Chapter 6: Thinking & Problem Solving Processes

Welcome to Last Minute Lecture.

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

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

For complete coverage, always consult the official text.

Okay, let's unpack this.

We're gonna get into a core ingredient of what makes us thinking.

I mean, for centuries, philosophers stretching all the way back to Aristotle,

they viewed the capacity to think as the single defining attribute of humanity.

The essential thing.

And you see it crystallize with René Descartes, right?

The famous phrase.

I think therefore I am, exactly.

He made thought the fundamental distinction between mind and, well, just physical matter.

And that really elevates thinking

to the highest possible status in studying human behavior.

But for us, for psychologists trying to get a scientific handle on this, the term thinking is, it's notoriously slippery.

Yeah, it's a problem.

It occupies this really privileged position in our vocabulary, yet it's, I mean, it's such an ambiguous label for these internal processes we can't even observe directly.

And that ambiguity is exactly where we have to start our deep dive.

Because if we're gonna study thinking scientifically, we have to know what we're actually measuring.

The issue is, when people use the word think in just, you know, everyday conversation, they could mean five very, very different things.

It's true.

It's a kind of psychological umbrella term.

On one hand, you have it used as a synonym for remembering.

Oh, right, like, I'm trying to think where I left my pen.

Exactly, you're asking your memory, you're not solving some complex equation.

Or it can be synonymous with imagining.

You might hear someone say, I'm trying to think what my new job will be like.

That's all projection, visualizing future outcomes.

A third one, which is really common, relates to concentration.

When you bump into someone and they say, oh, sorry, I wasn't thinking what I was doing.

They just mean they weren't paying attention, an absence of focus.

Right, and the fourth one is highly subjective, using think to express a belief, attitude, or opinion.

I think everyone should be paid more money.

Yeah, that's a political or social stance.

It's not a cognitive process in the scientific sense we're interested in.

So that brings us to the fifth one.

The crucial fifth definition, and this is the one that actually points us towards scientific measurable psychology, is reasoning.

This is the process of organizing, planning, working out the details of a task.

So I am thinking about how to write this chapter

that's reflective, it's organized.

Precisely, and historically, most psychologists narrowed their focus right down to that last type, what they called reflective thinking, and effectively made it synonymous with problem -solving behavior.

That was the definition championed by people like Humphrey back in the 50s.

But a lot of people felt that was too limiting.

I mean, if we restrict thinking only to solving a defined problem, we miss out on a massive amount of internal mental activity.

Right, there's more to it.

So a wider, slightly looser definition was also popularized by figures like Osgood around the same time, thinking as simply the internal representation of events.

And that wider definition, that covers everything from, I don't know, classifying a rose as a flower to mapping out a chessboard in your head.

So our mission today is to start with that rigorous study of the narrower, measurable definition problem -solving and trace its origins in animal behavior.

Then we'll expand out to cover the wider internal processes of representation, categorization, concept formation, and ultimately, creativity.

So we're really charting the entire landscape of mental activity from simple trial and error all the way up to the highest forms of human creativity.

And the goal here is to give you a comprehensive overview of how psychologists approach this essential subject, providing that clear structure and those aha moments that make these complex topics really click.

Right, let's get into it.

The history of psychological inquiry into problem -solving, it really starts with this monumental conflict between two scientific giants.

It basically marks a fundamental divide in how we view the mind.

It really does.

On one side, you have the behaviorist, E .L.

Thorndyke, and on the other, the Gestalt proponent, Wolfgang Köhler.

In Thorndyke's work, which came first, it established this powerful school of thought we now call trial and error learning.

His findings, published way back in 1911, suggested that mental activity was, oh, a lot more mechanical than we tend to think.

And his methodology was all centered on the famous puzzle box.

You really have to picture this setup.

It was a small cage,

and an animal, usually a domestic cat, was confined inside, and the motivation was extreme.

The cat was hungry, and the only way out was to activate some kind of mechanical mechanism to get to food that was placed conspicuously right outside the cage.

Thorndyke rigged these boxes with increasingly complex escape routes.

In one version, for example, the cat had to claw at a loop of strain inside the box.

That loop was then connected via a pulley system to a latch on the outside.

And once that latch released, a weighted pulley system would raise the door just enough for the cat to squeeze out, so it's not a simple task.

Not at all.

And what did Thorndyke observe?

Well, the cat's behavior was slow.

It was laborious, and it looked completely random.

The cat would just push and claw and scratch and squeeze at every single part of the cage.

There was no apparent realization, no sudden moment of, oh, I see the lever.

He documented that the correct response, say clawing that specific loop, was stumbled upon pretty much by accident.

The second time in the box, the process was a bit quicker.

The third time, quicker still.

But it was always a slow, gradual descent of the time curve.

And this led him to a very strong and very famous conclusion.

He concluded that problem solving in animals is simply a mechanical process.

It's a gradual stamping in of correct responses and a stamping out of incorrect ones.

So the cat wasn't reasoning.

It wasn't thinking in our sense.

It was just forming habits.

Exactly, and this process was governed by what he called the law of repetition or exercise.

The more you do something, the stronger that connection becomes, and more famously,

the law of effect.

The law of effect is pivotal here because it's the direct philosophical and theoretical ancestor of all modern reinforcement principles.

It really is.

