Chapter 20: Counselling Theory & Practice

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome to the Deep Dive.

Today we're taking on one of the most fundamental yet

really challenging areas of psychological research,

the scientific evaluation of the relationship itself.

That's right.

We're stepping into the very complex world of counseling where that human connection is assumed to be the core mechanism of change.

But we're scientific validation actually proving that those relationship qualities cause the outcome.

Well, that's proven incredibly elusive.

It led to decades of real scientific struggle.

So the mission for you, the learner, today is a critical one to understand how these early psychologists tried to measure the essential really subtle ingredients of a helpful humanistically oriented relationship.

What were the tools they built?

What did the research actually prove?

And maybe more importantly, what were the profound scientific paradoxes it exposed about the nature of human health?

Because when we talk about agencies offering personal help for life problems, we're talking about services that often rely on this intense personal contact as the sole path to resolution.

The key scientific question isn't whether people feel better.

It's whether we can isolate and prove which specific observable behaviors reliably cause constructive personality change.

And that, well, that requires a level of scientific rigor that the field just struggled immensely to achieve.

Okay, let's unpack this.

And a good place to start is defining our terms.

Because the labels counseling and psychotherapy are, you know, they're often used interchangeably.

But historically, and especially for the research we're looking at, they have very distinct flavors.

You really did.

Stuff Larry, way back in 1965, highlighted some of these characteristic differences, even though, you know, we all acknowledge there's massive overlap in practice today.

So what was the first big difference?

Well, the origin in the setting were key.

Counseling usually came from psychologists working in academic or educational settings.

So think school or vocational guidance.

Okay.

Psychotherapy, on the other hand, historically leaned more toward a medical orientation.

It was often practiced in clinical or psychiatric settings.

And you can see that difference just in the language they use, right?

Right.

For the recipient.

Exactly.

Counseling recipients are generally called clients.

And they're less likely to get a formal psychiatric diagnosis than patients in psychotherapy.

It reflects that whole medical heritage.

The goals were different, too.

Hugely different, at least in theory.

Counseling emphasizes relatively modest specific goals,

often related to fixing a deficiency in the client's current role.

So like improving the study habits of an underachieving student.

Whereas psychotherapy was aiming for something much bigger.

Much bigger.

Traditionally, it was aiming for deeper, more pervasive personality change that would impact, you know, all facets of a person's life.

And methodologically.

Counseling tends to involve fewer sessions, sometimes just a handful.

There's less rigorous exploration of the client's distant past.

And a much greater concern with cognitive clarification.

Meaning helping the client understand their problems logically, consciously.

Right.

Rather than getting deeply tangled up in all the emotional ambiguities that might be rooted in past trauma.

But, you know, it's important to remember these are just typical characteristics.

A skilled counselor can absolutely go deep.

Sure.

But the research we're focusing on, this whole attempt to validate the relationship factors, it really grew directly out of that humanistic, client -centered model you find in counseling theory.

Which brings us right to the central assumption of humanistic counseling.

This is the foundational belief that drove decades of research and, well, ultimately created the biggest scientific challenges.

That foundation is rooted in a fundamental humanistic ethic.

It's the belief that every single individual has this inherent potential for growth and self -actualization.

And that leads directly to this core assumption, which practitioners often just accepted without much criticism.

Yeah.

That a specific type of counseling relationship, one with certain universally good qualities, is just inherently good and effective for any client, no matter what their issue is.

It's an incredibly optimistic idea, isn't it?

That the right kind of human connection is like a universal solvent for psychological pain.

It is.

But as our sources point out, this whole professional tradition grew up largely before any real scientific validation.

So the scientific community needed something measurable, something empirical to test this huge global assumption.

And that's where Carl Rogers comes in.

Exactly.

In 1957, he tried to formalize this assumption so it could be evaluated empirically.

He argued that there are specific relationship conditions that are, this is the key phrase, necessary and sufficient for a constructive personality change.

Necessary and sufficient.

Yeah.

And he said they have to be present in all forms of genuine helping relationships.

So he laid out three central conditions that really became the gold standard for measurement, mostly through the work of Truax and Karcuff in the 60s.

