A Short CBT Interview with Supervisor Feedback- Post 2/5
Introduction
This is the second post in a five-part Counselling Skills Studio experiment exploring whether AI can help counselling students practise distinct interviewing styles and receive useful feedback on their process.
The same fictional client, Mara, is used across three short demonstrations: Cognitive Behavioural Therapy-style interviewing, Motivational Interviewing, and Carl Rogers / person-centred interviewing. Each interview contains only five counsellor responses. The aim is not to resolve Mara’s difficulty, but to see how quickly each approach creates a recognisable therapeutic pathway — and whether Supervisor Mode can identify both skill and overreach from a compact sample.
In this demonstration, ChatGPT acts as the counselling student using a CBT-informed approach. The interview begins with Mara’s broad description of feeling stuck and irritable, then moves into one recent incident at home. The counsellor helps her separate the situation, automatic thoughts, bodily sensations, emotions, behaviour, and later self-judgement.
The strength of CBT becomes visible quickly: diffuse distress becomes a sequence that can be examined. But the Supervisor also identifies the corresponding risk. Once a plausible pattern appears, the counsellor may begin organising the client’s experience faster than the client can fully discover and own it.
This fictional roleplay is intended for counselling-skills education. It is not therapy, clinical advice, diagnosis, or a substitute for professional supervision or care.
Advanced Mode, short CBT demonstration
I’m Mara, 38. I asked for counselling because I’ve been feeling “stuck” and unusually irritable at work and at home. I’m functioning, mostly. I still show up, still get things done, but it feels like everything takes more effort than it should. I’ve been sleeping badly, replaying conversations, and then getting annoyed with myself for being so affected by small things.
I’m not really sure what I want from counselling. Part of me thinks I just need to get a grip and stop making everything into a big deal. But another part of me is worried that I’m becoming someone I don’t really like.
Thanks, Mara. It sounds like there are two parts of you here: one part saying, “I should be able to handle this,” and another part noticing that something feels off enough that you do not want to ignore it.
To begin, could we look at one recent moment where this showed up clearly — either at work or at home? What happened, and what did you notice in your thoughts, feelings, body, or behaviour at the time?
There was a moment yesterday evening. I got home and my partner asked if I’d remembered to book the car in for a service. It was a completely normal question.
But I snapped back with something like, “I can’t keep track of everything.” Then immediately felt awful because he wasn’t attacking me. He looked surprised, and I could tell I’d made it into a bigger thing than it was.
In my body, it was like my shoulders were already up before I even answered. Tight jaw. Sort of buzzing in my chest.
Thought-wise, it was quick. Something like, of course, another thing I’ve failed to do. And then right after I snapped, it became, great, now I’m being horrible at home too.
Thank you — that’s a clear example. Let me check the sequence with you, and you can correct anything that doesn’t fit. The situation was: your partner asked whether you had booked the car service. Your body was already tense: shoulders up, tight jaw, buzzing in your chest.
The first quick thought was something like: “Another thing I’ve failed to do.” Then the behaviour was snapping back. And afterwards the thought shifted to: “Now I’m being horrible at home too.”
Does that feel accurate? And when the first thought hit — “another thing I’ve failed to do” — what feeling was strongest in that moment: irritation, guilt, overwhelm, shame, something else?
Yes, that’s accurate.
I’d maybe say the snapping happened almost at the same time as the thought. It wasn’t like I sat there thinking it through. It was just this instant sense of, of course I’ve dropped something else.
The strongest feeling in that first moment was probably overwhelm, but it came out as irritation. Guilt came in very quickly afterwards.
Shame is… maybe there, but I don’t like that word. It sounds too dramatic. It was more like this horrible little confirmation that I’m not managing things as well as I’m pretending to.
That makes sense. It happened fast — almost like your body and voice reacted before there was much room to choose.
I’ll stay with your wording: “I’m not managing things as well as I’m pretending to.”
When that thought showed up, how strongly did it feel true in the moment, from 0 to 100?
And what did snapping back seem to do for you right then — even briefly? Did it create space, push the pressure away, protect you from the feeling, or something else?
In the moment? Probably 85. It felt very true. Not like a thought I was choosing, more like a fact that had just been exposed.
And snapping back… I think it pushed the pressure away for about half a second. Like if I made the question sound unreasonable, then I didn’t have to feel quite so exposed by it.
But then immediately it made things worse. Because now I had the original thing I hadn’t done, plus the fact that I’d snapped at him. So the thought became, see, you’re not just failing at work stuff, you’re failing at being decent at home as well.
So there are two moments there. First: “I’m not managing things as well as I’m pretending to.” Then after snapping: “Now I’m failing at being decent at home as well.”
Let’s look at the first one before we move too far. When your partner asked about the car service, what evidence did your mind use for “I’m not managing”?
And what evidence, if any, was there on the other side — that forgetting or delaying one task might not mean you are failing more generally?
The evidence my mind used was basically: I hadn’t booked it. And it had been on the list for maybe two weeks. So it wasn’t random that he asked.
But my mind didn’t stop at, I forgot the car service. It jumped to, this is what you do now — you drop things, then act irritated when someone notices. That felt much bigger.
Evidence on the other side… I mean, I am managing a lot. I’m still working, still keeping most things going. The house isn’t falling apart. I did sort out two other admin things this week.
But that almost annoys me to say, because it sounds like I’m defending myself. Or lowering the bar. Like, “Congratulations, you did basic adult things.” So yes, there’s evidence on the other side, but it doesn’t feel very satisfying.