It stated that any response that's followed by a pleasurable state, like getting food, strengthens the association between the stimulus situation, so being in the box and that response clawing the loop.

And conversely, a response followed by an annoying state, like not escaping or getting a shock, that would weaken the association.

It was pure behaviorism.

The environment controls the outcome by reinforcing or punishing these SR connections.

Mine doesn't really enter into it.

But this mechanistic view was intensely challenged, and that challenge came from the Gestalt school in Germany and primarily from Wolfgang Kuhler.

Right, Kuhler argued that Thorndike's entire setup was fundamentally flawed.

He said, look, you've taken a cat, you've put it in this artificial restrictive environment, a box designed by an engineer, and you've forced it to try levers and pulleys, things that are completely unnatural for a cat.

So he's saying you've essentially forced trial and error behavior.

Exactly, thereby concealing the animal's true potential for intelligent thought.

So Kuhler's own research was very different.

It took place on a colony of apes and chimpanzees in Tenerife, and his findings were documented in his book, The Mentality of Apes, in 1925.

He designed food getting problems that were, as he saw it, much more naturalistic.

They allowed the animals to perceive the entire situation holistically.

And the most iconic example of this is the stick problem.

So picture this.

You have a chimpanzee locked in the cage, and a piece of fruit, let's say a banana, is placed clearly out of reach outside the bars.

Inside the cage, far from the fruit, is a short stick, too short to reach the banana.

And just outside the cage, within reach of that short stick, but not the chimpanzee's hand, is a longer stick.

So what does the chimp do?

Initially, it tries the simple, obvious things.

It reaches with its hand, fails, and then tries to use the short stick, which also fails.

And this often leads to signs of frustration, a moment of looking away, maybe wandering around the cage, but then something dramatic happens.

The animal suddenly stops.

It looks at the short stick, it looks at the long stick, and it looks at the banana.

Then in one coordinated, continuous movement, it uses the short stick to rake the long stick into the cage, and immediately, without any hesitation, uses that long stick to retrieve the banana.

And Kohler argued this demonstrated insight.

True reasoning.

The solution wasn't built gradually through accidental success.

The problem was solved internally, mentally, before the physical execution.

And he identified three key features that he said separated this kind of insight from simple trial and error.

First was suddenness.

The solution doesn't build slowly.

It appears dramatically, often in a flash.

And once it's achieved, it's retained permanently.

So if you put the chimp back in the box the next day, it skips all the random flailing and just goes straight to the stick raking solution.

The second feature was smoothness.

Once the solution is seen, the physical behavior sequence is executed continuously and smoothly.

It's an efficient plan, not the clumsy random movements you see in a trial and error progression.

And the third, the really critical feature, was that the solution precedes execution.

Right, the animal appears to have internally reorganized the elements of the problem, seeing the stick not just as a toy, but as a tool to reach another tool before it even initiates the physical action.

So we're left with this huge theoretical chasm.

Is problem solving purely a mechanical building of habits or does it involve this sudden holistic reorganization of the perceptual field?

And as so often happens in science when you have two extreme, very compelling viewpoints, the truth was really found somewhere in the middle ground.

Which is the concept of learning set formation.

Exactly.

The primary criticism that was leveled against Kohler was that he had ignored the ape's prior history.

He just assumed that the ability to use a stick as a tool was born only of that immediate perceptual situation.

But what if they had practiced using sticks before?

That's exactly what a researcher named Birch tested in 1945.

He used six young, naive chimpanzees.

Initially, only one of them solved the complex stick problem right away.

But then Birch gave the other five animals three days of unstructured,

spontaneous play with short sticks in their enclosure.

And during this play, they gradually developed the understanding that sticks could be used to poke and shove and pry.

When he retested them on the stick problem, the one requiring them to use the short stick to get the long stick, all of them solved it quickly and smoothly.

So that shows that past experience was absolutely crucial.

The play wasn't just random.

It prepared them to perceive sticks as functional tools.

And this really set the stage for Harry Harlow's definitive study in 1949, which provided the full compromise account.

Harlow wanted to systematically measure the role of past experience.

So he had monkeys solve hundreds of simple discrimination problems.

The setup was always the same.

Two objects, say a red square and a green circle, covered two food wells.

Only one object always covered the food reward.

The stimuli changed constantly, but the basic rule was fixed.

Always choose the correct object.

And Harlow visually documented this transformation in their learning curves.

When you look at the results for the initial problems, let's say problems one through eight, the monkeys showed a very shallow, low -performing curve.

They really struggled.

They made mistakes trial after trial, showing that slow, laborious trial and error learning, much like Thorndyke's cats did initially.

But here's the profound shift.

As the monkeys solved more and more problems of the same fundamental kind, the learning curves changed dramatically.

By the time they reached problems 257 through 312, the efficiency just spiked immediately.

They achieved what looked exactly like Kohler's insight.

It was incredible.

If they chose the correct object on the first trial, they made no mistakes after that.

If they chose incorrectly on the first trial, they just shifted to the other object on the second trial and were perfect from then on.

So they went from zero efficiency to near perfect efficiency in a single trial.

So was this trial and error or was it insight?

Harlow's groundbreaking conclusion was that trial and error and insight are but two different phases of one long continuous process.

So they aren't separate mental capacities at all.

They represent the orderly learned development of thought.

