Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Psyched to Practice

Psyched to Practice artwork

Beyond the Bot: AI in Supervision- How Much Help Is Too Much?

Nov 20, 202632 minAudio
ShareXLinkedInFacebookWhatsAppEmail

AI is quickly becoming part of everyday clinical work, but what happens when it starts shaping how the next generation of mental health professionals learns their craft? In this episode of Beyond the Bot, Dan Florell and Adam Lockwood explore the growing role of AI in clinical supervision and the tension between using technology and developing real clinical skills. They discuss the risks of overreliance, cognitive offloading, and clinicians who may struggle to recognize when AI gets something wrong. They also explore how supervisors can rethink training, evaluate competency, and help emerging professionals build the judgment, empathy, and critical thinking they need to practice independently. As AI takes on more of the technical work, are we creating room for better clinical care, or simply expecting clinicians to do more with less? Website: https://www.learn.psychedtopractice.com Social Media: https://www.facebook.com/PsychedToPractice https://www.instagram.com/psychedtopractice/ https://www.linkedin.com/company/psyched-to-practice/ “Be well, and stay psyched"

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

SPEAKER_00The following is a site to practice production.

SPEAKER_01Hey, welcome back to another episode of Beyond the Bot AI and Mental Health. I'm Dan Florel. I'm with Eastern Kentucky University. I'm a professor there. And with me, my co-host, Adam. Adam, take it away.

SPEAKER_02Hey, Dan. Adam Lockwood, Kent State University.

SPEAKER_01Okay, great. And today we are going to focus on the issue of supervision in the age of AI. I can see some pluses and minuses to this particular topic. I think some aspects of AI are probably enhancing supervision in a way that we've never seen before. But I also have a sense that AI can be unduly influencing who our supervisees may be listening to. But Adam, you've been around the country doing a lot of talks on AI and been able to chat with some people.

SPEAKER_01What are some of the issues that you're hearing kind of out there in the in the world as far as what people are kind of complaining about or noticing?

SPEAKER_02I'm looking at the survey results we have coming in because we're doing a survey. And it looks like at this point, only 13% of school psychologists are saying that they haven't been using AI at work. So and that's that's changed from about 21% a year ago. So we're we are probably reaching close to the ceiling of adoption. It's pretty crazy. Um so what I'm hearing is people are using AI, and I'm hearing the question I'm getting is what do I do for supervision?

SPEAKER_02Do I let them use it? Do I not let them use it? What do you tell them?

SPEAKER_01Um well, you know, I'm the business we're in, right? We're uh about AI and how to use it, usually in an ethical manner, in a thoughtful manner. So I would say if they're gonna use it, use it sparingly at first to make sure that skill development is firmly in place before you're going to embrace it. If you are using it, definitely not turn them loose, let them use it for everything, but maybe more structured, say, okay, next four cases. One of them I want you to try AI out, and I want you to do the other three kind of by hand, by hand, right?

SPEAKER_01I mean, you're still using all the other technologies that you've learned in your grad training. But my big concern, you know, is is something you brought up real early on is kind of that de-skilling effect or no skilling, as it might be for supervisees. And that's just saying that they haven't spent the time that's required to master a skill like report writing, is one of the ones I'm thinking of. So I think it's fine. I think you need to model a little bit of how to use it, or even model how to kind of join together in how to use it, because I would imagine most of the supervisors also are not very comfortable yet in their AI skills, especially at the speed that it's changing.

SPEAKER_01So you could almost model a joint way of how do you deal with some new technology that comes in to your profession, which I think also has a lot of value because now that I'm thinking about it more, just think in the last decade, you had cloud computing rush in, and how do you use remote scoring in an ethical manner? Then you had telehealth and all the chaos that COVID had, and using Zoom for clinical services. That was a big change.

SPEAKER_01And now you've got AI. So I'm going to go out on a limb and say there's going to be some other technological disruption in a supervisee's future career, and they need to learn how to deal with things that are not explicitly taught to them in their graduate training or their early years. So that's my very long response to your simple question. What do you what do you tell them?

SPEAKER_02Well, one problem I have with what you said is if I heard you right, I'm gonna have them do half with AI and half on their own. That means they're getting half the experience that they would have three or four years ago. So I'll tell you, for for my program, we have a expectation that they're not gonna use AI for any sort of clinical task. Our expectation is that they don't use AI.