Let's walk through them because the logical and practical problems researchers ran into are just fascinating.

First up, congruence, often called genuineness.

This just means the helper has to be truly real or authentic in the relationship.

A high level of congruence doesn't mean the therapist blurts out every single feeling they have.

But critically, they must not deny them.

They have to be freely and deeply himself, not putting on some kind of phony or defensive professional face.

It's all about sincerity versus a professional facade.

Okay, but here's the immediate,

just crushing logical problem for a researcher trying to rate this on a five point scale.

How can an external rater possibly know if a therapist is truly being himself?

They can't.

It seems impossible without knowing their actual internal state at that moment.

You can't assess authenticity just by watching a recording.

You absolutely can't.

The raters can't read minds.

They can't access the therapist's private thoughts.

So in practice, what they had to do is rely on these indirect observable cues that they learn to associate with genuineness.

Like what?

Things like the therapist's manner of expression, subtle voice tones, specific speech disturbances that might signal some internal conflict, or even just unique turns of phrase that seems spontaneous and unscripted.

So the rating of performance of genuineness?

Pretty much.

Shapiro, in a 1973 study, even suggested that cultural differences in these subtle things could affect the ratings.

He found distinct scoring differences between British and

which suggests they were rating a cultural performance, not some absolute internal sincerity.

That already tells you the measurement of this core condition is fundamentally flawed.

Okay, what's next?

Unconditional positive regard or non -possessive warmth?

This one, also usually on a five point scale, requires the helper to warmly accept the client's experience as part of that person without imposing any conditions.

So a high level means you value the client as a unique individual, you care for them, but it's not contaminated by any evaluation of their specific thoughts or behaviors.

Exactly.

The therapist should be searching for the meaning of the client's thoughts within the client's own frame of reference, not expressing their own personal approval or disapproval.

The unconditionality is the key.

But again, the research immediately ran into problems.

Big ones.

First, the rating scale itself was kind of flawed because it mixed up general warmth with the conditionality of that warmth.

But second, and this is a huge scientific paradox,

research from Truax himself suggested that warmth might actually be most effective when it's applied conditionally.

Wait, wait.

So Truax, who's a key researcher for this theory, found that strategically applying warmth so, giving positive regard only when the client shows what he called good therapy behavior, was more effective.

That's what he found.

That's a complete empirical demolition of the unconditional requirement, isn't it?

It's a massive empirical challenge to a cornerstone of the theory.

The ideal of universal non -contingent acceptance seemed to be less effective than strategic warmth.

It suggests the relationship actually works through reinforcement and conditioning, which is, you know, completely antithetical to the core Rogerian idea.

Wow.

Okay, what's the third condition?

Empathic understanding or acuate empathy.

And this one's on a nine -point scale, so it's even more complex.

It is.

It demands two things at once.

The therapist's deep sensitivity to the client's current feelings, and the verbal skill to communicate that understanding back to the client in language that's perfectly attuned to them.

So it's not enough to just get it, you have to reflect it back perfectly.

Right.

And at a deeper levels, it even includes sensing feelings the client is only partially revealing or communicating meanings the client is barely even aware of themselves.

It's supposed to deepen the client's own self -understanding.

Which, again, introduces a huge logical problem for the rater.

To really evaluate the counselor's empathy against the client's full unstated experience, the rater would have to be incredibly empathic themselves.

It's this impossible three -way matching game between the client's feeling, the counselor's response, and the rater's evaluation.

But again, the practical research revealed the workaround, or maybe the operational failure.

A 1966 study by Truax found that if you edited out the client's statements from the recordings, it had no apparent effect on the empathy ratings given to the therapist.

Let that sink in.

The raters could rate the counselor's empathy without even hearing what the client said.

Yes.

They were just rating a performance, not an actual connection.

Precisely.

It indicates the raters weren't matching the response to the client's context.

They were relying on these superficial cues.

Voice quality, vocabulary, how smooth the interaction was, and these obvious correlates of attentive responding.

So were they really measuring accurate empathy or just a professional performance of paying attention?

That's the million dollar question.