Right, and the intervening process, the secret sauce, if you will, is what Harlow termed learning to think or learning set formation.

The subject learns an organized set of habits,

a sophisticated mental framework that allows them to meet new problems of that particular category effectively and efficiently.

It's like the subject is constantly building a better operating system for its own brain.

That's a great way to put it.

It's an incredible insight, really.

It means the magic of insight isn't magic at all.

It's just super efficient, highly practiced trial and error that has been completely internalized.

So past experience doesn't just provide solutions.

It shapes the very way we perceive and approach new problems.

So if animals can be trained out of that mechanical trial and error and towards these efficient learning sets,

what kind of psychological hurdles do humans face when we try to solve novel problems?

Well, shifting to human studies, the investigation gets a lot more complex because now we can use language and introspection and all this prior knowledge.

Psychologists use two really classic problem situations to dig into the process of human solution.

One focused on intellectual complexity and the other on more practical physical manipulation.

Let's start with the first one studied intensively by the German psychologist Dunker, the radiation problem.

The core challenge was elegant and complex.

The problem was this, given a human being with an inoperable stomach tumor and rays which destroy organic tissue at sufficient intensity, by what procedure can one free him of the tumor by these rays and at the same time, avoid destroying the healthy tissue which surrounds it?

It's a beautifully structured challenge because it has a clear goal and clear constraints but absolutely no obvious starting point.

And Dunker's focus wasn't just on the final correct solution which is often overlooked, but on the process of getting there.

He used a methodology called protocol analysis.

Which means he just asked subjects to verbalize every single thought they had while they were attempting the solution.

And this created these incredibly rich transcripts of human intellectual struggle.

A subject might suggest, say, sending rays through the esophagus, hoping that internal tissue is less sensitive.

Or another might suggest desensitizing the healthy tissue with a chemical or physically exposing the tumor through surgery and then applying the rays.

The experimenter would guide the subject by gently repeating the core constraint or offering a productive prompt.

So for instance, when a subject suggested neutralizing the rays effect on the way, the experimenter might ask, that's a good track.

How specifically could one decrease the intensity of the rays on route?

And this questioning eventually guided the subject toward the standard accepted solution which is to send multiple weak beams of radiation from different angles.

Crucially, the beams are weak individually so they don't damage the surrounding tissue.

But they're focused through an imaginary lens or system in such a way that they all converge exactly at the tumor.

Right, where the cumulative intensity is high enough to destroy the disease tissue.

So Dunker took all these protocols, all the false starts and dead ends, and he organized them into a hierarchy,

a classificatory tree of solution proposals.

And this tree demonstrated that subjects didn't just randomly jump between ideas.

All their concrete suggestions, no matter how strange they seemed,

fell under three main strategic groups.

The first strategic group was proposals to avoid contact between the rays and healthy tissue, like inserting a protective cannula.

The second was proposals to desensitize the healthy tissue, for instance, by injecting some chemical counter agent.

And the third group of proposals was to lower the intensity of the rays on their way through the healthy tissue.

And that's the one that leads to the correct solution of concentrating weak rays.

The powerful insight Dunker drew from this was that successful problem solving doesn't just involve finding a solution, it involves a continuous reformulation of the problem.

You start with the general constraint, don't destroy healthy tissue, and each step forces you to define that constraint more productively.

How do I avoid contact?

How could I decrease the intensity?

He summarized this by saying, what is really done in any solution of problems consists in formulating the problem more productively.

The second classic situation is Mayer's two -string problem from 1931, which focused on overcoming mental rigidity in a physical setting.

This is a classic test of insight.

Picture a large room.

In this room, there are two cords hanging from the ceiling.

They reach the floor, but they're positioned far enough apart that if you hold the end of one, you cannot possibly reach the other.

And there are various objects scattered around the room, poles, pliers, maybe a table.

The task is simple,

tie the ends of the two strings together.

The obvious strategies, like trying to extend your reach with a pole or tying a secondary cord to one string to make it longer, they often fail because the distance is just slightly too great or the materials aren't suitable.

The correct solution demands an act of restructuring.

You have to use an object like the pliers, tie it to one string and swing that weighted string like a pendulum.

And while you're holding the other string, you catch the swinging pendulum and then you can tie the two strings together.

Mayer was particularly interested in what variables, both internal and external, helped or hindered people in achieving that sudden, insightful realization.

Right, so whether it's the high stakes radiation challenge or the simple swinging string,

psychologists identified several crucial factors that determine success or failure.

The first one is the effective instructions.

It seems obvious, but the way a problem is framed or the initial constraints you provide can really limit alternatives and provide necessary direction.

For example, a researcher named Safran studied subjects solving anagrams.

And those who were told to look specifically for animal names solved the anagrams far faster than those who were given no instruction at all.

But the effect goes beyond just clarity.

Aronson and Landy found that expectations matter just as much as clarity does.

They gave subjects a task that should take about five minutes, but they told one group they had 15 minutes to complete it and the other group they only had five minutes.

And intriguingly, the group given more time actually took longer to finish.

Which suggests that high expectations for task difficulty or even just generous time limits can actually decrease efficiency.

A second variable is the priming of solutions.

So familiarity enhances speed.

Studies by Dominovsky and Ekstrand show that if subjects learned a word list that happened to include the solutions to anagrams before the test, they solved them faster.

They were implicitly primed.