That makes sense: part of you can list evidence that you are managing many things, but another part immediately dismisses it as “basic adult things,” so it does not really count.
Can we look at that standard for a moment? When your mind says, “That doesn’t count, that’s just basic,” what would count as evidence that you are managing well enough?
And if a friend in the same situation said, “I forgot one admin task, but I handled two others and I’m keeping most things going,” would you hear that as excuse-making — or as relevant context?
Supervisor mode.
Supervisor Mode — CBT short demonstration, after 5 interactions
This is the first layer of supervision, not the whole supervisory system. I’ll keep this brief and focused for demonstration use.
1. Orientation
This was a clear, compact CBT-style opening. Within five exchanges, you established a specific incident, mapped thoughts, feelings, body sensations, behaviour, belief strength, and began examining evidence and standards. The main caution is that you were already approaching a broader self-standard quite quickly, so the next move would need to preserve Mara’s ownership rather than turn the case into a neat formulation too soon.
2. What worked well / strongest process moves
You began with a recognisably CBT-consistent frame: you asked for a recent moment and invited Mara to identify thoughts, feelings, body sensations, and behaviour.
You also checked the sequence rather than assuming it. That helped Mara correct the timing: the snapping and the thought happened almost simultaneously. This was good CBT process because it kept the formulation grounded in her actual experience.
Your strongest move was asking what snapping “did for her” in the moment. That moved beyond simple thought-record work and began identifying the function of the behaviour: briefly pushing away exposure or pressure.
3. Where the student became too leading, interpretive, advice-driven, or prematurely intervention-focused
There was no major advice-giving. The work was mostly collaborative.
The mild risk was moving quickly from the specific thought — “I’m not managing” — into the broader standard of what would “count” as managing well enough. That is a useful CBT direction, but in Advanced Mode it can become slightly fast if the client has not yet had enough time to stay with the emotion of exposure, guilt, and shame-adjacent feeling.
Your final friend-comparison question was CBT-congruent, but it also gently steered toward a double-standard formulation. That is probably a good hypothesis, but it still needs to be earned by Mara rather than supplied too cleanly.
4. Missed opportunities
The phrase “as well as I’m pretending to” was a major opening. You used it well, but you could have lingered slightly longer there before moving to evidence testing. It carried emotion, self-judgment, and possible shame.
You also might have explored the after-snap thought — “I’m failing at being decent at home too” — as a second maintaining loop. In this short format, you rightly prioritised one thought, but that second thought could become important later.
5. Stronger alternative direction
The best next move would be to let Mara answer the standard question and then ask her to define, in her own terms, what “managing well enough” would realistically mean.
Keep it behavioural and specific. Avoid turning it too quickly into a global schema about perfectionism, worth, competence, or shame. Those may be present, but this short CBT demonstration only needs to show how the counsellor helps the client examine one painful appraisal.
6. Likely effect on rapport, alliance, and depth of exploration
Your tone was steady, structured, and respectful. Mara likely felt understood rather than interrogated.
The CBT structure helped her become clearer quickly. The only alliance risk is that if you continued at the same pace, she might start feeling analysed or organised before she fully owns the meaning.
7. Suggested next move
For a short demonstration, the next CBT move would be:
- get Mara’s answer about what would “count,”
- compare standards for self versus friend,
- identify one more balanced but believable thought,
- stop there rather than moving into a full behavioural experiment.
That would make the demonstration readable and contained.
8. Confidence in this feedback
High confidence. The style fidelity was clear, the sequence was short enough to assess cleanly, and the main developmental issue is visible: strong CBT structure with a need to slow slightly around shame-sensitive meaning.
For deeper review, the most useful options would be a turn-by-turn process audit to show exactly where the CBT process was strongest, a moment-of-shift analysis around “as well as I’m pretending to,” or a strongest-move analysis to identify the best teaching examples from this short demonstration.
What this Demonstration Shows
Even within five exchanges, the CBT structure made Mara’s distress more observable.
A normal question from her partner activated the thought that she had failed again. That appraisal appeared alongside physical tension and an immediate irritable response. Snapping briefly pushed away the sense of exposure, but it then created a second layer of guilt and self-criticism. What began as a forgotten car service was rapidly becoming evidence, in Mara’s mind, that she was failing both practically and relationally.
The counselling student’s strongest move was not simply identifying the automatic thought. It was asking what the snapping achieved in the moment. That revealed the behaviour’s short-term function: it pushed the pressure away for half a second, even though it made the situation worse immediately afterwards.
Supervisor Mode recognised the interview as clearly CBT-consistent. It also identified the central growth edge: the counsellor was already moving from a specific incident toward broader standards about competence and what should “count” as managing well enough.
That direction may eventually prove useful. But the Supervisor cautioned against reaching a polished formulation too quickly, particularly when the client’s language contains shame-sensitive meaning. Mara’s phrase — “I’m not managing things as well as I’m pretending to” — may need to be stayed with before it is organised into a larger theory about perfectionism, competence, or self-worth.
This is where supervision adds educational value. It does not merely confirm that the correct CBT elements appeared. It asks whether those elements were introduced with the right pacing, collaboration, and restraint.
The lesson from this short demonstration is therefore double-sided:
CBT can make an overwhelming experience understandable very quickly. The counsellor must still ensure that clarity is being discovered with the client, rather than completed for them.
In the next post, the same fictional client is interviewed using Motivational Interviewing. Instead of mapping the mechanics of one incident, the conversation follows Mara’s ambivalence: the part that wants to keep pushing through and the part that worries something needs to change.
Discover more from The Elderwell Initiative
Subscribe to get the latest posts sent to your email.