These Socratic questioning examples are for therapists who use CBT. You'll find 100 questions sorted by what each one is for (clarifying, evidence, alternatives, consequences, perspective and meaning), plus summary stems and the synthesizing questions that tie a dialogue together. Then come Christine Padesky's four stages of guided discovery, which turn a list of good questions into a conversation that changes something, and three complete invented session dialogues, with the client's answers, for depression, panic and perfectionism.
After that: the mistakes that turn Socratic questioning into an argument, how to adapt it for teens and couples, how it fits with thought records and the cognitive triangle, and how to write it in a progress note.
Every client in this article is invented. Any resemblance to a real person is a coincidence.
What Socratic questioning is in therapy
In CBT, Socratic questioning means asking questions that help a client examine their own thoughts and reach their own conclusions, instead of being told what to think. A widely used definition comes from Christine Padesky's 1993 keynote. She wrote that Socratic questioning "involves asking the client questions which: a) the client has the knowledge to answer b) draw the client's attention to information which is relevant to the issue being discussed but which may be outside the client's current focus c) generally move from the concrete to the more abstract so that d) the client can, in the end, apply the new information to either reevaluate a previous conclusion or construct a new idea" (Padesky, 1993).
Guided discovery is the wider process: the client discovers something with your help, through questions, written summaries, behavioral experiments, imagery or role play. Socratic questions are its main tool.
Padesky's central point was about purpose. Many published therapy vignettes, she noticed, showed a therapist who already knew the answer, asked a quick run of factual questions and then asked "well, then how can you think thus and so?" She called this pattern "One-two-three-aha!" Her alternative was questioning out of real curiosity: "In the best cognitive therapy, there is no answer. There are only good questions that guide discovery of a million different individual answers."
If you searched for Socratic questions for a classroom, the teaching version is different. Teaching guides, such as the University of Connecticut's list, sort questions into six types (clarification, assumptions, probing, implications and consequences, viewpoint and perspective, and questioning the question) to test reasoning about a topic. Therapy borrows the idea, but the subject is the client's own painful thought, so pace, emotion and the relationship matter as much as logic. The groups below are built for that.
Does Socratic questioning work?
The research is young but encouraging:
- In 55 adults in cognitive therapy for depression, early sessions with more Socratic questioning than the client's own average were followed by larger symptom drops by the next session, even after accounting for the alliance (Braun et al., 2015).
- In 123 clients in CBT for depression, cognitive change appeared to carry the effect of Socratic questioning on symptoms, and the link was stronger for clients who started with fewer CBT skills (Vittorio et al., 2022).
- In a video study, 144 lay observers rated a Socratic presentation as more helpful than a didactic one, and preferred it (Heiniger, Clark and Egan, 2018).
- A 2023 review calls the research "scant" and notes there is no research yet on improvement at the end of therapy (Overholser and Beale, 2023).
It is also hard to learn. In her 2019 address, Padesky noted research finding it one of the most difficult CBT skills for therapists to learn (Padesky, 2019). That is a good reason to practice with a structure.
Padesky's four stages of guided discovery
Padesky's keynote, "Socratic Questioning: Changing Minds or Guiding Discovery?", was an invited address at a European CBT congress in London on September 24, 1993. In it she observed that "many therapists, including myself, ask dozens of good questions in a session without ever helping the client put the answers together in some meaningful way," and proposed four stages of the guided discovery process.
Stage 1: Informational questions
Questions the client can answer, which bring relevant and possibly helpful information into awareness and make the concern concrete. Padesky warned against questions the client can't answer: asking a client who is unaware of their emotions "what are you feeling now?" underscores a deficit. Her better version: "Are you aware of any tension or changes in your body as we talk about your father?"
Stage 2: Listening
"It is critical that the therapist not just ask questions," she wrote. Listen for the client's own words, metaphors and images, and for "a word that seems oddly placed in a sentence." Two lines worth keeping: "If I am not regularly surprised by my clients' answers, I suspect I am either not asking interesting questions or not listening to the replies," and "If you are not truly curious to know the answer, don't ask the question."
Stage 3: Summarizing
"One of the most common mistakes I notice therapists making in the Socratic questioning process is that they don't summarize enough." She suggested a summary every few minutes, written down by the client when it matters. A summary checks that you both understand the same thing, and lets the client see all the new information at once.
Stage 4: Synthesizing or analytical questions
A question that applies what was found to the original thought or belief. In its simplest form: "how does all this information fit with your idea, 'I'm no good?'" Padesky noted that therapists often stop short of this stage, though it is where the client gets to put the pieces together.
What changed by 2019
In her 2019 address in Berlin, Padesky described how the model had grown (Padesky, 2019):
- She began calling it Socratic dialogue soon after 1993, "to capture the collaborative nature of these processes."