SPEAKER_02However, I'm starting to change my thoughts about that because prohibition historically has not worked, whether it's for alcohol, drugs, or anything else that's super reinforcing. Right? So I'm not sure that I I still have that hardline stance and and the the work I do in probably hasn't helped, you know, if I'm um financially benefiting from AI, maybe I'm softening for that reason. But I'm thinking more about well, how can we train supervisees to use AI but still get the skills they need to practice independently because you do need to be looking over the output of AI.

SPEAKER_02And the way I always think about it, Dan, is I use the analogy of AI as an intern. It's a that overly confident but good intern. And those are the ones that I worry the most about, the ones who think they know it all, because they don't ask questions and they they don't they're not careful and they can be reckless, but they're not there yet. So I think of AI as an intern, and then I ask myself, well, would I let a practicum student supervise an intern?

SPEAKER_02No. Would I let an intern supervise an intern? No. Would a first-year school psychologist be allowed, according to NAS rules, to supervise an intern? No, they have to practice for several years before they're allowed to fill that role. So, so what what does this look like? And I think what it looks like is that we have trainees do some of the work themselves first,

SPEAKER_01and

SPEAKER_02then have them use AI, and then we compare the different outputs. That is the that's the best outcome because what you mentioned is a real concern in the medical field at this point, and it's known as never skilling. And there's things I'm reading where where publications of we're worried that doctors are gonna be out there who never learn their job, and they're out there practicing medicine and they've been relying on AI their whole career, their whole training, and when a hallucination happens, they're not gonna be able to catch it, and it's gonna be catastrophic.

SPEAKER_02So that how do you how do you negotiate that is tricky, but I think one thing you need to do is you cannot you can no longer rely on the artifacts as an indicator of competency. Before, how I knew if an if an intern was competent was based on their report. I would look at it and go like, oh yeah, they're really getting this down.

SPEAKER_02And I would watch them progress throughout the year. At this point, you cannot trust that report. You have to dig in more, and really that looks like, hey, I see you gave this recommendation. Why? I see you included this. This is great. Tell me more about how you came to that conclusion. I think it actually is going to create more work for supervisors because not only do they now have to look over their reports and sign off on them, but they really should be digging in to make sure that the trainee didn't just have AI automate the report writing process, for example.

SPEAKER_01Well, I think one of the good things with supervision, at least in theory, right? You're one-on-one with them. It shouldn't all be pay-per-review supervision, especially at the early stages. And I've always said, you know, when people wring their hands about, oh, we have no good AI detectors, I mean, you hit the nail on the head right there. You just get them in front of you and ask. If they can't feasibly put things together and what they quote did, then they probably have used AI. And once again, I'm not against AI.

SPEAKER_01I think you could use it. You know, if you're using it especially as a collaborator, you should have a lot of back and forth with it. You're going to know that content with AI's help and hopefully even get a better product than either of you would have gotten alone. But I think at that early stage, I think you're right. You know, you do need to be resistant. You did uh mishear me a little bit. I wasn't saying like you spend half and half. Uh I was like, if you have four cases, maybe one would go to AI. And the ratio can be much higher.

SPEAKER_01It goes back to the thing I talked about, the foxtrot approach, where you work out a ratio of being able to gain skills, and then how much will you benefit from the advances of the AI? And that the less skilled you are in an area, the more you need to be using AI in a very limited fashion. And that as you get more experienced, then you can twist that ratio around where maybe you're using AI most of the time, but you're still going back and doing reports from scratch, and that the moment they start to become hard again by doing them by scratch, it means you you're using AI too much, that that skill is starting to rust away.

SPEAKER_01And so the other the other thing I was going to point out is we we do have interns sometimes supervising practicum students. I took a supervision course, right, as my fourth-year, third-year doctoral student, and I was supervising you know, one of the first, first or second years. So, you know, not to say that there isn't a little bit, and I'm not saying that that supervision was great supervision, but I think it would point to a nice analogy model that we do need to try it a little bit because good supervisors only get to be good supervisors by starting somewhere.

