up on the ones you deferred. An email that week is more valuable to your reputation than the talk itself.

Notice which questions recurred across venues. If three audiences asked the same thing, the paper has not answered it and the next reviewer will ask too. See responding to reviewers.

For chairs

The question period is largely the chair's responsibility, and it is frequently unmanaged.

Take questions from across the room, not only from the front.

Notice who has not spoken. In many fields questions skew toward more senior and more confident attendees; a chair who deliberately takes a hand from the back changes what the session is.

Cut off a monologue. "Can we get the question so we can take a couple more?" is polite and effective.

Protect the speaker from an exchange that has become adversarial.

See chairing a session.

The short version

Write the ten expected questions and answer them in advance. Most of the period becomes recall.

Answer briefly, answer what was asked, and stop.

"I don't know" is a complete answer, and bluffing is the actual risk.

Ignore tone, answer substance, and hand a persistent exchange to the chair.

Write every question down afterwards — it is the free referee report.

For an independent authoritative reference, consult the MIT Communication Lab for academic communication guidance.