This was also tried on Mayer's problem.

Judson and his colleagues found that having subjects learn a word list containing the critical words rope, swing, pendulum did facilitate performance.

However, because it's an insight problem, the priming effect wasn't always perfectly reliable.

Priming seems to be most potent in highly structured problems like anagrams where the solution is one specific word.

Next up is the effects of hints.

And this is where Mayer made one of his most compelling observations about the subconscious.

If a subject struggled on the two string problem for 10 minutes, the experimenter would walk by and accidentally brush against the center string, causing it to swing just enough to suggest the pendulum movement.

And the result was staggering.

Of the 23 subjects who were failing,

19 of them solved the problem within seconds of the seemingly accidental hint.

And yet, this is the crazy part.

When they were questioned afterward, most of the subjects denied that the hint had affected them at all.

It's a fantastic illustration of covert thinking.

The physical action was too subtle to register as a conscious instruction, but it provided the necessary sensory input for that mental reorganization to occur, which led to the sudden conscious insight.

Burke and colleagues later found that hints were most helpful when they were provided early in the process, which suggests they helped the subject establish a useful initial framework or mental set.

And now we get to the flip side.

How past experience can actively harm performance.

The first major cognitive hurdle is functional fixedness.

This is the interference that happens when an object used in a specific way in a prior task has to be used differently in a new, similar task.

We struggle to see an object outside of its established function.

Dunker first highlighted this concept, but a 1951 study by Burke and Rabinowitz gave it a really strong empirical basis using a variation of the two -string problem.

In their version, they only offered two objects that could serve as the pendulum weight,

an electrical relay and an electrical switch.

And they divided subjects into three groups.

Group S had to use the electrical switch to complete an initial pretest circuit task.

Group R had to use the electrical relay in the pretest, and group C was the control with no pretest.

Then all the subjects were given the two -string problem and told they needed a weight.

So let's look at the data.

Of the nine subjects in group S who had just used the switch as a switch, only two of them used it as a pendulum weight.

The other seven chose the relay.

Conversely, of the 10 subjects in group R who had just used the relay as a relay, all 10 of them chose the switch as the pendulum weight.

That is just clear, undeniable evidence of rigidity.

The prior use of the object, using it functionally as a switch, made it functionally fixed.

It was difficult to restructure their perception of that object, even for something as simple as using it as a weight.

This led to the concept of functional value.

Suggesting a task is much easier if the subject can specify or has shown the appropriate functions of the tools beforehand.

The second major way past experience hurts us is mental set.

This describes rigidity in the solution process itself, rather than just the object utilization.

Luchens and Luchens brought this under beautiful experimental control with the famous water jar problems in 1950.

In these problems, you're presented with three jars, A, B, and C, of varying fixed capacities, and you have to obtain a specific amount of liquid.

But the crucial part was the sequencing.

Right, the early problems, problems two through six, were the Einsteinling or set -inducing problems.

They could only be solved by this laborious multi -step formula, jar B minus jar A minus two times jar C.

So it required focus and repetition to establish that pattern.

Then came the critical problems, seven through 11.

And these problems could still be solved using that complex, established B, A to C formula, but they also had much simpler, more direct solutions.

Like just filling jar A and removing jar C, or simply adding A and C together.

And what did they find?

A massive level of rigidity.

Most adults and nearly all the children continued to apply the laborious, complicated B, A to C method to the critical problems, even when the simpler solution was, you know, blindingly obvious and available.

They had developed a mental set, a fixed way of approaching this entire category of problem that prevented them from efficiently reevaluating the current situation.

It's like when you discover a complex keyboard shortcut for a task in some software, and you continue to use that 10 key combination, even when a new version introduces a simple one -click button.

You're just mechanized in your thinking.

And this phenomenon beautifully demonstrates that being highly experienced or highly practiced can paradoxically be the biggest barrier to innovative or efficient solutions, because the old method works well enough to suppress the search for a better one.

Okay, so with the findings on animal learning sets and human cognitive rigidity established,

psychologists needed models to explain how the process of problem solving actually unfolds.

Right, and three major theoretical approaches were developed to try and account for these phenomena.

The first,

and historically a very significant one, is the Gestalt theory approach, championed by the likes of Wertheimer, Kafka, and of course, Kohler himself.

Their main tenet, that the psychological whole is greater than the sum of its parts, meant they saw thinking as an indivisible process.

You couldn't just break it down into tiny stimulus -response bonds.

Their approach had some really clear distinctive features, all flowing directly from those chimp experiments.

They emphasized insightful solutions that required a sudden perceptual reorganization of the problem elements.

And they also characterized the solution process as the continuous development or reformulation of the problem, which links directly to Dunker's work we just talked about.

Crucially, they really de -emphasize past experience.

For Gestaltists, the solution was driven by the immediate objective direction that was determined by the current perceptual situation, not by accumulated habits.

And while Gestalt theory provided these powerful descriptive concepts like functional fixedness, which they were the first to identify, it ultimately proved to be a bit too loose as a scientific description.

How so?

Well, it couldn't provide rigorous testable hypotheses about the mechanism of the reorganization.

It could only really say that it occurred.

So research along those strict lines eventually just kind of lapsed.

Okay, so that brings us to the second, and perhaps the most generative approach, especially leading up to the mid 20th century, the associative theory approach.