SPEAKER_01So you got to learn that skill just as much. Yeah, I I don't I've gone back and forth. You know, I'm working on a presentation right now at one of the state associations I'm gonna go do, and it came out of a keynote I did over at Clark University about the role and decisions we need to make about skills that we need to develop that are core to who we are as a profession, skills that we can develop or have in collaboration with a AI as far as helping extend your thought process.

SPEAKER_01And then there are skills that maybe just aren't worth retaining anymore. Like, I don't know, looking things up on a table for a test. You know, I think we've most of us have ever have given it up. I haven't. I still like scoring like the whisk by hand. Frankly, I think it's faster sometimes, and many times I'm in a school where there's no Wi-Fi that I can get on. And so I don't want to wait and I want to get a quick idea of what the child's IQ is, and I'll I'll just do it by hand. But my point being, a majority of the field, I would say, has decided that that's just not a skill that they need to retain anymore, and it doesn't really impact your skills as far as being a psychologist, school psychologist.

SPEAKER_01And so, do we need to go back in medicine? The old way that they figured out what your blood counts and stuff is they got your stuff under your blood under a slide, and they counted the blood cells using their little area mapping technique. Well, if doctors don't know how to do that and they trust the readout from the machine on what your blood count is, I'm not really too concerned about that. So, where's the dividing line that we are so much in a moral panic that AI is doing something that maybe, yes, those skills were considered integral back in the day because there was no other way to do it.

SPEAKER_01But now, are they really integral to what we do as a profession and our understanding? I I think one time, I certainly one of the things I'll say is like giving an IQ test. Like giving an IQ test, I believe, will be a task that we can eventually automate totally away, and that we really don't need people spending two, three, four hours on per case sitting across from an individual giving them these tasks.

SPEAKER_01Now, some may argue, you know, you're gonna miss out on them changing their emotions or how they react to things. But I think there are other ways of getting at that and getting at it quicker than requiring four hours of a very standardized test while you're scribbling away. But I've been going on. What do you what's your reaction to that, Adam, and those bold statements?

SPEAKER_02Oh, I have a lot of thoughts. I I am consulting with some folks who are thinking about the feature of cognitive assessment. And I don't think that 20 years from now it's going to be a paper and pencil. I do often wonder what are skills we need to have. I think I've talked before about reading and writing. Do we need to read and write? But the other thing I think about sometimes is, well, any one skill might not be necessary. But what happens when every part can be automated?

SPEAKER_02What happens when we're offloading all of these tasks? And that's that's what it's known as in the AI literature, cognitive offloading. If you have no ownership, you haven't given the test, you haven't written the report, you you're not retaining anything. You you're not you are essentially someone who's just checking off and making sure that the machines have done the whatever they do. I think report writing, I don't know if it needs to exist or not, but I do think that wrestling with the data helps the case conceptualize, helps you to know the client.

SPEAKER_02If you're just or the patient or the student, if you are just taking the test and handing it to AI, well, I think something's lost there. And there's also the problem of anchoring, and that is whenever someone you know suggests, hey, I think this might be a good idea, whether it's AI or human, we switch from is that the best idea a lot of times to confirming whatever it is we've been provided. So instead of is this best to is this good enough?

SPEAKER_02And so I do worry that something could each first of all, each component could be not necessary, but by giving up all of them, we've lost something. And I also just worry like, even if what AI provides is accurate, there's no hallucinations, it is appropriate that we will lose our competency. Because for me, Dan it comes down to not only is it right, not only is it evidence-based, but is it within our competency?

SPEAKER_02Because we can have a recommendation that is correct for this person, is evidence-based. But if I'm in a median and someone goes, Oh, you you recommended PCIT, what is that? And I go, I don't know. You know, I it's one thing to say, well, it's an evidence-based approach for parent training. I'm not an expert in it, but I do know there's a lot of literature to report to to support it. And I do know this person I'm referring to, they they have an understanding of it and they can tell you much more about it.

SPEAKER_02That's defensible, but to me, just to provide some recommendation that you have no idea, you've no idea, but you don't have enough to like and like discuss it, that's problematic.