- Stage 2 became empathic listening, and she now listens for the parts that evidence-based models predict, noticing when one seems to be missing.
- Stage 3 favors written summaries in the client's exact words. She recommends being a "humble parrot": "only use clients' exact words, metaphors, and images in summaries."
- Stage 4 adds change-oriented questions, so the dialogue can build new beliefs and goals, not only test old ones.
- The dialogue works best around action: interactive writing, behavioral experiments, imagery and role plays, then the four stages to debrief them.
Her 1993 keynote also describes the shape a good dialogue takes. It "often goes from the abstract ('I'm no good') to the concrete (the family reunion) to the abstract (qualities of a good father) to the concrete (a behavioral experiment)." The full model, with dialogues for depression, anxiety, children and adolescents, groups and supervision, is in Dialogues for Discovery, edited by Padesky and Helen Kennerley (Oxford University Press, 2023).
100 Socratic questions for therapists, by purpose
Use this as a menu, not a script. Padesky gave her trainees lists of good questions too, and found that advanced therapists knew "these generic questions were not enough." Pick the group that fits where you are, and change the wording to the client's own words. Most of these are Stage 1 questions; the last two groups belong to Stages 3 and 4.
Clarifying questions: make it concrete (16)
Use them first, before testing anything, and whenever the client speaks in global terms ("I'm a mess", "everything's falling apart").
- Can you take me to one specific time this week when you felt that way?
- Where were you, and what happened just before the feeling started?
- What went through your mind right then?
- Was it words, a picture in your mind, or both?
- When you say "[client's word]", what does that word mean to you?
- Is that true of everything, or of one part of your life in particular?
- If I'd been watching on a video, what would I have seen?
- Which part of that bothers you most?
- Of all the thoughts you had, which one brought the strongest feeling?
- How much do you believe that thought right now, from 0 to 100?
- What feeling came with it, and how strong was it, from 0 to 100?
- What did you notice in your body?
- What did you do next?
- Is this a familiar thought, or a new one? When did you first notice it?
- How often does it show up in a usual week?
- What would you need to see to know this thought was true?
Evidence questions: what fits the thought, and what doesn't (18)
Use them once you have one specific thought and a belief rating. Ask what supports the thought first, so the client feels heard before you look at the other side.
- What makes this thought feel true?
- What happened that tells you this is true?
- What part of that did you see or hear, and what part is your reading of it?
- Is there anything, even something small, that doesn't fit this thought?
- Have there been times when this thought wasn't true?
- The last time you had this thought, did things turn out the way you expected?
- How many times has this happened before? How many times did the worst happen?
- Are you basing this on one event or on many?
- What did the other person actually say or do?
- What might you be leaving out because you're feeling low (or anxious) right now?
- When you're feeling better, do you see this any differently?
- What would someone who knows you well point out that you haven't mentioned?
- If someone had to judge this on the facts alone, what would they want to know?
- Is there information you don't have yet? How could you find out?
- Is this a thought, or a fact you could check?
- "Have you ever been in similar circumstances before? What did you do? How did that turn out?" (Padesky, 1993)
- "What do you know now that you didn't know then?" (Padesky, 1993)
- Is there any experience that shows this isn't completely true all the time?
Alternatives questions: other ways to see it (14)
Use them after the evidence is on the table, or when the client is stuck on one explanation.
- Is there another way to explain what happened?
- What are all the possible reasons this could have happened? Let's list them before we pick one.
- Which explanation fits the facts we just listed best?
- If this thought were only half true, what might the other half be?
- What else could [the racing heart, the silence, the short email] mean?
- What would you have thought about this a year ago, or on a good day?
- Where between "all" and "nothing" would you put this, from 0 to 100?
- What would "somewhat" or "sometimes" look like here?
- Is there a way to put it that includes what's true in the original thought?
- What part of this is in your control, and what part isn't?
- If you didn't believe this thought, what would you think instead?
- What would it take for the other explanation to be true?
- Is there a way of seeing this that would be more useful to you?
- Which of these explanations would you bet on, and why?
Consequences questions: what follows from the thought (14)
Two kinds. The first looks at the feared outcome (worst, best, most likely, and coping). The second looks at what holding the thought costs and gives.
- What's the worst that could happen?
- If that happened, how would you cope? Who could help?
- What's the best that could happen?
- What's most likely to happen?
- How bad would it be, from 0 to 100? And a week later? A year later?
- Have you gotten through something like this before? What helped?
- What do you do when you have this thought? What happens after?
- What do you think would happen if you didn't [check, leave, apologize, sit down]?
- What does believing this thought do for you? Does it help in any way?
- What does it cost you: in mood, sleep, work, relationships?
- If you kept thinking this way for the next five years, where would that leave you?
- What could change if you believed it a little less?
- What would you do differently if this thought weren't in charge?
