01
Today’s verified habit quote
“people who ask more questions, particularly follow-up questions, are better liked by their conversation partners.”
Huang et al., It Doesn't Hurt to Ask
02
Why this habit matters
A conversation can change direction before the other person's answer has really landed. One follow-up question creates a small pause in that pattern: notice a detail they chose to share, ask about that same detail, and hear the response before moving to your own story or advice.
The useful target is responsive attention, not popularity or disclosure. Laboratory and field studies found better interpersonal ratings in particular conversations when people asked more questions or showed higher-quality listening. They did not show that one question repairs a relationship, makes every person feel safe, or obliges anyone to explain more.
03
Start with one small practice
Hear one detail → ask one follow-up → make room
- Choose one ordinary conversation that is already welcome to both people. For about three minutes, reduce any distraction you can safely set aside and let the other person finish a thought. Use eye contact, silence, or another attention cue only in a way that fits the person, culture, setting, and access needs.
- Pick one detail the person just volunteered and ask one brief question that lets them elaborate. You might ask, “What happened next?” or “What mattered most about that?” Another option is “What was that like for you?” Ask only if the topic feels appropriate. Avoid rapid-fire questions, demands for private details, advice disguised as a question, or an accusatory “why.” Accept a short answer, a refusal, or a change of subject immediately.
- Listen to the reply before adding your own story or solution. If advice might be useful, ask whether they want it rather than assuming. When the three-minute window ends, record only whether you listened and asked one welcome follow-up. Do not score how much the person disclosed, whether they liked you, or whether the conversation continued.
The 180-second window is a convenient practice boundary, not a research-tested dose. The question-asking experiments used 15-minute text chats, the field study observed four-minute speed dates, and the registered report paired about 13 minutes of training with a six-minute conversation. None tested one follow-up question for three minutes. End sooner if the other person seems unwilling, busy, distressed, or unsafe; never extend a conversation to satisfy the timer.
04
What the science can—and cannot—say
Two experiments changed the number of questions in 15-minute text chats
In the first experiment, 199 pairs of strangers chatted by text for 15 minutes. One person in each pair was randomly told to ask at least nine questions or at most four. In the second experiment, 169 pairs used the same format, but both people received an independently randomized high- or low-question instruction. In both studies, conversation partners rated people assigned to ask more questions more favorably, and independent coders rated them as more responsive.
The researchers then classified question types across those conversations. A follow-up question stayed with something the partner had just said instead of switching to a new topic. A higher follow-up-question rate was associated with coded responsiveness and partner liking. However, the experiments randomized total question quantity, not one follow-up question by itself, so the subtype analysis does not prove that today's single question causes a change.
The speed-dating result was a field association, and the paper has a correction
The field study analyzed 110 adults who each completed 15 to 19 four-minute heterosexual speed dates. People who asked a higher proportion of follow-up questions were more often chosen by partners for a second date. Because participants chose their own questions and partners differed across dates, the authors explicitly described this result as correlational. It should not be generalized to every relationship, culture, identity, or conversational goal.
In March 2025, APA published a correction after an independent audit found several minor errors in how results had been reported. The correction states that the substantive result of every hypothesis test and all conclusions remained unchanged, and that the audit found no reason to doubt data-collection integrity. This article uses that current record and avoids the affected detailed coefficients. The correction does not turn a field association into a causal result or make one question a tested dose.
A registered report tested a whole listening package, not one question
A 2024 registered report paired 606 adults who did not know one another. Listeners were randomly assigned to watch about 13 minutes of either high-quality- or moderate-quality-listening training, then listened during a six-minute conversation about the speaker's character strengths. The high-quality condition produced a small increase in perceived listening quality and greater shared positive connection.
The listening manipulation was only partly successful, and the training combined several behaviors rather than isolating follow-up questions. The condition did not change participants' authenticity or state anxiety. Together with the question-asking study, it supports practicing attentive, responsive conversation while keeping the claim narrow: one three-minute attempt may rehearse a useful behavior, but it is not a guaranteed social outcome or a substitute for context-sensitive communication.
05
Reflect and connect it to your tracker
What detail did I stay with, and did my question make room for the other person's answer rather than steer them toward mine?
Create an “ask one follow-up, then listen” action in 66 Days and record the respectful attempt—not liking, disclosure, or a promised relationship result.
Track this cue in 66 DaysResearch and primary sources
- It Doesn't Hurt to Ask: Question-Asking Increases LikingJournal of Personality and Social Psychology, 2017 · DOI 10.1037/pspi0000097Two experiments analyzed 199 and 169 dyads in 15-minute text chats between strangers. One or both partners were randomly instructed to ask at least nine or at most four questions. A third study analyzed 110 adults completing 15 to 19 four-minute heterosexual speed dates.In both chat experiments, partners reported greater liking for people assigned to ask more questions, and coders rated those askers as more responsive. A later classification analysis linked follow-up-question rate with responsiveness and partner liking, but question type was not randomized separately. The speed-dating result was observational and cannot establish causation. A 2025 correction following an independent audit changed minor reported details but stated that substantive hypothesis-test results and conclusions remained unchanged.
- The effects of listening on speaker and listener while talking about character strengths: an open science school-wide collaborationRoyal Society Open Science, 2024 · DOI 10.1098/rsos.221342A registered report randomized 606 English-speaking adults in 303 unfamiliar dyads. Listeners watched about 13 minutes of high-quality- or moderate-quality-listening training before a six-minute conversation about the speaker's character strengths.The high-quality-listening condition produced a small improvement in perceived listening quality and greater shared positive connection, but did not change authenticity or state anxiety. The manipulation was only partly successful, the training combined several listening behaviors, and the study did not isolate one follow-up question or test a three-minute practice.