SPEAKER_01Yeah, I I going back to your earlier point, I do make it when I when I present, I talk about swimming in the context, which I'd borrow from a couple historian friends of mine who are absolutely you know, hate AI. And even though I thought, well, I mean, it would be so great, it could allow you to scan through so much more. And they said, you know, the value though is not just seeing the old newspaper article, it's seeing the articles that surround it and even the ads, because it sets the context. And so obviously that's for historians, but aren't we a historian of the child?

SPEAKER_01Not ancient history, but very modern recent history, but it's history nonetheless. And, you know, knowing that, say, COVID was going on in 2020 can certainly do some explanations about maybe while there was a grade slip or something, you know, in the background history. And yes, AI might know that, but it's not it's not gonna know it unless you specifically maybe call on it. It's just gonna spit out the numbers and whatever its generic interpretations are. So, you know, to your point, I think there is that and that value.

SPEAKER_01And that kind of coming back full circle to what we originally started our podcast with, right? These are the things that supervisors and supervisees need to have discussions about. It isn't always about the product. And I think supervision in the age of AI is gonna get into a more philosophical approach, just like if you're learning a new counseling technique and your supervisor starts digging in the weeds of what the theory means, not just that you're a technician, but what's the point behind all of this and what at the core makes you a professional?

SPEAKER_01And I think that's the only way that you really stir change and the motivation to spend the extra time it takes. Because this is this is the hard work stuff, right? We are trying to convince somebody where there's a literally an easy button that you can press and bypass the hard work for anything, almost anything anymore. And so, how do you convince them not to be tempted and press that button? And I think the only way you do it, it comes from within them and having an understanding of not eat it, it's good for you, but knowing why it's good for you and what is behind that.

SPEAKER_01And I think if we don't establish that with our next generation, they may tell you all they want in supervision or graduate training, but they're using AI. They they are shortcutting it. Um, because I know I would have when I was in grad school. I definitely would have shortcut it a little bit because everybody knows grad school is a devious kind of invention to pour more things than you're really capable of taking in in the time that they give you. Devious.

SPEAKER_01Some could argue tedious too, but I think that's what I would

SPEAKER_02argue. It's drudgery, it's largely drudgery and busy work.

SPEAKER_01It could be, yes.

SPEAKER_02I'm and I'm a professor. I still think a lot of this stuff is is busy work unfortunately.

SPEAKER_01We're seeing rebellious Adam in grad school right now. And if he had a I

SPEAKER_02was very compliant, very compliant.

SPEAKER_01I I just see you, I see you gloming onto AI in a half a second in grad school. But to the point, right? Well what do you think about that as far as uh supervision kind of morphing from a technician making sure that everything's going right, you're doing things ethically, but now taking maybe even more so than than typical, the bigger picture and discussing professionally like what it means and what should AI maybe help with, and maybe what it doesn't belong in.

SPEAKER_02Well, I hope that supervision and practice will shift more to showing people how to what it looks like to conve convey empathy. And what does it look like to listen and to advocate and to, I mean, I know I'm getting touchy feely here, but you know, if we can automate some of this assessment and report writing, we can get to some of the things that school psychologists have always said they want to do. And we can we can train and model how to do those things for the trainees as which are more soft skills than than hard skills.

SPEAKER_02And and the best mentor I have was actually a pediatrician. And the best and the reason why he was the best mentor is he taught me how to be patient-centered. So, and just one example is he had me schedule follow-up calls with every single child I I did an assessment of. And he just the way he interacted with people and just made them feel I just really wanted to emulate. So there could be a move away from hard skills to more to like this is how you this is how you really you know treat people.

SPEAKER_01Yeah. Well, I I think that it that's the hope of at the end of the day. My jaded self, uh, where we've commodified all of this, is that hey, we have 200 avowals and good, look, now you know, 100 was really pushing it for people and it was really stressful, but now we can get 200 out of them because they have this AI. And they have this automated cognitive assessment thing, and uh, you know, join the factory line. Uh, and if you don't want to do that, then we don't really need to hire you guys because the federal mandate says we only need to do these certain special ed uh vows, and otherwise you're discretionary, you're a discretionary position.

SPEAKER_02Well, I'm gonna I'm gonna plug then, Dan, because I have I've blogged about this on multiple occasions at this point. The Red Queen effect written in the communique about it. And I also just did a blog today, which for the listeners will be a couple months ago talking about AI could change everything and it could improve our lives, or it could just perpetuate all the things that are wrong in our world. And it really is up to us to make sure that we that the things we want manifest.