- If the thought is accurate, what could you do about the situation itself?
Perspective questions: step outside the thought (12)
Use them when the client holds themselves to a standard they'd never apply to anyone else, or is too close to see.
- If a good friend told you this about themselves, what would you say to them?
- Would you judge a colleague the same way for the same thing?
- What would someone who cares about you say if they heard you think this?
- How would someone you respect handle this?
- How might the other person in the situation describe what happened?
- How will you see this in a month? In five years?
- What would you tell your younger self about this?
- If you were a coach instead of a critic, what would you say to yourself?
- Is there anyone in your life who would see this differently? What would they notice?
- "What would you advise a friend who told you something similar?" (Padesky, 1993)
- Does your family, culture or faith have a way of seeing this? How does it fit with yours?
- What would you say if you overheard someone saying this to your child?
Meaning questions: what the thought says about the person (12)
Use them later, once the alliance is solid, to find the rules and beliefs under thoughts that keep coming back. The first two, asked in turn, are what many therapists call the downward arrow. Go gently: these questions reach painful beliefs quickly.
- If that were true, what would it mean to you?
- And if that were true, what would it say about you?
- What's the worst part of that for you?
- What would it mean about you if someone saw [the mistake, the panic, the tears]?
- Is there a rule you live by here? How would you finish "If I ..., then ..."?
- Where do you think you learned that rule?
- Who else in your life lived by it?
- Does the rule still fit your life now?
- What does this thought say about how you see other people, or the future?
- What matters to you underneath this worry?
- Would you want someone you love to live by this rule?
- What would a more flexible version of the rule sound like?
Summary stems: Stage 3 (5)
- "Let me see if I've got this. You said ... Did I get that right?"
- "Here's what we have so far, in your words: ... What would you add or change?"
- "On one side we have ..., and on the other ... Is that fair?"
- "Would you write that down, so you have it in your own words?"
- "What's the most important part of what we just said?"
Synthesizing and analytical questions: Stage 4 (9)
The first five are Padesky's (2019). The rest are close cousins.
- "What do you make of this?"
- "How do these ideas fit with your original belief?"
- "Which of these ideas do you think would be most helpful for you?"
- "How could you use these ideas to help yourself this week?"
- "What are the next steps you could take?"
- Looking at all of this, how much do you believe the original thought now, from 0 to 100?
- How could you find out whether this new idea holds up?
- What do you predict will happen, and how will we know?
- If the thought comes back on Thursday night, what would you want to say to yourself?
Socratic questioning example 1: depression, "I'm useless now"
Invented client: Grace, 58, a retired accountant, in her fifth session of CBT for depression by video. Since she retired eight months ago, she spends long stretches on the couch. At the start of the session she was asked about thoughts of death and said she has none. She brought a thought record with one entry. The italic labels mark where each of Padesky's stages starts; real dialogues loop back and forth between them.
Stage 1: informational questions
Therapist: You wrote that on Tuesday afternoon you thought, "I'm useless now." Can you take me back to Tuesday? What happened just before?
Grace: My daughter called to say she'd found a sitter for the kids on Thursdays. I used to do Thursdays. After she hung up I sat on the couch for two hours.
Therapist: And as you sat there, what went through your mind?
Grace: That nobody needs me anymore. Forty years I was the one people came to. Now I'm useless.
Therapist: When you say "useless", useless to whom, or at what?
Grace: I don't do anything that matters to anyone.
Therapist: How much do you believe that right now, from 0 to 100?
Grace: Eighty-five. Maybe more.
Therapist: Is this thought new since you retired, or have you had it before?
Grace: New. When I was working I never thought about it.
Therapist: Let's look at just the past seven days. Was there anything you did, even something small, that someone else relied on?
Grace: Not really. I drove my neighbor to her eye appointment on Monday, but anyone could have done that.
Therapist: How would she have gotten there otherwise?
Grace: A taxi, I suppose. She hates taxis. She baked me bread to say thank you. And my brother calls me whenever he can't make sense of a bill. But that's not a job.
Stage 2: listening
Therapist: "That's not a job." I'm curious about that.
Grace: When I worked, what I did was worth something. People paid for it. Now it's just favors.
Stage 3: summarizing
Therapist: Let me see if I've got this, and maybe you could write it down as we go. Since you retired, the thought "I'm useless now" shows up, especially when something you used to do goes to someone else, like Thursdays. This week you drove your neighbor to an appointment so she didn't have to take a taxi, and she baked you bread. Your brother calls you about his bills. And underneath, there's a rule: if it isn't paid, it doesn't count. Did I get that right?
Grace: Yes. Said all together, it's more than I thought.
Stage 4: synthesizing questions
Therapist: What do you make of this?
Grace: I'm not useless. I'm just not useful the way I'm used to.
