Ask the Second Question: What the Research Found About Follow-Ups and Second Dates
They mention, almost in passing, that they spent last spring learning to sail. You have a choice. You can say that sounds fun and move on to the next thing you had ready, or you can ask the second question: what made you want to learn? This shelf already gives that advice, most directly as the Follow-Up Rule in the first-date conversation guide and as “ask the second question” in the piece on responsiveness. What it has not shown is the evidence. There is a paper that measured this, and the most useful thing in it is not that asking works. It is that the people doing the asking did not expect it to.
The Finding
In 2017, Karen Huang, Michael Yeomans, Alison Wood Brooks, Julia Minson and Francesca Gino published a paper in the Journal of Personality and Social Psychology whose title is also its conclusion: It doesn't hurt to ask: Question-asking increases liking. The set-up was plain. Strangers were paired up to get to know each other, some of them were told to ask a lot of questions and some to ask only a few, and afterwards their partners said how much they liked them.
The people who asked more were better liked. The abstract adds the sharper part: more questions, “particularly follow-up questions”. A follow-up, in the paper's definition, is a question that invites the other person to say more about what they have just said, as opposed to one that moves to a new topic (page 432). The authors' explanation is responsiveness, which the abstract describes as listening, understanding, validation and care. People told to ask more questions were rated as more responsive, and in the authors' analysis, responsiveness is what carried the effect on liking.
What the finding is, in one sentence
In first conversations between strangers, the people who asked more questions, follow-up questions especially, were better liked by the person they were talking to, and the researchers traced the difference to how responsive the asker seemed.
From a Chat Window to a Speed Date
The experiments were typed conversations. Pairs of strangers chatted by instant message for fifteen minutes, first in a research lab, from separate cubicles and never face to face (page 433), and then online (pages 436 and 438). That leaves the obvious question: does any of it survive a real table and a real face?
For that, the authors turned to recordings from speed-dating events, collected for an earlier study by other researchers. Each dater went on a run of four-minute dates and said after each one whether they wanted to see that person again (page 440). There were too many conversations to sort every question by hand, so the authors trained a program on the questions they had already hand-sorted from the chats, which the abstract calls a “follow-up question detector”, and ran it over the speed-dating transcripts.
The result, in the abstract's words: speed daters who asked more follow-up questions were “more likely to elicit agreement for second dates” from the people they met. And it was the follow-ups specifically. When the program sorted the questions by kind, the follow-up rate tracked the yeses, while the other kinds it sorted, including questions that switched to a new topic, were unrelated to them. The pattern was similar for men and women (page 441).
The Sentence Worth an Article
The abstract ends on the line worth keeping: people “do not anticipate that question-asking increases interpersonal liking.” In both chat experiments, the people told to ask lots of questions did not expect to be liked any more than the people told to ask a few (pages 434 and 437). They were liked more anyway.
Outsiders did no better. In one study, a separate group read the chat transcripts and rated the people in them, and they did not see it coming either. What they did instead is telling: reading from the outside, they liked the people asking lots of questions no more than the people asking few, and warmed to the people doing the answering (page 438). The liking that questions produced belonged to the person being asked. A follow-up is not a performance for an audience. It lands with the one person it is about.
Why people hold back is a question the paper raises rather than settles. The authors suggest the blind spot may help explain why people default to talking about themselves (page 443), and list other possibilities they did not test: not thinking to ask, too little curiosity about the answer, and worry that a question will come across as rude or intrusive (page 444). If the second question feels nosy to ask, that last worry may be the feeling. In these studies, the person on the other end liked the asker more for it.
The limits, stated plainly
Everything here comes from first meetings between strangers, and short ones: fifteen-minute typed chats and four-minute speed dates. The authors note that their work only looked at first encounters (page 443). The speed-dating result is a correlation, and the paper says so in as many words: “one cannot interpret this relationship causally” (page 441). In that data, asking follow-ups behaved like a trait, a habit people carried from one date to the next, and once the analysis allowed for how sought-after each person was in general, the link with a yes shrank (page 441). So the speed dates cannot separate what the questions did from what kind of person asks them. Nothing in the paper says that more questions without end are better, and the authors raise the opposite possibility themselves (below). None of it is a prediction about the person across the table from you.
What It Implies for a Second Date
The finding does not tell you what to say on your next date. The shape of it does carry, and the shape is narrower than ask questions: when someone tells you something, ask about that before you move on. A prepared question can be a good one, and the first-date guide has plenty. A follow-up cannot be prepared at all, because it is made of what they just said. You can only ask one if you listened to the answer (page 432), which is why the authors treat it as a sign of responsiveness rather than just another question (page 439).
A second date adds one thing the studies never covered: you arrive with material. Whatever they told you last time is still there to ask about. The research looked only at single conversations, so treat that as an implication, not a finding. The paper did notice something nearby inside a conversation: as conversations went on and more had been said, people's questions shifted toward follow-ups, while other kinds became less common (page 443). What a second date is for is its own question, and the second-date guide answers it. This is only about the talking that happens inside one.
The version that is actually supported
Not ask more questions. Something narrower: when they tell you something, ask about that before you change the subject. In first conversations, that was the kind of question that went with being liked, and in the speed-dating data, with a yes to another date. And the people asking questions did not expect them to matter.
And the honest counterweight
A follow-up is not an interrogation. The authors raise the possibility themselves: the link between asking and liking “may not be monotonic”, and someone who asks question after question while saying little about themselves may end up liked less (page 443). They think there is likely a balance between asking and answering, and say it still needs testing (page 444). So answer their questions too. What the research says is that asking is worth more than people expect, and the follow-up most of all. It does not say a question is the only thing worth saying.
Where This Connects
The responsiveness piece is the idea this paper puts a measurement on, and the place to read what responsiveness is and how to offer it. The first-date guide has the Follow-Up Rule and a stock of questions to start from; this article is the evidence under the rule, not a replacement for the stock. The liking gap is a close relative: there, people underestimate how much a new acquaintance liked them; here, they underestimate what asking does. One note on words. On this shelf, the follow-up usually means the message after a good date, the one the liking gap piece and the second-date guide tell you to send. That one comes after the date. This one happens during it.
Source
Karen Huang, Michael Yeomans, Alison Wood Brooks, Julia Minson and Francesca Gino, It doesn't hurt to ask: Question-asking increases liking, Journal of Personality and Social Psychology, volume 113, number 3, 2017, pages 430 to 452, DOI 10.1037/pspi0000097. What was read: the paper itself, in the published version hosted by Harvard Business School, and its abstract as indexed in Europe PMC (PMID 28447835); page numbers above are the journal's. Also read: the correction the journal published in March 2025 (volume 128, number 3, page 669, DOI 10.1037/pspi0000491), as indexed in Europe PMC, and the audit reports it points to at osf.io/rymv8. The correction says an independent audit found several minor errors in how some results were reported, and that the audit confirmed “all the conclusions in the paper are valid.” Because some reported figures were corrected, this article prints none of the paper's statistics. Quotations are verbatim. Retrieved October 1, 2026.