SPEAKER_02Uh so we're not just doing 200 more assessments or 100 more assessments. They say we need less school psychologists, that we fill the roles that we can and that we improve student lives through making sure that that happens.

SPEAKER_01So for our listeners that aren't maybe up on the latest NAS communique, though I know you wrote that maybe a year ago, Red Queen. What what exactly is the Red Queen?

SPEAKER_02The Red Queen is is an idea that was that came out of evolutionary um biology. And it's the idea that with through evolution, no animal or no organism truly gets ahead of any other organism because as they evolve and they they pull ahead, the other animals or organisms evolve to stay ahead. So so basically, I'm a I am a tiger and I am faster and I start to get all the gazelles, while the slower gazelles die off and the faster ones breed, and then before you know it, that faster tiger is just getting the same number of gazelles.

SPEAKER_02So no matter how fast an an organism progresses in its capabilities, it never really gets ahead. And it's been applied to technology that is no matter how good technology gets, we never really get ahead because everything we're just expected to move more quickly. The first time I read about it was the person was talking about a man sits in London traffic in his car during rush hour moving no more quickly than the person in a horse and buggy did 200 years ago.

SPEAKER_02So we have all this technology and it's supposed to be great, but but we just it never really it does, of course. We have infant mortality rates that are much lower. I don't mean to I don't mean to make light, but in a lot of ways everything speeds up around us. And so my fear and the fear of others is that AI, if it speeds things up, then they'll just add more caseload, for example, because it won't be like now you can go to working the 40 hour a week that you're actually paid to do.

SPEAKER_02They're gonna say, no, now you have 200 evaluations instead of 100, whatever.

SPEAKER_01Right. Yeah, okay. Yeah, I think that's uh it's a very good point. And I think in a subtle way, and this is also something that could be brought up in a supervisory AI world, is the issue of advocacy, right? So in the end, at least now, humans are the ones in charge, and we get to make the rules and we get to influence the standards and the regulations and the laws through our advocacy, and maybe passing that on into the next generation and its importance, maybe even more so, and pointing out really what are the advantages of not having somebody say doing all the assessments all the time, but spreading it out to using, for instance, the NAS practice model, which is a very distributed um way of doing school psychology and puts an emphasis on a lot of other things like consultation and intervention and even systemic changes, where that could be even the exciting part, right?

SPEAKER_01With AI being able to see whole pictures that was just really hard to do with all the data out there and encapsulate it to where you can get your mind around and make good choices about systems change or how you want to address with the new time that maybe you have, uh, because AI is able to take some of that slack up. But uh there's just so many opportunities that I had in supervision to discuss things with supervisees that I don't know if that's always the standard, you know, the the idea that you're always focusing on cases, which of course is very warranted, especially uh when they are first starting out in the field, but still making time because I I've also over the years, you know, we've had supervisees and I mean supervisors who are very good ones, but they are also really busy.

SPEAKER_01And so they're they're supervising practicum or interns, but it's really just focused on the caseloads, making sure that it's competent and then passing it on, and they're not spending the full time that they're supposed to. They do, but we're in the same office, we're always talking back and forth, and and that's really not supervision. And it is, but it isn't the full idea of supervision, which is to give time to really discuss things beyond just the day-to-day, I got this report done, or there's this little tricky thing about it, but more the theory behind it and the some of the things that we've been talking about uh earlier in this episode.

SPEAKER_02Well, I coordinate supervision. I gotta say, 90% of supervisors are fantastic and they do a very good job, despite most of them having no training and and supervision, you know, even I had it in the doctoral program, although I I don't think that I it made me a good supervisor necessarily, but the majority of them are phenomenal and are and are so giving up their time. And so yeah, I'm always impressed by our the supervisors that I do.

SPEAKER_01Well, and and I don't, like I said, you know, even those that maybe didn't allow as much time for the really deeper stuff, I don't I felt they did a good job in preparing our students. You know, we we had a at an or uh university had a really good reputation of having really good graduates, and that's a large part of the good supervision that they did receive. But does it need to go even further when we're talking about the age of AI and its impacts that maybe previously really good supervisors didn't have to go to that level because it wasn't there, right?