Therapist: How does that fit with "I'm useless now"?
Grace: It isn't true the way I said it. Sixty, maybe. I still miss being needed.
Therapist: That's a real loss, and it makes sense to miss it. How would you like your weeks to be, now that work isn't filling them?
Grace: Something regular. Somewhere people expect me.
Therapist: What's one thing you could try this week to find out whether that would help your mood?
Grace: The library runs a tax-help program for older people. I've walked past the sign. I could call them.
Therapist: Would you call before Thursday, and rate your mood before and after?
Grace: I can do that.
What to notice:
- One concrete moment first.The dialogue starts on Tuesday afternoon, not with "I'm useless" as a life verdict.
- No argument.The therapist never says "you're not useless." Grace finds the evidence, so the conclusion is hers.
- Listening found the rule."That's not a job" was the oddly placed phrase; it holds the rule "if it isn't paid, it doesn't count", a target for later meaning work.
- The new belief is balanced, not positive.Eighty-five to sixty, and the loss is named as real.
- It ends in action she chose:a call, with ratings, which is a behavioral experiment on her mood.
- Risk comes first.If Grace's answers had turned toward hopelessness or thoughts of death, the questioning would stop and a risk assessment would take its place. Our safety plan template covers that work.
Socratic questioning example 2: panic, "My heart is going to give out"
Invented client: Theo, 27, a graphic designer, in his fourth session of CBT for panic disorder, seen in the office. His primary care physician examined him last month, including an ECG, and found nothing wrong with his heart. He has had about 30 panic attacks since March.
Stage 1: informational questions
Therapist: Let's take Wednesday's attack on the train. What did you notice first?
Theo: My heart. It started pounding out of nowhere. Then I got dizzy.
Therapist: When your heart started pounding, what went through your mind?
Theo: That it's going to give out. That this time it's real.
Therapist: How much did you believe that in the moment, from 0 to 100? And now, sitting here?
Theo: Ninety-five on the train. Maybe forty now.
Therapist: What did you do when you thought your heart was going to give out?
Theo: Got off at the next stop, sat on a bench, checked my pulse, and called my sister to stay on the phone with me.
Therapist: Of the 30 or so attacks since March, how many times did your heart give out?
Theo: None. But that's because I sat down every time.
Stage 2: listening
Therapist: "Because I sat down every time." So sitting down feels like what's kept you safe.
Theo: Yes. I don't know what would happen if I didn't.
Therapist: Have you ever felt your heart pound like that when you weren't scared?
Theo: When I run for the bus. Or after three coffees.
Therapist: When you run for the bus, what do you do about it?
Theo: Nothing. I get on the bus, and it slows down.
Therapist: And what did your doctor say after the tests?
Theo: That my heart is healthy. She thought it was anxiety.
Stage 3: summarizing
Therapist: Let me pull this together, and tell me what I've got wrong. When your heart pounds, the thought "it's going to give out" jumps to 95. You've had about 30 attacks, your heart kept going every time, and you've sat down every time, so you've never found out what happens if you don't. Your heart also pounds when you run for the bus, and it slows down while you keep going. Your doctor's tests were normal.
Theo: That's it. The sitting down is the part I'm not sure about.
Stage 4: synthesizing questions, then action
Therapist: What do you make of that?
Theo: Maybe I don't actually know that sitting down is what saves me.
Therapist: How could we find out, without waiting for the next train?
Theo: I don't know. Make it happen here?
Therapist: That's one way. We could run in place for a minute to make your heart pound on purpose, and you'd stay standing afterward. You can say no. If we did, what do you predict would happen?
Theo: I'd panic. Maybe a 30 percent chance something bad happens to my heart.
(They run in place together for 60 seconds. Theo stays standing. Two minutes later:)
Therapist: What happened?
Theo: It pounded hard and I wanted to sit. It came down in a couple of minutes anyway.
Therapist: How does that fit with "my heart is going to give out"?
Theo: It's anxiety making it pound. Twenty, maybe.
What to notice:
- The prediction was tested, not argued. This follows the cognitive model of panic, in which attacks come from catastrophically misreading body sensations (Clark, 1986). Sitting down is a safety behavior: it stops Theo from learning that his heart is fine without it.
- Listening caught the key phrase. "Because I sat down every time" told the therapist what to test.
- The "hot" belief and the "cool" belief differ (95 on the train, 40 in the office). Rate both, or progress in the office will look bigger than it is.
- Medical evaluation came first. Don't question a client's fear about their heart, breathing or health until a physician has looked, and follow your panic protocol for in-session exercises.
Socratic questioning example 3: perfectionism, "One mistake and they'll see I'm not good enough"
Invented client: Lena, 33, an architect, in her seventh session of CBT for perfectionism and work stress, by video. She has been checking drawings late into the night, and her last two submissions were late.