This is the direct application of classical and operant conditioning principles to the complexities of human problem solving.

Associative models rely on explaining behavior through three main constructs.

First, the description of SR, that's stimulus response connections via reinforcement and extinction.

Second, they tried to tackle the unobservable internal processes by describing them as mediating responses.

Okay, what are mediating responses?

They're those internal covert links that occur between the stimulus and the overt response.

They are learned SR connections that happen inside the organism, allowing for thought without immediate action.

Got it, and the third.

And third, associative theory heavily relied on response hierarchies.

This concept explains that when you're faced with a problem, the organism generates a whole sequence of potential responses, but past experience has placed them in an order of probability.

So the most reinforced, most successful response sits at the top of the hierarchy,

and that's the response that's most likely to occur first.

Exactly, and it's fascinating how associative theory sought to explain color's insight without invoking reorganization.

A researcher named Davis in 1966 was pivotal here.

He distinguished between type O overt problem solving and type C covert problem solving.

Type O is pure old school trial and error.

The organism has no idea what will work, so it has to carry out responses overtly and check the outcomes.

Right, but type C is what happens after a subject has solved hundreds of similar problems.

Due to past experience and a well -developed response hierarchy, the organism can successfully predict the outcomes of various responses internally.

So in type C, the subject just needs to choose the appropriate response covertly in their head and then execute it perfectly.

And this covert choice appears as sudden brilliant insight, but it's actually the product of established highly efficient response hierarchies.

It's the conceptual equivalent of Harlow's learning set formation, just described using the language of SR bonds.

Okay, so finally we arrive at the newest contender, the information theory approach, which gained significant momentum in the 1960s with the rise of computing.

Right, the central idea championed by figures like Newell and Simon is that if you can program a computer to successfully simulate human problem solving, that successful computer analog provides a rigorous testable theory of human behavior.

And this approach really transforms the view of the human problem solver.

We're no longer seen simply as a black box of SR associations, we're seen as an active information processing unit.

And this approach focuses heavily on the mechanics of selection.

It concerns itself with selection processes, things like searching the problem space and scanning memory or the environment for relevant information.

It also emphasizes feedback or regulatory mechanisms, the continuous monitoring and modification of behavior based on the result of the search.

It's a very clean logical model, but it hits a computational snag when it tries to model actual human behavior.

What's the problem?

The main difficulty ironically is not accounting for our successes, but accounting for the discontinuities and failures in human problem solving, the rigidity, the functional fixedness, the sudden frustration, the seemingly illogical jumps.

So a computer program generally won't fall victim to a mental set if it's following a purely logical search algorithm.

Exactly.

The fact that humans do fail in these predictable illogical ways remains a challenge for the information processing model to fully simulate.

But I mean, it continues to be the most promising avenue for developing sophisticated, detailed and measurable theories of how we structure thought.

We've established that structured problem solving is often about finding the single, efficient, correct answer.

But what happens when there is no single correct answer or when the problem itself hasn't even been fully defined yet?

That moves us to the frontier of original or creative thinking.

Historically, this area has been investigated along two distinct lines.

First, understanding the process of creation.

So how do original ideas form?

And second, measuring the individual differences.

Who is creative and how do they differ from others?

To understand the process, we have to rely largely on introspective accounts.

First hand, reflective reports from eminent creators, whether they're mathematicians, musicians or poets.

And the most famous and detailed example of this comes from Henri Poincaré, describing his journey to discovering fusion functions in mathematics.

It provides an almost chronological narrative roadmap of the creative mind at work.

He started with intense, focused, conscious effort.

The stage that Wallace would later call preparation.

For 15 days, Poincaré strove unsuccessfully to prove that these functions couldn't exist.

He sat at his work table, he tested countless combinations and he failed.

So he gathered all the necessary knowledge, but the solution remained hidden.

Right, then the first moment of illumination occurred.

One evening after drinking black coffee, he couldn't sleep.

He described ideas rising in crowds, colliding with one another until stable pairs interlocked.

By morning, he hadn't formally proved it, but he knew the class of fusion functions existed.

That's the flash.

That's the flash.

He continued conscious work guided by analogy, but then his routine was interrupted.

He left Karen for a geological excursion, a shift in environment and attention leading to a period of apparent quiescence or what we call incubation.

And this led to the famous second elimination.

He arrived at Kutonsis and as he was stepping onto an omnibus, the idea suddenly materialized with perfect certainty that the transformations he used for fusion functions were identical with those of non -Euclidean geometry.

It required no conscious thought leading up to it.

It just came to him.

That's incredible.

Yeah.

He had the idea standing on an omnibus step.

It really emphasizes the power of that quiescence part of incubation.

When the conscious mind steps away, the subconscious machinery is often still grinding away, making those previously invisible connections.

A third illumination occurred later after another failed conscious attempt at arithmetical research.

Disgusted, he went on vacation to the seaside thinking of completely unrelated things.

Walking on a bluff one morning, the idea came with the same characteristics,

brevity, suddenness, and immediate certainty.

The connection between arithmetic transformations and non -Euclidean geometry.

Exactly.

And a final breakthrough happened while he was serving military duty, another period of forced incubation.

Walking down the street one day, the final solution to a difficulty that had stumped him suddenly appeared.

And following each of these illuminations, he then moved into the final phase.

That final phase, which he did back in his study, was verification.

