Separating Recyclables

You run a recycling center. Your trucks bring in the recyclable items all mixed together: plastics, cans, and glass jars.

How can you easily separate them by some automated process?

At the end of the process, the plastics should be together, the cans should be together, and the glass jars should be together.

Please try this puzzle on your own before googling current solutions. You might come up with something better.

Drying Tennis Shoes?

I got caught in a bizarre downpour yesterday and my tennis shoes got soaked.

How can I dry them in a dryer without them banging around?

Drying them with a full load of towels (or something) helps, but there are more creative ways. How can you do it?

Reinventing the Shopping Cart

Because the shopping cart is so familiar to us, when we try to reinvent it there are so many ways that we can fall into mental ruts and functional fixedness (i.e., the tendency to consider an object only through the lens of its designed use--and so be oblivious to other possibilities).

Watch this group from IDEO counteract these negative forces that inhibit creativity.

(part 1: 7 minutes)
http://www.youtube.com/watch?v=oUazVjvsMHs&feature=related

(part 2: 7 minutes)
http://www.youtube.com/watch?v=_civr9fr4iw&feature=related

(part 3: 7 minutes)
http://www.youtube.com/watch?v=zddv5Bv7da8&feature=related

Aha! Productions Mission Statement

Here is a draft of a mission statement for a new company that I'm designing called Aha! Productions.

The human mind is the source of all innovation; and unfortunately, often its worst enemy. Those who understand how to get the human mind “unstuck” when it is solving problems will have a distinct advantage in any race to innovate. Aha! Productions first uses behavioral and neuroimaging experiments to test the effectiveness of our innovation-enhancing techniques; then communicates our scientifically demonstrated techniques and insights in media products such as television shows, board games, video games, books, and educational materials. Our first three products are the following: Aha! Guess the Invention (TV show), What Box!? Thinking Outside the Box by Dissolving It (a book for a general audience), and Mad Scientist (a board game for ages 8 and up).

Building Our Theories

Choose one of the following tasks to do for next week, or come up with your own related task.

1) Look over all the diagrams that we created and create a master diagram that sums up the commonalities in them all.

2) After you look over all the diagrams that we created maybe you will notice that they cannot be reduced to one master diagram. If this is the case, reduce them to two (or three) diagrams. Begin to think about how the two (or three) diagrams you came up with differ, what different predictions they make, and how we might test them.

3) Look at the eight theories from the literature that I presented. Which of the theories do you think is the most promising to build upon? Is there some way to combine two or more of the theories into a grander theory?

4) Compare the eight theories from the literature with the diagrams that we created. Are there important things mentioned in the eight theories that we left out of our diagrams. If so, create a new diagram that incorporates it.

5) Come up with your own task for next week. That is, come up with your own next step for building a theory of insight problem solving.

Status of Our Lab & the Big Picture

As you can tell from our meetings, there is a great deal of room in the area of insight problem solving for new work (new theories and new experiments).

I am very impressed with all the work that our PuzzleLab group has been doing. You have really given yourselves to the puzzles and all the tasks that I have asked you to do. Thanks for giving yourself to the topic and making lab so enjoyable.

As you also know, I am very interested in coming up with practical results that can improve people's problem-solving abilities. First, I believe it is a scientist's responsibility to society to address research questions that can potentially benefit society. Second, I believe that given the times we live in improving people's problem-solving abilities is a very important task. In the coming decades there is going to be so much innovation required as we search for more and more "green" products, more energy-efficient devices, and new sources of energy (to name a few areas). Any psychology results that can improve problem-solving skills and enhance innovation could play an important role in this transition to a "green" world. Further, as my TV show design highlights, many clever inventions are needed to improve the basic quality of life for so many of the world's poor. Again, psychology results could assist the process of invention for this noble and important cause.

Based on the scholarship I received this year, the UMass engineering department agrees with my basic vision and sees potential in my research for results that could directly help engineers become more innovative.

Recently, other areas of application have come to my attention.

1) Shiraj Sen is a doctoral student in computer science working in the robotics lab. He is trying to make a robot that can learn about the objects in its environment and then solve problems using those objects. Although the problems are simpler than the ones we humans deal with, still they are a type of insight problem in that the representations of the objects are incomplete and the robot needs to notice new features about the objects. Shiraj and I have been meeting and he is going to be using an idea of mine in his dissertation work. I have not shared this idea with you, but I will sometime in the coming weeks.