And then finally, you look at all three scales together and you see this huge issue of conceptual overlap.

I mean, if a counselor shows zero empathy,

it's logically impossible for them to show high non -possessive warmth because high warmth requires searching for meaning within the client.

The concepts are all tangled up.

They're completely interdependent, even though they were supposed to be three separate, measurable dimensions.

So we have these three complex interwoven concepts and they're being rated by observers using manuals and short clips of interviews.

Before we even get to outcomes, we have to ask the basic scientific questions.

Are these ratings reliable and are they valid?

And that's where things get even murkier.

Let's start with reliability.

So consistency.

Right.

Inter -rater reliability was all over the place.

For empathy and warmth, the ranges were generally OK in controlled studies, sometimes hitting 0 .95, which is great.

But for genuineness.

I'm guessing that was the problem, child.

It was a disaster.

The range went from 0 .25 to 0 .95 and it was usually much lower than the others.

A correlation of 0 .25 is statistically abysmal.

That means the two judges are barely agreeing more than chance.

They're not even watching the same performance, let alone judging it by the same criteria.

Exactly.

It suggests that at times the just confirms that logical difficulty we talked about.

This inconsistency really suggests that raters are relying on nonverbal cues and other subtle things that are not explicitly spelled out in the rating manuals.

The whole thing depends on the rater training, not the scale itself.

So then we move to validity.

Are these scales measuring what they claim to measure?

Researchers did some clever experiments using only parts of the information.

And there's evidence that reliable ratings could be made using just visual cues alone.

Without sound.

Without sound.

Or just the therapist's words, for warmth and empathy.

Or even just from the raw transcripts, without voice tone or visuals.

So this implies that different parts of the therapist's behavior all kind of move together in this redundant way.

If they look warm and attentive, they probably sound empathic even if the content isn't really deep or accurate.

Right.

It's a halo effect.

The style of responding is overwhelming the substance and cueing the rating.

And this leads to what the source calls an alarming possibility, which just throws the entire validity question into doubt.

This is the finding by Karasina and Vickrey in 1969.

They found that empathy ratings correlated strongly with the number of words spoken by the therapist and their proportion of the dialogue.

You're kidding me.

Empathy should mean deep, accurate understanding.

Not just being chatty.

I know.

So if I just talk a lot and keep the conversation flowing, I get rated as highly empathic.

That's an operational nightmare for a concept as subtle as empathy.

It is.

It suggests that raters are forced to rely on these superficial, objective behaviors.

How much the therapist talks.

How fluid they sound.

Because they're just easier to quantify than the abstract idea of genuine empathy.

And there were also major statistical critiques, right?

Yeah.

Shinsky and Rappaport in 1970, they warned against analyzing ratings from a small number of therapists who are heard over and over again with different clients as if they were all independent scores.

Because they're not.

It creates a problem called correlated errors.

Exactly.

If you have ratings from only four therapists, but dozens of clients, treating all those client outcomes as independent measures just inflates the power of your results.

You've really only tested the method as delivered by four people, not a large random sample.

It's a major concern.

Okay.

What about the idea that these three dimensions,

congruence, warmth, and empathy are independent?

Truax claimed they were.

That was the theoretical claim, yeah.

And you can design experiments where therapists deliberately vary warmth and empathy separately from genuineness.

But in ordinary therapy, the three scales are almost always positively inter -correlated.

So if a therapist is high on one, they tend to be high on all three.

Right.

Which lends a lot of credence to the idea that raters are actually just scoring a single global good therapist factor rather than three distinct dimensions.

Which suggests that maybe Roger's three concepts, while clinically useful, might not be the best scientific framework for describing therapist behavior.

It does.

And when researchers looked at broader interpersonal behavior outside the narrow Rogerian framework, they found some overlap, but also some really significant gaps.

What did they find?

Well, broader interpersonal research identifies a major dimension of solidarity, which lines up pretty well with Roger's warmth.

Another is activity responsiveness, which might relate to empathy.

But there's a major dimension in that broader research that has no counterpart in Roger's scheme at all.

And that's dominant status potency.

This measures how much control or influence one person has over the other.

And its absence is fascinating.