Stage 1: informational questions
Therapist: You said you checked the drawings until 1 a.m. on Sunday. What were you looking for?
Lena: Anything wrong. A dimension off, a label in the wrong font.
Therapist: And if one got through, what would happen?
Lena: The partners would see it, and they'd realize I'm not good enough to be there.
Therapist: How much do you believe that, from 0 to 100?
Lena: Eighty.
Therapist: What would "good enough to be there" look like?
Lena: No mistakes. Ever.
Therapist: Does anyone at the firm meet that standard?
Lena: (pause) No. Mark sent out a set last month with the wrong north arrow.
Therapist: What happened to Mark?
Lena: Someone caught it, he fixed it. Nobody said anything. He's up for promotion.
Therapist: What did you think of Mark when you saw it?
Lena: That it's an easy thing to miss.
Stage 2: listening
Therapist: For Mark, one mistake is "an easy thing to miss." For you, one mistake would mean you don't belong. What do you make of the difference?
Lena: I hold myself to a different standard. I always have.
Therapist: And if the partners did see a mistake of yours, what would that mean about you?
Lena: That I've fooled everyone. That I'm a fraud.
Therapist: That's a painful thought to carry. Let's hold it for now and come back to it. What does the checking cost you?
Lena: Sleep. And my manager mentioned the late submissions.
Stage 3: summarizing
Therapist: Here's what we have, in your words, and change anything that's off. You check until 1 a.m. because one mistake would show you're not good enough. Nobody at the firm meets "no mistakes, ever." Mark's mistake got fixed, and you saw it as easy to miss. The checking has cost you sleep and made two submissions late, and the lateness is what your manager noticed.
Lena: So what actually got noticed was being late, not a mistake.
Stage 4: synthesizing questions
Therapist: How does that fit with "one mistake and they'll see I'm not good enough"?
Lena: It doesn't, really. The checking is causing the problem I'm afraid of.
Therapist: How much do you believe the original thought now?
Lena: Fifty. It still feels true at night.
Therapist: That's an honest answer. What would you like to try this week?
Lena: Check each set once, at a set time, and send it by 9 p.m.
Therapist: What do you predict will happen?
Lena: Someone will find something, and I'll feel sick.
Therapist: Let's write that prediction down, with how bad you think it would be, and see what actually happens.
What to notice:
- The perspective question did the work. Lena's own judgment of Mark ("an easy thing to miss") was more persuasive than anything the therapist could say about standards.
- A meaning question surfaced the belief underneath ("I'm a fraud"), and the therapist noted it for later instead of chasing it at the end of a session.
- Consequences questions moved the focus from the feared mistake to the real cost of checking.
- The plan tests a prediction, written down so next session can compare it with what happened.
Common mistakes with Socratic questioning
Leading questions
A leading question has one acceptable answer, and the client knows which one you want. It is Padesky's "One-two-three-aha!": the client may say "I see what you mean," but the discovery is yours, not theirs. Some rewrites:
- Leading:"Don't you think you're being a bit hard on yourself?" Open: "How would you judge a friend who did the same thing?"
- Leading:"Isn't it true that most people wouldn't even notice?" Open: "What do you know about how other people reacted?"
- Leading:"So that's not really evidence, is it?" Open: "What part of that did you see, and what part is how it felt?"
- Leading:"Can you see that you're catastrophizing?" Open: "What's most likely to happen?"
- Leading:"Wouldn't it be more helpful to think you did your best?" Open: "If this thought weren't in charge, what would you think instead?"
A quick test: if you'd be disappointed by an honest "yes" or "no", the question is leading.
Debating the client
When the client starts answering "yes, but...", you are in an argument, and clients win arguments about their own lives. Go back to Stage 2: reflect the "but", because it often holds the meaning ("But that's not a job"). Ask what supports the thought, not only what doesn't. Ask, "What part of this still feels true?" A thought that drops from 85 to 60 is real progress; pushing for 10 usually costs you the alliance.
Going too fast
- Before the client feels heard.Questioning a thought in the first minutes of telling it can sound like "you're wrong." Understand it first.
- Before there's one specific thought.Testing "everything's falling apart" goes nowhere. Clarify until there is one hot thought in one situation.
- Question after question, no summary.A run of questions without a summary feels like a cross-examination. Summarize every few minutes, as Padesky advised.
- No silence.Padesky (2019) writes that big questions "are not answered quickly", and that the therapist "needs the courage to be silent" while the client thinks. Don't answer for them.
- Meaning questions too early.Downward arrow questions reach core beliefs fast. Wait for a solid alliance, and leave time in the session.
- A rushed ending.Starting a new thought with five minutes left leaves the client holding it open. Save Stage 4 time.
Other mistakes
- Questions the client can't answer, such as "what are you feeling?" to someone who can't yet name feelings. Ask about the body instead.