All the subsequent testing of hypotheses, writing out the proofs, working out the intricate details, and formalizing the discovery.

This collection of anecdotal evidence, which is remarkably consistent across fields, whether it's poetry or physics, led to Wallace's four -stage model from 1926, which remains the fundamental descriptive framework for creativity.

And we've covered them.

Preparation, acquiring knowledge, incubation, stepping away, letting the subconscious work, illumination, the aha moment, and verification, the hard work of checking and finishing the idea.

While it's descriptive, not explanatory, it beautifully structures the experience of creation.

So beyond the process, researchers started asking, are certain people simply wired for creativity?

Early studies like Galton's Hereditary Genius and Anne Rowe's work on eminent scientists focused on high intelligence, success, and family lineage.

And the most famous longitudinal work was Turman's study, which he initiated in 1922, tracking thousands of high IQ children.

Turman found his gifted sample excelled in almost every respect, educational, social, emotional, compared to their peers.

But Turman himself noted a crucial point.

While his subjects were highly successful, none of them reached the stratosphere of true genius.

None of them became an Einstein or a Darwin.

And this observation fueled a pivotal critique by J .P.

Guilford in 1950.

Guilford argued that conventional intelligence and attainment tests were fundamentally flawed because they only measured one type of thinking,

convergent thinking.

Convergent thinking requires the subject to find the single correct expected answer to a quotient.

It's what most of school and standardized testing is built upon.

Guilford proposed that this approach completely overlooked a whole range of imaginative original abilities he called divergent thinking.

Divergent thinking is the ability to generate a wide array of novel solutions or ideas.

And Guilford argued these abilities are continuously distributed in the population, not just limited to a few geniuses.

So to measure this, he developed new tests.

For example, the unusual uses test.

You're asked to list as many uses as possible for an ordinary object, like a brick.

And the score is determined not just by the sheer number of uses, which is fluency, but by how unusual, clever, or far -fetched they are.

That's originality.

Another one was the plot titles test, where subjects generate many different titles for a story plot, and that's scored on quantity and cleverness.

These tests shifted the focus from finding the answer to finding many possible answers.

And this led to the famous research comparing high IQ students with high creativity students, the most well -known being the Getzels and Jackson study of 1962.

They selected two groups of adolescents, one with very high IQ, but low scores on Guilford's creativity tests, and the other with high creativity scores, but significantly lower IQs.

And the findings were genuinely surprising.

Despite having statistically lower IQ scores, the high creativity group showed no difference in terms of school achievement.

So they were just as successful academically as the high IQ group.

Exactly.

This suggests that the measured elements of creativity can compensate for, or perhaps even override,

lower conventional intelligence when it comes to academic output.

But the differences went deeper,

touching on personality and social dynamics.

The high IQ group was consistently preferred by teachers, which is, I guess, not surprising, given that their values aligned with school success.

More tellingly, the high IQ group highly rated qualities that they believed were important for teacher liking and adult success, indicating a tendency toward conformity and external validation.

The high creativity group, however, rated those same qualities low, showing a strong tendency toward independent thought and unconventionality.

And they also demonstrated this independence in their aspirations.

The high creativity group indicated a wider and more unconventional range of career goals.

So while the debate continues, is creativity independent, or simply another facet of intelligence that conventional tests missed?

Guilford's work dramatically broadened the psychological view of thinking, asserting that our mental life is just as much about generating possibilities as it is about finding the one correct pathway.

So to fully understand thinking, particularly that internal representation of events, Osgood's broader definition, we have to address the fundamental mechanism that allows us to cope with reality, and that's categorization.

I mean, think about it.

We are bombarded by individual,

unique stimuli objects, events, people, every single second.

If we had to treat every single rose, every cloud, every pencil, as a completely new, unique event, our minds would be instantly overwhelmed.

In order to cope, we must classify and categorize this vast array into manageable sets or concepts.

And concepts are, by definition, hierarchically arranged.

The concept of a plant is higher order than flower, which is higher order than rose.

At the core of any concept are its attributes.

An attribute is any discriminable feature of an event that can vary shape, size, color, weight.

Right.

And if you think about George Kelly's theory, his term construct is essentially a synonym for an attribute in his personal psychology framework.

So psychologists investigate concepts in two key areas,

concept attainment, which is how adults learn new concepts, and concept formation, which is how basic concepts develop in children.

Let's start with adults.

The classic work here is Bruner, Goodnow, and Austin's, a study of thinking from 1956.

They began by establishing the three main logical types of concepts we use to organize our world.

First up, conjunctive concepts.

These are defined by the joint presence of several attributes.

For a concept to be a red pencil, it has to be both red, A, A, possess all the defining attributes of a pencil.

Both conditions must be met.

Second, disjunctive concepts.

These are defined by the presence of one or more of a number of attributes.

For instance, the traditional diagnostic concept of schizophrenia was historically disjunctive.

The criteria were met if any two or three defining attributes were present, even if no single attribute was obligatory.

So it's a messy or situation?

Exactly.

And third are relational concepts.

These aren't defined by presence or absence, but by a specifiable relationship between attributes.

Income tax levels are a classic example.

Your tax bracket is determined by the relationship between variables like your income level and the number of dependents you claim.

Bruner and his colleagues designed experiments to study how adults actually learn these categories.