It's so revealing because it reflects the Rogerian theory's fundamental, almost ideological rejection of dominant or submissive behavior.

The ideal Rogerian therapist sits right in the neutral middle of that dimension.

But scientifically, dominance is a central part of human interaction that the Rogerian framework just ignores in its pursuit of equality.

It's a huge blind spot.

A huge variable that's just left out.

And then before we move on to outcomes, there's the concept of self -disclosure.

Girard's work in the seventies showed that

or transparency, which is conceptually like genuineness, has powerful and predictable social effects.

What does it do?

It stimulates reciprocal self -disclosure.

If the counselor acknowledges their own feelings or is transparent, it encourages the client to do the same, to go deeper.

This just reinforces the idea that the helper's behavior is actively shaping the client's behavior through social exchange.

Which brings us to the ultimate core issue.

Do these globally measured conditions actually lead to positive client change?

The crucial question,

if we measure these conditions, what happens to the client?

So researchers use this wide battery of measures, psychological tests, hospital staff ratings, and what they considered the most trustworthy source, objective data.

Things like how long someone stays out of the hospital or their academic grades.

Right.

And the standard practice was to use a diverse battery on the principle that many different signs pointing the same way are more credible than any single one.

But that approach was heavily criticized.

Very heavily.

Because many studies, especially from TrueAxis Group, would unjustifiably report multiple correlated scores as if they were independent measures of change.

Leading to these spuriously high success rates.

Like saying 14 of 19 measures showed change.

That sounds great, but it's a statistical sleight of hand if 10 of those are just different ways of measuring the same tiny improvement.

Exactly.

The broader critique is that this broad spectrum approach is just too vague.

A much better strategy is to specify a precise objective outcome criterion -like improved grades for underachievers and then combine that with measures directly relevant to that goal, like test anxiety scores.

So the research on outcomes fell into two main design types.

Let's start with design type one, comparing high -conditions counseling with untreated controls.

This design actually had some promising, if tentative, results.

High -conditions counseling seemed to help delinquents stay out of institutions, and it improved the academic performance of underachievers.

So it proved that this type of counseling was better than nothing.

But the results were disappointing in more severe cases.

Very.

The Big Wisconsin Schizophrenia Project, led by Rogers himself, only showed a non -significant three -month difference in hospitalization over a year compared to controls.

It suggested that for

hospitalized populations, just having a good relationship wasn't a potent enough therapeutic agent on its own.

There was another study with a mixed mental hospital population, but it had a big catch.

It did.

Karcoff and Truax reported some efficacy, but there was also evidence of actual deterioration in the treated group.

Plus, staff bias was a huge problem, since the ward staff who did the ratings knew which patients were in the treatment groups.

Okay, now for design type two, the correlational design.

This is where they retrospectively split clients into high and low conditions groups based on session ratings and then compare the outcomes.

This is the more stringent test of the core hypothesis.

And this is where the methodology gets really murky and heavily criticized.

The procedure usually involved rating tiny three -minute segments from a very small number of sessions.

Wait, how small?

Two segments from three sessions out of a total of 24.

They'd use that tiny six -minute sample to calculate a mean rating for the entire course of treatment.

They were judging months of an intense therapeutic relationship based on about 18 minutes of recorded tape.

That's the entire basis for judging whether a client was in a high or low conditions environment.

That's it.

It's an astoundingly small sample of behavior to base such huge conclusions on.

So what did the big review of these studies find?

Well, a major review by Shapiro in 1976 found that while, yeah, the majority of measures trended in the expected direction, high conditions tended to have better outcomes, the differences rarely reached statistical significance once you corrected for all the comparisons they were making.

And here's the crucial failure that really undermines the whole thing.

This is the killer.

When you look at the most objective data, the institutionalization rate time spent outside the hospital, it failed to show significant results in any of the three studies where it was obtained.

None.

So the overall assessment is that the evidence for a reliable link between the three conditions and a favorable outcome is only suggestive and weak.

It certainly doesn't justify the optimistic claims that were being made.

And the problem with any correlation, as always, is causality.

The association might just be because of the clients themselves.