- Stopping before Stage 4. A great list of evidence the client never connects to the original thought changes little.
- Testing an accurate thought. If the rent really is late or the discrimination is real, the work is problem-solving, support or advocacy, not a reframe.
- Questioning in a crisis. When a client may be at risk, assess and plan for safety first.
- "Why" questions that sound like blame. "Why did you do that?" invites a defense. "What led up to that?" invites information.
- Nothing written down. Without a written summary, much of the session can be gone within the hour. Padesky (2019): "If it isn't written down, it didn't happen."
Adapting Socratic questioning for teens
Teens can do this well, but they are quick to spot a question that's really a lecture. Robert Friedberg's chapter in Dialogues for Discovery covers this work, including asking permission to test thoughts, why adolescents sometimes want to hold on to a thought, using games, and matching the young person's level of information processing. In practice:
- Ask before you test. "Want to check that thought, or does it feel too true to check today?" A teen who chooses to look is more likely to look honestly.
- Keep it concrete and short. One situation, three or four questions, a summary. Use 0 to 10 instead of 0 to 100.
- Make it a game or a mission. Being a "thought detective" who only counts what was seen or heard works for many younger teens; older teens may find it childish, so ask.
- Use their words, including slang, and don't correct them.
- Respect the reason for the thought. Expecting rejection can feel safer than hoping. Ask what the thought protects them from before asking them to drop it.
- Mind who's in the room. Questioning a teen's thought in front of a parent can feel like being corrected in public, and parents should not be called in as witnesses for the prosecution.
A short example, with Jordan, 15 (invented):
Therapist: Want to check the thought "everyone at lunch thinks I'm weird", or leave it for today?
Jordan: We can check it.
Therapist: Detective rules: we only count what you saw or heard. Who's "everyone"?
Jordan: The table. Six people.
Therapist: What did they do that tells you they think you're weird?
Jordan: Maya laughed when I said something.
Therapist: What happened after she laughed?
Jordan: She said it was funny. And Dev asked if I'm going to the game on Friday.
Therapist: What do you make of that?
Jordan: Maybe it's mostly me who thinks I'm weird.
Therapist: Out of 10, how sure are you now that everyone thinks so?
Jordan: Like a five. It was a nine.
Adapting Socratic questioning for couples
In couple work, Socratic questions usually target how each partner explains the other's behavior. APA's description of enhanced cognitive-behavioral couple therapy notes that partners often hold "negative, distorted cognitions when interpreting the other partner's behaviors, attributing such behaviors to malicious intent, self-centeredness, and so forth" (APA, Enhanced Cognitive-Behavioral Couple Therapy; see also Epstein and Baucom, 2002).
- Screen each partner alone first.Before joint cognitive work, ask each partner privately about violence, threats and control. If one partner is afraid of the other, don't use Socratic questions in joint sessions to test that fear; it can put them at risk. Follow your training and setting's policy.
- Balance.Ask both partners the same kind of question, in turn. If you question one partner's thinking three times in a row, the other partner hears that you've picked a side.
- Question the meaning, not who's right."What did you tell yourself it meant?" works better than "Did that really happen?"
- Let the partner add data, not a rebuttal.Ask the speaker first what they'd want to know, then invite the partner to describe their side of that moment.
- Compare standards."What does caring look like to you? And to you?" Many fights are two different rule books.
- Ask about the pattern."What happens next, after you say that?" moves the focus from blame to the cycle.
A short example with Sam and Alex (invented):
Therapist (to Sam): When Alex came home late without texting, what did you tell yourself it meant?
Sam: That I'm not a priority.
Therapist: How strongly, from 0 to 100?
Sam: Ninety.
Therapist: Before we hear from Alex, what would you want to know about that evening?
Sam: Whether Alex even thought about me.
Therapist (to Alex): Would you tell Sam what that evening was like for you?
Alex: My phone died at four and the meeting ran over. I thought about texting from someone else's phone and figured I'd be home soon.
Therapist (to Sam): How does that fit with "I'm not a priority"?
Sam: It's less about that night. It's that it keeps happening.
Therapist: So the pattern matters more than one night. Alex, what do you make of that?
How Socratic questioning fits with thought records and the cognitive triangle
The cognitive triangle comes first. Clarifying questions fill its three corners: what happened, what went through your mind, what you felt and what you did. Our cognitive triangle worksheet and guide shows how to teach it in a first session; Socratic questioning is the next step, once the client can catch a thought.
The 7-column thought record is Socratic questioning on paper. Padesky wrote in 2019 that "Socratic methods are embedded in the 7-column thought record," which she developed and which appears in both editions of Mind Over Mood (Greenberger and Padesky). Each question group fills a column:
- Situation, moods and automatic thoughts (columns 1 to 3):clarifying questions, including the belief and mood ratings and picking the hot thought.