They used a card array of 81 cards, with each card representing a unique combination of four attributes, color, figure type, number of figures, and border type, each with three values.

And they studied attainment under two key paradigms.

In the selection paradigm, the subject actively chooses which cards to view, testing their own hypotheses, much like a scientist designing an experiment.

In the reception paradigm, the experimenter picks the cards, and the subject is the passive recipient of the information, only told whether that card is positive or negative for the concept rule.

Which is analogous to a human psychologist studying naturally occurring mental operations, receiving information passively.

So focusing on that reception paradigm for conjunctive concepts,

Bruner identified two main logical strategies that adults use to solve the problem.

The first, and generally superior one, is focusing, sometimes called the whole list strategy.

The focuser uses the very first positive instance they see, for example, three black crosses with a troubled border, as their initial overarching hypothesis.

They then systematically vary one attribute at a time, and based on the feedback, rule out irrelevant attributes.

And focusing is logically superior, because it requires far less cognitive strain.

The focuser only needs to remember which attributes have already been definitively eliminated.

The second strategy is scanning, or the part strategy.

Here, the subject starts with part of the first instance as the hypothesis, say, just black crosses,

and tests it.

The moment they encounter a positive card that invalidates that partial hypothesis, they have to change their hypothesis entirely.

So the scanner has to keep track of every possible combination.

They're trying to juggle too many attributes at the same time.

Right, and Bruner's findings confirmed that while subjects were highly consistent in the strategy they chose, more of them used focusing.

And crucially, focusers were significantly more successful than scanners.

Especially when the task became difficult, or when time pressure was introduced.

Failures usually happen when subjects abandon the strict logical rules of their chosen strategy.

Okay, so if that's how adults attain new concepts, we need to look at how children form the most basic concepts.

And for this, we have to turn to Jean Piaget.

His intensive, longitudinal studies on how children adapt to the environment define the stages of conceptual development we use to this day.

Piaget's theory relies on two basic functional concepts for adaptation.

Assimilation involves the active manipulation of objects.

The infant explores, glasps, and sucks, integrating the object into their existing mental schemes.

And accommodation is the passive adjustment to external stimuli.

The object resists, or hurts, or rewards the infant, forcing the mental scheme to adapt to the reality of the object.

These interactions build initial schemas, which later develop into operations internalized, reversible actions that allow for planning and logical thought.

Piaget outlined four stages of development detailing the gradual construction of concepts.

Stage one is the sensorimotor period, from birth to about two years.

This period is focused on sensation and motor action.

The paramount conceptual achievement here is object permanence.

So for a five -month -old, if an object is hidden, it literally ceases to exist for them.

But by eight months, the infant will actively search for the hidden object, demonstrating that the object is stable in their mind, even when it's out of sight.

Stage two is the preoperational thinking stage from two to seven years.

Symbolic representation language, imagination develops rapidly.

The central challenge the child faces here is conservation.

Which is the understanding that the quantity, so mass, weight, or volume of a substance remains invariant, despite changes in its appearance or transformation.

Piaget's conservation experiments are legendary, and they beautifully illustrate the cognitive limitations of this stage.

Take the conservation of mass.

A child agrees that two balls of plasticine are identical.

One is then rolled into a sausage shape right in front of them.

A typical four -year -old will instantly say the sausage has more clay.

Because they're just focusing on the length and ignoring the corresponding decrease in diameter?

Exactly.

Conservation of mass is achieved around age seven.

Conservation of weight is tested similarly, using scales, and it's independent of and more difficult than mass conservation.

Children achieve weight conservation around age nine.

And the conservation of volume is the hardest.

Water from a short, wide beaker is poured into a tall, thin beaker.

The young child focuses solely on the height dimension and confidently states the tall, thin beaker has more water.

Or if you pour the water into six small beakers, the child says there is more because there are simply more peakers.

The child fails to conserve volume, which is typically achieved around age 11.

They're locked into concentrating on a single salient perceptual dimension height or number to the exclusion of others, like beaker diameter.

Bruner and his colleagues identified two main reasons for this failure.

First, that tendency to concentrate on a singular perceptual dimension.

And second, a failure to match language with experience.

The verbal label same does not carry the same structural meaning for the young child as it does for the adult.

Stage three is the concrete operation stage from seven to 11 years.

This is where conservation becomes a stable, logical structure.

The child develops other internalized operations,

including classification understanding that one class can be included within a broader class and seriation, the ability to logically order stimuli based on continuous variation, like shortest to tallest.

And stage four is the formal operation stage from 11 to 15 years.

The final stage where the child moves beyond reliance on concrete objects and can apply operations in an abstract way, enabling them to form and test hypotheses mentally.

And this is the stage required for pure verbal logic.

The classic example is solving a hypothetical verbal problem like, Edith is fairer than Susan, Edith is darker than Lily, who is the darkest of the three.

This capacity to manipulate abstract rules detached from immediate perception is the culmination of concept formation.

So if concepts are the essential building blocks of our thinking, how we categorize and anticipate the world, the next logical step is to ask,

can we actually quantify and measure an individual's personal system of concepts?

Psychologists develop techniques that are highly flexible and centered on the individual's unique system of interpretation.

We'll look at two seminal examples.

The first is the Osgood semantic differential developed by C .E.

Osgood.

This technique actually has fascinating origins in the study of synesthesia.

