Evidence status: conditional support In several getting-acquainted studies, people who asked more questions—especially follow-up questions—were liked more on average, partly because partners experienced them as responsive. A 2025 independent audit identified several minor reporting errors in the original article but left the substantive tests and conclusions unchanged. The evidence does not support interrogation or the proposition that more questions are always better. This is a version 0.1 research draft, with evidence reviewed to August 2026.
How Everyday Psychology Takes Shape · Season Two: How Do We Read Other People? · Article 8
When preparing for an important conversation, we often concentrate on what we will say. How can I appear interesting? Which experience should I describe? Can I produce a clever answer? Afterwards, the other person may remember something simpler: whether they felt heard.
A question does not inherently express care. An interviewer can mechanically follow a list, and an acquaintance can use questions to intrude. The informative behaviour is often the follow-up. The other person has said something, and the next turn shows that it entered a shared conversation rather than merely filling time until we could speak again.
Psychologists place this experience within the broader idea of responsiveness: feeling that another person attends to, understands and values what one communicates. Question-asking can produce responsiveness. It is not the only route and not a mechanical substitute for it.
How researchers studied the claim
Karen Huang, Michael Yeomans, Alison Wood Brooks, Julia Minson and Francesca Gino published a set of studies in 2017. They analysed and manipulated the number of questions that people asked in getting-acquainted interactions, paying particular attention to follow-up questions based on a partner’s preceding response.
In live dyadic conversations, people instructed to ask more questions were liked more by their partners on average. Perceived responsiveness partly accounted for the relationship: question askers seemed to listen, understand and care. Beforehand, participants did not fully expect this benefit, perhaps because they feared that asking would make them appear less competent or make the interaction less interesting.
The researchers also trained a language model to identify follow-up questions in speed-dating conversations. Daters who asked more follow-ups were more likely to receive agreement to a second date. This portion is an association in naturally occurring data and cannot alone establish that questions caused the outcome. It gains meaning alongside the experimental manipulations.
In 2025, the journal published a correction after an independent audit found minor errors in how some results had been reported. The notice stated that the substantive outcome of every hypothesis test remained unchanged and that the audit found no reason to question the integrity of data collection. This update belongs in an evidence-based article. Credibility is not assessed only on publication day. A correction is not a retraction, and it should not be concealed; it tells readers what was rechecked.
What is different about a follow-up?
“Where do you work?” may simply be the next item. “You said your team changed its process last year—what prompted the change?” uses information that the other person supplied. The follow-up turns listening from a private claim into an observable act.
It also alters the information structure. The questioner no longer relies entirely on a prior model but allows the speaker to determine what needs elaboration. Understanding moves from one-way inference to the acquisition of new material. The question need not be brilliant. It enables an answer to enter a path shaped by both people.
Impressive answers still matter. Knowledge tasks, emergencies and accounts of responsibility require accurate content. If a person asks why a decision has harmed them, replying with another question can be evasion. The sometimes in the title is essential. In getting acquainted and trying to understand another person, a responsive question may create more relational value than an answer designed mainly to display the speaker.
Why more can become worse
Question count is not a dose to increase without limit. Repeated questions without any self-disclosure can feel like interrogation. Topics involving health, income, family, identity or trauma require sensitivity to relationship and willingness. A good question contains room not to answer.
Nor can care be calculated from punctuation. Sales scripts, chatbots and manipulative speakers can generate fluent follow-ups. Whether someone is responsive also depends on what happens to the answer: is “I would rather not discuss it” respected, are important details retained, and do later actions remain consistent?
Cultural norms alter the rhythm. Direct questions signal involvement in some settings; other relationships develop through more indirect elaboration. Rapid follow-up can burden someone using a second language. Getting-acquainted and speed-dating samples do not define the etiquette of all conversation.
Power is especially important. A supervisor may ask a subordinate a private question in a warm tone while making refusal difficult. Responsiveness cannot be determined solely by the questioner’s intention. It also depends on whether the person asked has meaningful freedom not to answer.
A practice that does not become a script
In an ordinary conversation, forget a quota such as “ask three questions”. Do one thing instead. After the other person finishes, identify something you genuinely do not know and that the speaker appears willing to expand. Ask about it, then allow the answer to change the next turn instead of dragging the exchange back to a prepared story about yourself.
If the person gives a short answer, changes the topic or appears uncomfortable, do not use “research says questions increase liking” as permission to continue. The first requirement of responsiveness is responding to the response.
Within a team, questions can also expose understanding to correction: “I heard that time is your main concern rather than budget—is that right?” The speaker can amend the account. Restatement does not pretend to agree. It makes one person’s model testable by the person it describes.
This is a crucial difference from generic advice to “be curious”. Curiosity that never accepts correction can still centre the questioner. A responsive question changes conduct when the answer is unexpected.
Questions are also constrained by power. An employee before a manager or an applicant before an assessor may disclose only what feels safe. The more sensitive the subject, the more important it is to explain the use of an answer, permit refusal and provide channels not controlled by the immediate authority. Receiving an answer is not the same as obtaining a complete perspective.
In consequential settings such as hiring, discipline or medical care, one conversation should not become the only evidence. Self-report belongs alongside task records, conduct over time and checkable facts, with a route for the person to correct the record. Participation does not transfer the whole burden of proof back to the person being assessed.
Understanding requires the other person’s participation
Earlier articles in this season show how readily people infer another mind from private feeling, a face or a brief action. Question-asking changes the source of knowledge. The other person is no longer only an object of observation; they become a provider and corrector of information about their experience.
Self-report is not infallible. People forget, edit and sometimes decline to disclose. Yet when the subject is someone’s preference, feeling or intention, excluding that person from the evidence altogether is usually worse than recognising the limitations of their report.
The version 0.1 judgment is that, if the goal is to help a conversation partner feel understood, preparing one fewer display story and asking one follow-up grounded in what they just said is often an evidence-supported improvement. The effect comes from responsiveness, not the number of question marks. A good question does not promise instant insight. It acknowledges missing information and returns some authority to the person being understood.
Primary research sources
- Huang, K., Yeomans, M., Brooks, A. W., Minson, J., & Gino, F. (2017), “It Doesn’t Hurt to Ask: Question-Asking Increases Liking”.
- American Psychological Association (2025), correction and independent-audit notice.
- Reis, H. T., & Gable, S. L. (2015), “Responsiveness”.
- Huang et al. audit materials, Open Science Framework.
Series navigation: How Everyday Psychology Takes Shape — Season Two overview
Discover more from Geoffrey Chen
Subscribe to get the latest posts sent to your email.