- Evidence that supports the hot thought (column 4):the supporting side of the evidence questions, asked first.
- Evidence that does not support it (column 5):the other evidence questions, plus perspective questions.
- Alternative or balanced thought (column 6):alternatives questions, then synthesizing questions, with a belief rating.
- Rate moods now (column 7):the closing synthesizing question.
Do it aloud in session before you assign it. Padesky (1993) found that the questions on her list, given to clients, "helped clients generate alternative responses on their written automatic thought records." Many therapists hand the client a short card of five or six questions from the list above to use at home.
When the thought is a prediction, test it. Theo's and Lena's thoughts were predictions, so the dialogue ended in a behavioral experiment. Padesky's later work puts action at the center for this reason.
Meaning questions feed the case conceptualization. The rules and beliefs they surface ("if it isn't paid, it doesn't count", "I'm a fraud") belong in your formulation; see our case conceptualization examples.
Socratic questions are not MI questions. Both avoid arguing, but they have different jobs: motivational interviewing questions draw out the client's own reasons for change, while Socratic questions examine a thought. If a client isn't sure they want to change, motivational interviewing questions usually come first.
How to write Socratic questioning in a progress note
Name the method, the thought it targeted and what the client produced. "Challenged distorted thinking" tells a reader little, and sounds like a debate. Record:
- The intervention by name: guided discovery, Socratic questioning or Socratic dialogue, and any behavioral experiment.
- The target thought, in the client's words, with the belief rating before.
- What the client generated: the main evidence, the alternative thought in their words, any rule or belief that surfaced.
- The response: the belief rating after, the mood rating if you took one, and whether the client did this alone or with prompting.
- Risk: what you asked and what the client said, when you asked.
- The plan: the experiment or thought record, and what you'll review next.
You don't need to transcribe the dialogue. A BIRP note for Theo's session (invented):
Phrases that work in any format:
- "Used Socratic questioning to examine the automatic thought '...' (belief X/100)."
- "Client generated evidence for and against the thought, and an alternative: '...'"
- "Belief in the original thought fell from X to Y."
- "Client identified an underlying rule: 'If ..., then ...'"
- "Client had difficulty generating alternatives today; thought record assigned to continue."
- "Planned a behavioral experiment to test the prediction '...'"
Phrases to avoid: "convinced client", "client admitted the thought was irrational", "disputed client's beliefs". They describe winning an argument, not what the client learned.
In DAP, all of this sits in Data and Assessment (see our DAP notes examples); in SOAP, the client's account goes in Subjective and the ratings in Objective (see the SOAP note examples). BIRP gives the method and the response their own sections, which suits technique-led CBT; the BIRP note examples show one session in all three formats. For more wording, the progress notes cheat sheet lists CBT intervention words and client response words.
How this works in TranqBay
TranqBay is practice software for therapists in private practice: a practice website, online booking, video sessions and session notes in one place. Here is where Socratic work fits.
- Socratic dialogue over video.Video sessions are end-to-end encrypted and HIPAA-ready, and TranqBay provides a Business Associate Agreement that applies when you accept the Terms. Before a session, ask the client to keep paper and a pen nearby, so they can write the summary in their own words, as Padesky recommends.
- Writing it up.Session notes come in BIRP, DAP, SOAP or GIRP format, or as a blank note in your own words. The structured formats are part of the advanced note templates, which some plans include.
- Checking whether it's helping.The form library has ready-made templates you can send a client for a session, including the PHQ-9 and GAD-7, plus intake and consent forms. The client fills them in online and the answers wait with the session. TranqBay doesn't score them; our PHQ-9 scoring calculator and GAD-7 scoring calculator give the total and severity band.
- Clients finding you.If your practice website says you offer CBT, the booking page is where a client who read that takes the next step.
practice.tranqbay.health shows a session page with a sample DAP note that reviews a client's thought record, which is where a lot of Socratic work ends up, so you can see where your notes would sit before you set anything up.
Frequently asked questions about Socratic questioning
What is Socratic questioning in CBT?
A way of asking questions that helps a client examine their own thoughts and reach their own conclusions. In CBT it is the main tool of guided discovery: the therapist asks, listens, summarizes and asks the client to put it together, instead of telling the client what to think.
What are some examples of Socratic questions?
"What makes this thought feel true?", "Is there anything that doesn't fit it?", "Is there another way to explain what happened?", "What's most likely to happen?", "What would you say to a friend who thought this?" and "How does all this fit with your original thought?" The list above has 100, sorted by purpose.
What are the six types of Socratic questions?
The six types used in teaching are clarification, assumptions, probing reasons and evidence, implications and consequences, viewpoints and perspectives, and questions about the question. In therapy, it is more useful to sort them by what they do for the client: clarifying, evidence, alternatives, consequences, perspective and meaning.