2) There is a fairly new area of research called CreativeIT (Creative Information Technology). The premise of the area is that computers can be used to help us humans overcome some of our limitations when it comes to being creative. My research definitely fits into this new area. Here is a quote from the National Science Foundation website.

"Information technology is playing an increasing role in extending the capability of human creative thinking and problem solving. The study of creativity and computing as a way to advance computer science and information technology, cognitive science, engineering, education, or science can lead to new models of creativity and creative computational processes, innovative approaches to education that encourage creativity, innovative modes of research that include creative professionals, and new technology to support human creativity."

These are some of the reasons I think the study of insight problems is important and exciting. I am glad that you are a part of it.

Clever Crows

Cracking nuts by dropping them in traffic over a pedestrian walkway. The cars run over the nuts to crack them. The stoplight stops the traffic and the crows safely collect the nut fragments.

http://www.youtube.com/watch?v=2PmkreJHw4k

A crow bends a straight wire so it can hook food that is out of reach.

http://www.youtube.com/watch?v=AeQppJmOWJA&NR=1

Chimp Solution

Here is a solution that this chimp took about 10 minutes to create. The clip is from "Ape Genius," a NOVA special for PBS.

http://www.youtube.com/watch?v=yrPb41hzYdw

Hints for Jail Cell Problem

First Hint
Try a visual juxtaposition diagram. The peanut is at the bottom and you want it near the top so you can reach it.

What can effect this transformation? Imagine the peanut moving from the start state to the end state? What can help the peanut do this?



Second Hint
Try a visual superimposition diagram.

What can effect this transformation? Again, imagine the peanut moving from bottom to top? What can help the peanut do this?

The Jail Cell Problem

You are locked in a jail cell!

If you solve the puzzle, the warden will let you go free!

Here is your jail cell!



Affixed to the inside of one of your bars is a tall (2 foot), slender (two of your fingers can fit in), transparent tube (illustration below). A peanut is dropped to the bottom of the tube. You must get it out but you cannot remove the tube from the bar or break it. You have to repeat your solution in front of the warden to make sure that you did not cheat.

All you have in your cell are the usual items: cot, pillow, desk, sink, and toilet. How can you win your freedom?

Hints are at the next post: "Hints for Jail Cell Problem"

My 'samples hurt' answer

Possibly, subjects fixate on the features of the sample that satisfy the goal requirements rather than focusing on the goal requirements themselves. A further factor may be the visual nature of the sample when compared to the abstract nature of the goal which is commonly presented as a verbal description. The visual nature of the sample may draw attention to itself in a way that the verbal goal description cannot overcome. If the goal could be made as visual as the sample, then perhaps the visual nature of the goal can successfully compete with the visual nature of the sample.

I hypothesize that subjects who have seen the Cant Hook can overcome their fixation on it by focusing on the visual juxtaposition diagram of the rolling log. What do you think? Does this sound like a reasonable hypothesis?

In lab, we will discuss a simple experiment to test this hypothesis.

Why do samples hurt?

Why might sample solutions inhibit innovation? Write out your own answer to this question. When you are done you can look at my answer in the next blog. But remember, I don't really know. I haven't tested anything yet. I am just theorizing.

A Log-Rolling Solution

Here is a device for one person to roll a log. It is called a Cant Hook.



And here is a picture of the Cant Hook in position for action.



The spike rests against one side of the log. The hook digs into the other side. The operator uses the handle to gain leverage. In this way, a large log can be slowly rotated in a ratcheting fashion.

Sample Solutions Inhibit Innovation

It has been consistently shown that the presence of a sample solution inhibits innovation. A series of studies suggests that presenting engineers with a problem to solve along with a picture of an object that solves the problem tends to produce new solutions that share major characteristics with the given sample, even when engineers are told to be innovative; while presenting a problem without showing a sample solution leads to more innovative solutions (Jansson & Smith, 1991; Purcell, Williams, Gero, & Colbron, 1993; Purcell & Gero, 1996).

Try this problem. I won't give you a sample solution that will tend to inhibit you. Suppose you want to be able to roll a fallen tree that has been stripped of its branches—so rolling is feasible. Only human power can be used to roll the log (no motors) and two people can work together to achieve the rolling.

In essence, you want to move the log from the start state to the end state. How can you do it?

Try to come up with your own solution without going to the next blog. The next blog has a sample solution and that will inhibit your creativity.