Right.

Maybe a client who is already getting better is just easier to be empathic with.

The client's behavior causes the therapist's high -rated performance, not the other way around.

Which highlights the absolute need for a true experimental comparison.

The gold standard.

To really test this, you have to randomly assign clients to groups of therapists who are known to differ in their levels of these conditions.

And that stringent design, well, it hadn't been done at the time of these reviews.

And without that true experiment, the positive results you see from just comparing treated versus untreated people might simply do the placebo effect.

Just getting attention from another human being.

Exactly.

If we can't show that clients assigned to good relationships do better than those assigned to bad relationships, then the specific quality might be irrelevant compared to just receiving structured human interaction.

Okay.

So since the global efficacy was so hard to prove, a lot of researchers turned their focus inward to the minute by minute processes within the session.

And this brings us to the scale of client self -exploration.

This is a direct nod to Rogerian theory.

The idea is that the three therapeutic conditions aren't the cure themselves, but they create the environment that facilitates the client's own deep exploration of themselves.

That's the real engine of change.

So TrueAxe and his team developed a scale to measure this.

It assesses how much and how spontaneously the client discusses personally relevant stuff.

Their feelings, values, fears, life choices.

It's the operational measure of what happens when a client really opens up.

And interaction studies where they actually manipulated counselor behavior in single interviews revealed this really complex dynamic.

It confirmed the client has power too.

The relationship is mutual.

So what do they find?

Well, high client self -exploration could only be maintained in the face of, say, reduced empathy from the therapist if both of them started out functioning at a high level.

But conversely, a skilled client who deliberately lowered their self -exploration could actually drag down the therapeutic conditions offered by a low -functioning therapist.

But the good counselors were different.

Yeah, they seemed remarkably unaffected.

They'd keep trying to offer empathy, even to seemingly trivial stuff.

They had this stable commitment to the process.

Now we circle back to that huge conflict with the idea of unconditional acceptance.

TrueAxe's work showed that even Rogers himself varied his expressions of empathy and warmth based on what the client had just said.

Which fundamentally suggests a mechanism of selective reinforcement.

The therapist, whether they mean to or not, is reinforcing certain client behaviors.

Behaviors that look like high self -exploration, or even just the client's expressive style becoming more like the therapist's.

It hints at imitation or modeling as a way clients change, which is not a Rogerian idea at all.

The client is learning the language of therapy.

Exactly.

But it's complicated.

It's hard to tell who is reinforcing whom.

Is the client rewarding the counselor for being empathic, or is it the other way around?

It's a mutual feedback loop.

So to map this influence more accurately, Bandura and his colleagues developed a detailed coding scheme to classify therapist responses as either approach or avoidance.

This is crucial for understanding how these subtle communications can shape the whole direction of a session.

Okay, so let's walk through this conceptual model.

Approach reactions are things that encourage the client to keep going with their current topic, like hostility or dependency.

Right.

And there are five main types of approach reactions.

First is approval.

So the client says they were mad, and the counselor says, under the circumstances, how could you have felt otherwise?

It legitimizes the feeling.

Okay, what's next?

Exploration.

Asking for more detail.

Aggravated.

Can you tell me a little about that?

It signals the topic is important.

Then there's instigation, which is directing the client back to the topic.

Yes.

Let's get back again to that evening, that irritated feeling you had.

Forcibly reinforcing the topic.

Then reflection.

Just simple mirroring.

You get mad.

And finally, labeling, which are interpretive statements that give a framework for the feeling, like maybe some of your resentment is partly displaced from your mother.

So all of those signal, I want more of this.

What about the avoidance reactions, the ones that try to shut a topic down?

There are five of those, too.

First is disapproval, a direct judgment.

Just for that, you hit her.

It has a heavy, suppressive effect.

Then topical transition, just changing the subject completely.

My mother annoys me.

How old is your mother?

Ignores the emotion entirely.

Then silence, a non -responsive four seconds or more.

Ignoring where you shift focus away from the emotional content to the practicalities.

And finally, mislabeling, an inaccurate interpretation of feeling, which can be really frustrating for the client.