What is the difference between Socratic questioning and guided discovery?
Guided discovery is the whole process of helping a client discover something new, through dialogue, writing, behavioral experiments, imagery or role play. Socratic questioning is the questioning part of it. Padesky's four stages describe how the questions become discovery.
What are Padesky's four stages of Socratic questioning?
From her 1993 keynote: (1) informational questions, (2) listening, (3) summarizing and (4) synthesizing or analytical questions. By 2019 she called the process Socratic dialogue and described the stages as informational questions, empathic listening, summaries and analytical or synthesizing questions.
Is Socratic questioning the same as cognitive restructuring?
Not quite. Cognitive restructuring is the goal: a more balanced, useful way of thinking. Socratic questioning is one of the main ways to get there, and it can also be used to build new beliefs, plan experiments or explore goals.
Can I use Socratic questioning outside CBT?
Yes. It is used heavily in cognitive processing therapy for PTSD, where one study linked therapists' Socratic skill to greater client improvement (cited in Padesky, 2019), and Dialogues for Discovery is written for therapists of every approach. The spirit (curious questions, the client's own conclusions) fits most kinds of therapy.
How many Socratic questions should I ask in a session?
There's no set number. A focused dialogue on one thought often takes 10 to 20 minutes and a dozen questions, with a summary every few minutes. Stop when the client has something to apply, not when you run out of questions.
What if the client says "I don't know"?
Treat it as information. The question may be too abstract, too early or too painful. Make it more concrete ("Let's take just Tuesday"), offer options ("Some people think A, some think B; is either close?"), allow silence, or come back to it later. Don't answer it for them.
Is there evidence that Socratic questioning works?
Early evidence, mostly in depression: more Socratic questioning in a session predicts larger symptom drops by the next session, and the effect seems to run through cognitive change. Reviews call the research limited, and no study yet shows its effect at the end of therapy.
Before your next session
Pick one thought a client brought last week. Clarify it until it's one hot thought in one moment, ask from two or three groups above, summarize in the client's words, and finish with "How does all this fit with your original thought?" Ask the client to write the summary down. Then note the belief rating before and after.
If you want booking, video sessions and session notes in one place, see how TranqBay works for private practice: you can look at a practice website design and the booking page your clients would use. If you're deciding whether to join, this page explains how joining TranqBay works.
Sources
padesky.com 2academic.oup.com 1Research 5Books 1Associations 2cetl.uconn.edu 1988 Lifeline 1
padesky.com2
- 1Padesky CA. Socratic questioning: Changing minds or guiding discovery? Invited keynote address, European congress, London, September 24, 1993 (Padesky's publications page)padesky.comhttps://www.padesky.com/clinical-corner/publications/
- 2Padesky CA. Action, dialogue and discovery: Reflections on Socratic questioning 25 years later. Invited address, Ninth World Congress, Berlin, July 18, 2019 (Padesky's publications page)padesky.comhttps://www.padesky.com/clinical-corner/publications/
Research5
- 4Braun JD, Strunk DR, Sasso KE, Cooper AA. Therapist use of Socratic questioning predicts session-to-session symptom change in cognitive therapy for depression. Behav Res Ther (2015)pubmed.ncbi.nlm.nih.govhttps://pubmed.ncbi.nlm.nih.gov/25965026/
- 5Vittorio LN, Murphy ST, Braun JD, Strunk DR. Using Socratic Questioning to promote cognitive change and achieve depressive symptom reduction. Behav Res Ther (2022)pubmed.ncbi.nlm.nih.govhttps://pubmed.ncbi.nlm.nih.gov/35016095/
- 6Heiniger LE, Clark GI, Egan SJ. Socratic and non-Socratic presentation of information in CBT, rated by lay observers. J Behav Ther Exp Psychiatry (2018)pubmed.ncbi.nlm.nih.govhttps://pubmed.ncbi.nlm.nih.gov/29055854/
- 7Overholser JC, Beale E. The art and science behind socratic questioning and guided discovery: a research review. Psychotherapy Research (2023)pubmed.ncbi.nlm.nih.govhttps://pubmed.ncbi.nlm.nih.gov/36878221/
- 8Clark DM. A cognitive approach to panic. Behav Res Ther (1986)pubmed.ncbi.nlm.nih.govhttps://pubmed.ncbi.nlm.nih.gov/3741311/
Books and publishers1
Professional associations2
- 10Epstein NB, Baucom DH. Enhanced Cognitive-Behavioral Therapy for Couples: A Contextual Approach. American Psychological Association (2002)apa.orghttps://www.apa.org/pubs/books/431790A
- 11American Psychological Association. Enhanced Cognitive-Behavioral Couple Therapy (video description)apa.orghttps://www.apa.org/pubs/videos/4310862