Visualizing the Ten-Coin Problem

In the previous post, the ten-coin problem was visually presented as the start state next to the end state. I will call that image a visual juxtaposition diagram.

Another way to present the two states is to superimpose them on top of each other. To present the visual superimposition diagram, I will make the end state transparent and place it on top of the start state, as shown below.



The superimposed representation provides the clue that seven coins of the two formations overlap and that the end state can be reached by moving the three unaligned blue coins into the position of the three unaligned transparent coins.

I have found no mention in the problem solving literature of the visual superimposition method and how the visual superimposition diagram can improve performance for certain types of spatial problems. In fact, insight problem studies continue to use the ten-coin problem with their subjects.

Does the visual superimposition method only work for the ten-coin problem? Give a description of the type of problem that the visual superimposition method would help with.

Ten-Coin Problem

In lab, we looked at verbal processes that might help or hinder performance on insight problems (Glucksberg & Danks, 1968). Presently, we will consider visual processes that might help performance.

Try the classic ten-coin problem below. Only after you have tried it on your own for a while should you go to the next blog for a hint on how to visualize the problem in a helpful way: "Visualizing the Ten Coin Problem."

Transform the start state of the 10 coins into the end state by moving only three coins.

Article: Glucksberg & Danks

Here is some relevant background that you will need to understand the paper.

A) Functional fixedness means that you get stuck (i.e., fixated) on an object's designed use that inhibits you from considering other uses for the object. For example, when looking at a screwdriver, knowing that it is designed to turn screws interferes with your ability to notice features that could lead to other uses.

B) Duncker's Candle Problem (1945)
Subjects need to attach a candle to a wall so it will burn without dripping wax on the floor using only a candle, a book of matches, and a box of tacks. Subjects tend to notice the "box of tacks" but not the "box" that contains the tacks. The traditional solution is to use the box as a container (or shelf) for the candle and then attach the box holding the candle onto the wall with the tacks (also the box catches the wax drips).

These paper includes two experiments. For each experiment, be prepared to discuss the answers to the following questions.

1) What is the experiment designed to test? What is the authors' hypothesis?

2) What is the design (e.g., the groups that are compared against each other)?

3) Is the design a good design? Does it test what it is supposed to test? Are there any flaws in the design? If there is a flaw, how would you fix it?

4) What are the results? Do the results support the hypothesis?

5) Overall, does the paper give you any ideas for follow-up experiments?

Noticing Omnipresent Features

If a feature of an object is always present how does one notice it? How does one notice that the wax of a candle is in one piece or that the candle does not change locations when it burns? As Whitehead (1929) observed, empirically we notice by difference in that if something is sometimes present and sometimes absent then we can notice it. If something never takes a vacation then we cannot notice it in this manner. An act of imagination is required to reveal it for the first time. But how does the imagination unconceal what is hidden in its omnipresence? In short, we must create a way to notice it. I will suggest ideas that might help with the unconcealing.

Of course, these ideas will not capture everything. The quest to notice the omnipresent features is most likely an endless quest. These questions, however, get us off to a good start.

See if any of the ideas below help you notice more features of a candle. If you can notice a new feature, then you can vary that feature to make new innovative designs!

Number
For every noun you notice (e.g., wick, wax), consider any number of these items (including zero). For example, if you notice one wick, consider zero wicks, two wicks, etc.

Negation
Noticing the presence of something (e.g., has a wick, is tall) can lead to considering its negation (e.g., does not have a wick, is not tall).

Two Parts Fixed Relative to Each Other
A pencil lead of a typical wooden pencil is fixed (i.e., does not move) relative to the wooden shaft. Making the two parts move with respect to each other leads to mechanical pencils. Now, the pencil lead moves with respect to the shaft of the pencil. In constrast, if two things move (i.e., vary) with respect to each other, maybe they can become fixed.

Contiguous versus Separated
If two parts of an object are next to each other or connected, maybe they can be separated--or vice versa. For example, separating the power button on your TV from the TV itself gives the idea of a remote control. Now, the power button is separate (and physically distant) from the TV.

1D to 2D to 3D
If multiple parts are in the same dimension (line or plane or volume) with respect to each other, maybe they can move to a different dimension with respect to each other. (Need an example.) (If you have an example, please let me know.)

Spatial Boundaries
Violating the natural spatial boundaries of an artifact may alter its design in interesting ways. (Need an example.) (If you have an example, please let me know.)

Rotation
Rotating an object from its normal operating position may have interesting consequences. (Need an example.) (If you have an example, please let me know.)