And the immediate effect of these was clear in Bandura's work.

Totally.

Clients were way more likely to continue expressing hostility or dependency after approach responses than after avoidance responses.

It shows the counselor's immediate behavior is a powerful, selective determinant of what happens next.

But trying to show that these responses shape behavior over the entire course of counseling was less successful.

It produced equivocal results.

Bandura thought it might be because a category like hostile is too broad.

Therapists probably distinguish between appropriate and inappropriate hostility and respond differently.

So the real reinforcement schedule is way more complex.

Which brings us back to that dyadic interaction.

The relationship has to be seen as this mutual interactive process.

It's not simple stimulus response.

And that complexity demands a real shift in the research focus.

A lot of the research difficulties we've talked about, the low reliability, the inflated stats, the failure to prove causality, they all stem from this failure to specify what intervention produces, what change, and what type of client.

That global assumption of universal efficacy was the biggest scientific roadblock.

It was.

And this realization spurred a critical vital shift towards specificity in counseling research.

Moving away from the good relationship idea to targeted intervention.

The first step had to be setting reliably measurable and clearly relevant goals that were actually achievable and objective.

Right.

And we need to look at some specific examples where this shift paid off dramatically.

Take the example of delinquency.

The goal is objective.

Keep the client out of further trouble.

Low recidivism.

So what worked?

Persons, in 1966, used a highly specific structured group therapy sequence.

He started with support, then moved systematically to interpretation, reinforcement, role playing, stress induction, and community discussion.

So a very structured, active approach.

Very.

And it heavily favored the treated group.

They had better disciplinary records and a significantly lower recidivism rate after a year.

It was a clear case of a structured, goal -oriented intervention being more effective than just an unstructured, good relationship.

What about another example?

Massimo and Schor's program for delinquent boys.

It was vocationally oriented so the counselors were involved in all aspects of the client's life, focusing on practical, job -oriented assistance.

And the results?

Fewer and less serious misdemeanors and better academic scores.

This contrasts so sharply with the earlier Truax study, which just linked success to high empathy and this vague need for a good relationship with authority.

The specific real -world intervention mattered more.

Okay, second example.

Facilitating hospital discharge for psychiatric patients.

Dryblatt and Weatherly, in 1965, found that very brief contacts, up to 12 sessions of just 10 minutes, especially those that did not discuss symptoms but focused on enhancing self -esteem, led to speedier discharge than control conditions.

That's remarkable.

How could that work?

The speculation was that it enhanced self -esteem, helping patients better navigate the hospital system.

What's amazing is how sharply this modest intervention contrasts with the intensive, failed Wisconsin schizophrenia project, which tried to build these deep, unstructured relationships.

The brief, focused, goal -oriented intervention was simply more appropriate and effective.

It was.

And it also suggests that more directive techniques might be useful for certain populations.

Another study found that using confrontations in group counseling actually increased positive behaviors in chronic schizophrenic patients.

For some clients, you need that structure to break through.

And the third, widely studied example.

Underachieving students.

The goal is objective academic improvement, GPA.

A big review by Bednar and Weinberg in 1970 found about half of the studies showed significant GPA superiority for counseled students.

And crucially, they identified factors associated with success that ran directly counter to that non -directive Rogerian ideal.

Yes.

The factors included counseling length of more than 10 hours, using group procedures, and most importantly, structuredness.

What do they mean by structuredness?

Directive, authoritarian, academic, and prescriptive approaches.

Not non -directive, client -centered, or purely effective approaches.

The science was pointing toward, if academic success is the goal, tell them what to do and how to do it.

And this led to multimodal treatments.

Right.

Mitchell and Eng studied test -anxious students who also lacked study skills.

Desensitization alone reduced their anxiety, but only the multimodal conditions, which combined behavior therapy and counseling, improved both anxiety and study skills, which led to academic improvement.

So you have to combine the practical, remedial help with the counseling that addresses the emotional dynamics.

Specificity means adding tools, not just relying on one global relationship quality.

Okay, so these successes established that counseling can work, but only when it's tailored and specific to the problem and the goal.

And this leads us to the final, most profound step.