That's the cross -sensory phenomenon where, say, hearing music elicits visual sensations like colors or shapes.

Osgood's team realized that many sensations and concepts could be related using symbol adjectives.

His technique asks subjects to rate the connotative meaning of various concepts, like father, sin, or communism, on a series of seven -point bipolar adjectival scales.

So these scales are anchored by opposites, things like rough, smooth, good, bad, or wise, foolish.

And subjects were found to rate concepts easily on these scales, even for seemingly unrelated pairings, like rating the concept lake on a scale anchored by yellow -blue.

From this, Osgood formalized the theory of meaning.

He proposed that meaning is, first, a set of mediating responses that have become autonomous from the initial SR pairings.

Second, and crucially, he argued that meaning is multidimensional.

Meaning is not one thing.

It is a complex structure.

To isolate these stable structures, Osgood used factor analysis, a statistical method that groups scales that intercorrelate highly with each other.

And this analysis revealed that across hundreds of concepts in diverse populations, meaning consistently collapsed into three stable dimensions.

The largest and most robust factor is invariably evaluation, with scales like good, bad, or wise, foolish.

This seems fundamental to how we judge any stimulus.

Second is potency, with scales like potent, impotent, or strong, weak, reflecting the perceived power of the concept.

And third is activity, active, passive,

fast, slow, reflecting the perceived dynamism of the concept.

These three dimensions, evaluation, potency, and activity, often called EPA, form the core structure of connotative meaning.

The format typically presents a concept at the top of a page, followed by a series of bipolar scales beneath it.

But the key takeaway is that the semantic differential is a flexible technique, not a standardized test.

Both the concepts being rated and the scales used must be carefully selected by the investigator to suit the specific study.

Right, you wouldn't use sweet sour to rate architecture, for example, but beautiful, ugly, and strong weak would be perfect.

Exactly, and that flexibility is shared by the second major measurement technique, the Kelly Repertory Grid Technique, developed by George Kelly in 1955.

This is deeply rooted in his personal construct theory.

Kelly viewed every individual as an active scientist, focused on anticipating future events.

And to do this, we all develop our own highly personal and idiosyncratic system of bipolar constructs, the unique ways we interpret and categorize our world.

And the Repertory Grid is designed specifically to sample and quantify that individual construct system.

The central method for eliciting these personal concepts is the triadic method.

The subject is presented with three elements, which might be people in their life, jobs, or pieces of art, and asked to state one important way in which two are alike and different from the third.

This process is repeated until the investigator has collected an adequate sample of meaningful constructs unique to that individual, for example, friendly -unfriendly or supportive -critical.

Early quantification involved simple matching scores.

So if a subject consistently rated people they called friendly, also as kind, the two constructs were highly correlated in their personal system.

Modern approaches typically use a ranking method where the individual ranks all the elements, the people, jobs, or art from, say, most friendly to least friendly on various constructs.

This allows for powerful statistical analysis to specify the precise relationships between their concepts, showing which of their concepts are closely linked and which are independent.

So the crucial difference between the semantic differential and the Repertory Grid lies in their starting point.

Right.

Osgood maps the general stable dimensions that EPA of connotative meaning shared across people.

Kelly, however, maps the unique idiosyncratic system of bipolar personal constructs that one specific individual uses to anticipate and navigate the world.

Both provide an invaluable look into the organized structure of individual thought.

Wow.

We've covered a remarkable range of content today, tracing the process of thinking from, you know, the mechanical trial and error of a cat in a box all the way up to the mathematical brilliance of Poincaré's sudden insights.

We saw the historical debate resolved by Harlow's concept of learning sets, which demonstrated that insight isn't some magical leap, but rather the internal accumulation of highly efficient habits that allow the subject to learn how to learn.

We then explored human cognitive hurdles, focusing on Dunker's emphasis on productive reformulation and the two great inhibitors of thought, functional fixedness, that's rigidity in object use, and mental set, which is rigidity in the solution process itself.

We outlined the three major theoretical attempts to model this process, the holistic reorganization of Gestalt, the response hierarchies of associative theory, and the selection processes of information theory.

And finally, we moved to the broader internal processes, charting creativity through Wallace's stages, contrasting conventional convergent thinking with Guilford's crucial divergent thinking, and detailing how we categorize the world through concept attainment strategies and Piaget's stages of concept formation.

Before closing with the personalized measurement tools of the semantic differential and the repertory grid.

So what does this comprehensive overview mean for you, the learner?

We've established that the capacity for efficient, insightful thinking is not some inherited genius, it is a learned capacity built through experience, which results in the formation of powerful mental frameworks.

And we know that the greatest barriers to insight are our own cognitive ruts, that functional fixedness and the rigid mental set we develop when a method has proven successful, even if it's cumbersome.

So if creativity or divergent thinking is the ability to break free from the expected answer, and if a rigid mental set is the surest way to fail a novel problem, here is a final provocative thought to mull over.

The research suggests that the deliberate practice of tackling unfamiliar unstructured problems is the only way to build new learning sets and cultivate the capacity for that sudden, point care -like illumination.

So which everyday tasks in your life allow you to lean comfortably on an established formula, that BA2C solution, and which ones actively force you into a state of cognitive struggle, forcing you to develop truly divergent thinking?

The more you choose the latter, the more sophisticated your thinking system becomes.