SPEAKER_01It wasn't an option. And I think that's that's what I'd be interested to kind of hear what you have to say about that.

SPEAKER_02I think there's a lot of ways that you can use AI to improve supervision. I've been talking about all the cautionary stuff, which I feel like I always need to lead with. And maybe it's good people anxious, and then you say, no, okay, now here's how you mitigate, and here's how you here's how you can use it to improve.

SPEAKER_01Maybe a good part two kind of thing. We seem to the lapse into a couple good series. We had the Gemini Notebook uh just a couple episodes that I think once again really did warrant. Probably could have done another two or three. We may, we may in the future, the way that they are really improving the Gemini Notebook. So, you know, I guess ours was more of a theoretical how is supervision being done or maybe needs to be done in the age of AI. And then uh maybe a follow-up episode of really talking about what are some of the real advantages of uh supervision in AI.

SPEAKER_01I mean, I'm just thinking off the top of my head, some of the counseling AI programs that are out there uh training-wise, are just from what I understand, just incredible and provide feedback that I wish I had as a graduate student. And I'm sure you've heard of other sorts of things too, or thought of other novel use cases of AI within a supervisory context that probably really enhances our skill sets. So

SPEAKER_02yeah. And I'll any

SPEAKER_01uh final

SPEAKER_02shout out to Carlos at Thinkai. He's a professor at U at one of the Texas universities. I can't remember, but he's he's doing a lot in this space uh for counseling psychology. But um yeah, next time if we do have time for a part two, I'd really like to dig in a little bit more about some of my practical the framework that I've been working on and and talking about, about some things that supervisors can do to really promote success amongst their supervisors while using AI and ways they can use it to improve supervision.

SPEAKER_02And and then I was hoping I could do a couple uh plugs as well, if you don't mind.

SPEAKER_01Hey, it's our podcast. We can plug away all that we want. So

SPEAKER_02by the time by the time this comes out, I will I should have my blog on this topic, AI and supervision. And uh I don't know if this will be out before December, but I'm pretty sure that in December I'm gonna be doing a training through the trainers of school psychologists on this topic. Uh so if anyone's interested in either of those, check it out.

SPEAKER_01I'm already starting to log in, Adam. So thanks so much, everybody. We appreciate you listening uh to another of our episodes. Hopefully, came away with a little bit of considerations as far as how supervision might be encountered in the age of AI. And with that, thanks again for listening to another Beyond the Bot AI and in mental health.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. THROWBACK EPISODE: The Difference Between Feeling Gratitude and Performing It

    Nov 13, 202653 minAudio

    A quick note for ABCT attendees: Paul and Ray are getting ready for the ABCT Conference in Baltimore, Maryland! Stop by the Psyched to Practice booth at Booth 305 at the Baltimore Convention Center and Hilton Baltimore...

    Credit the people on this episode

  2. Beyond the Bot: Lonely, Anxious, and Talking to AI

    Nov 6, 202636 minAudio

    AI chatbots are becoming more than homework helpers. For many teenagers, they’re turning into confidants, sources of emotional support, and even substitutes for human relationships. But what happens when young people...

    Credit the people on this episode

  3. Practice in Action: Turning Play Into Progress

    Oct 30, 20261 hr 12 minAudio

    What if the games already sitting in your therapy office could do more than help kids pass the time? In this episode of Psyched to Practice, Paul Wagner and Dr. Ray Christner explore how therapists can use everyday...

    Credit the people on this episode

  4. Beyond the Bot: From Research Overload to Real Understanding: The AI Tool Worth Exploring

    Oct 23, 202634 minAudio

    What if you could turn a collection of research papers into a podcast, create practice questions from professional guidelines, or build an infographic from dozens of sources? In this episode of Beyond the Bot, Dan...

    Credit the people on this episode

  5. Practice in Action: Beyond CBT, Integrating Different Approaches for Better Outcomes

    Oct 16, 202654 minAudio

    CBT has strong research behind it, but does that mean it’s the right approach for every client? And what should therapists do when progress stalls? In this episode of Psyched to Practice, Paul Wagner and Dr. Ray...

    Credit the people on this episode