Challenging the assumption that the same method works for all clients.

The emergence of client -treatment interaction effects.

This is where specificity truly shines.

It shows that the client's own internal characteristics dictate which methodology is appropriate.

The focus shifts entirely from the ideal counselor to the ideal match.

So Gilbreth's studies in the late 60s looked at student dependency needs.

What did he find?

He found that structured counseling was highly effective for students with high dependency needs, clients who need direction and guidance.

Conversely, unstructured counseling, the Rogerian ideal, was effective only for students with low dependency needs who were ready for self -direction.

And the failure mode was clear.

Very clear.

If students got the inappropriate counseling structured for the independent students, or unstructured for the dependent ones, they did no better than the untreated controls.

It's powerful proof that therapy can be actively harmful or just useless if not matched correctly.

A similar pattern emerged with anxiety level.

Brown found that students who admitted to high manifest anxiety were helped by unstructured counseling.

It may have given them a safe space to process it, but students with low manifest anxiety who tend to deny their distress were helped by structured counseling.

And inappropriate placement led to increased anxiety.

DeLaredo's 1971 study added personality type to the mix.

Yeah.

Studying students with interpersonal anxiety, he found that rational emotive therapy, a very directive cognitive approach, was effective with introverts.

Client -centered counseling, the unstructured one, was effective with extroverts.

And systematic desensitization, a pure behavioral method, worked for both.

The huge takeaway there is that when matched correctly, the counseling methods were just as good as the behavioral method.

But with the wrong client, they were no better than a placebo.

It empirically demolishes the idea of universal application of the humanistic approach.

And finally, these interaction effects aren't just about personality.

They even looked at socioeconomic status.

Love and his colleagues did.

For children with emotional and behavioral problems, they found that a concrete intervention information feedback using videotapes for parents improved grades for children in upper socioeconomic levels.

But for lower SES families.

More orthodox parental counseling focused on emotional processing and relationships was what improved grades.

It suggests that cultural or class differences dictate the preferred way of getting help.

Right.

The concrete factual approach might resonate better with higher SES parents, while the emotional process -focused approach suited the lower SES group.

This evidence is just profound.

It completely dismantles the idea of a one -size -fits -all, good relationship.

It firmly introduces the idea that the characteristics of the client determine the effectiveness of the type of help being offered.

So let's bring this deep dive to a close.

Despite all the literature and decades of effort, the main conclusion is firm.

Proven efficacy for broad global humanistic counseling practices has not yet been achieved.

Firm recommendations for general practice based on what we've reviewed today are just impossible.

The central assumption that we started with, that one type of relationship is universally good,

is fundamentally challenged and largely disproven.

The research undermined those cherished beliefs of unconditional acceptance.

We saw that unconditional acceptance doesn't seem to happen in practice.

The conditionality of the counselor's response operating through selective reinforcement and modeling those approach avoidance mechanisms is clearly a huge ingredient that drives the client's behavior.

And the evidence also undermines the global nature of Rogerian theory.

We have strong evidence that more structured approaches are helpful for dependent, anxiety -denying, or introverted clients.

While the Rogerian unstructured model seems better for independent, anxiety -admitting, or extroverted clients.

In short, the evidence overwhelmingly favors modest specific goals like improving grades or facilitating hospital discharge over these ambitious, vague attempts at far -reaching personality change that you can't even measure effectively.

So if the initial research seemed bogged down by all these methodological problems, it just reflects the immense difficulty in studying such a subtle, complex subject.

Researchers may have relied too heavily on these clinically -derived theories, starting with an idea of the good counselor, as Rogers and Truax did, and then trying to fit the science around that ideological premise.

So here is our final provocative thought for you to chew on.

Future research needs to pivot entirely.

Instead of starting with a clinical ideal, the perfectly -condruent, unconditionally warm counselor, it might be better to start by empirically identifying how effective counselors actually differ in their specific behavior from ineffective ones, regardless of the theory they claim to follow.

What do the winners actually do in practice when they're talking to a specific type of client?

The research has now reached the point where we can expect increasingly precise answers to that question.

What specific type of counseling works for what specific type of client, with what specific problem?