Inversion
Considering inner parts moved outward or outer parts moved inward may have interesting consequences. (Need an example.) (If you have an example, please let me know.)

Motionless to Motion
If part of an object is motionless (or even the whole object), perhaps it could be put into motion—or vice versa. Considering putting a burning candle into motion led to the designs described above to spin or raise the candle as it burns.

Features and Innovations

For an innovator, noticing a novel feature (or relation) is a pathway to innovation. A feature is a site of potential differences. For example, if you notice a candle's color, then vary the color for new types of candles (not a very creative idea, but I think you get the point). From every feature you notice comes new innovations as you vary that feature. It follows then that the more features you notice, the more innovative you are! By this measure, the person who notices the most features of an object is really the most innovative.

Standard Features
Consider color, size, shape, and material. Vary these and you have various types of candles.

Internal Relations
Consider the relations between the two parts of the candle: wax and wick.

The wax keeps the wick straight.
Vary this and you have candles where the wick spirals, slants, zig-zags (or other patterns) through the wax.

The wax contains one wick.
Vary this and you now have multi-wick candles.

Combine: non-straight wicks and multi-wicks
Several wicks wind their way through the wax. Or, several wicks slant and converge into one wick halfway through the wax.

The wax of the candle consists of one piece.
Keep the wick as one piece and break up the wax into multiple parts.



To compensate for this change, a structure is built to hold the discrete wax sections in mid-air.














Causal Relations
A careful analysis of a causes and effects of a candle in action can lead to various innovations. Each cause has one or more effects. For each cause-effect pair, we can ask, “Does it have to be this way? Could it be different?”

The wax pools.
The heat melts the wax which then pools at the base of the wick before it starts to climb up the wick. What if we changed the candle’s design so that only some of the wax pooled? The rest drained away before cooling. Altering the amount of pooled wax would change how fast the candle burns. I do not know if this is a useful design, but it is one that is possible.


The candle is motionless when it burns.

Putting one side effect to work.
How can we get the candle to move from its own dynamics? The heat that rises from the flame is often considered a side effect because we are primarily interested in the candle’s light. However, the rising heat can be made “useful” and become the cause of further effects. The rising heat can turn a small windmill-like blade above the flame. The turning blade is connected to a gear system that slowly turns a stand that the candle sits upon.

Putting another side effect to work.
Another side effect of a burning candle is that is loses weight as it burns. Place a candle on a stand that is counterweighted to keep it in the same vertical position. Because a candle slowly loses weight as it burns, by placing it on this stand it will slowly elevate itself as it burns. Maybe it can elevate itself upward toward a candle snuffer and put itself out when it reaches the top.

Go to the next posting for more ideas to help you create further innovative candle designs.

New Candle Designs!

Your job this week is to design some new types of candles. Be as innovative as you can be.

That is all the instruction I will give you right now. I don't want to inhibit your natural creativity.

After you have created several new designs and sketched them out, you can go to the next post to some things that might help you (or maybe you were already doing them naturally).

Which Came First?

Which came first the feature or the relation?

If you guessed 'relation' you are correct.

A feature is a shorthand for a relation (or set of relations).

Because features are reducible to relations, relations are more fundamental than features.

But we usually learn an object's features and forget about the underlying relations of each feature.

Where Do NEW Features Come From?

Q: Where do new features come from?
A: They come from new interactions.

If a feature comes from how an object interacts with another thing, then we can never know every feature about an object because we have not seen it interact with every other possible thing in the world (Nehamas, 1985).

It follows then that to find a new feature of an object, have the object interact with something it has never (or rarely) interacted with before.

Science is replete with stories of new features of objects emerging after getting them to interact with things they have never interacted with before. Superconductivity in ceramics, for example, emerged from interacting a specific ceramic material with electricity and a near-absolute-zero temperature.

But the same principle works for common objects, such as a basketball, as we have seen.

Which came first the feature or the relation?

Go to the post: "Which Came First?"

Where Do Features Come From?

Q: Where do features come from?
A: They come from interactions.

"The features of a thing are effects on other 'things':
if one removes other 'things,' then a thing has no features..."
-Nietzsche (Will to Power, #557)

Every feature of an object is the result of some interaction with something else (Nehamas, 1985). If a new object never interacted with anything then we could know none of its features (e.g., color, heaviness, hardness, texture). A set of relations describes the interaction.

Consider a stone.

Being gray is a feature of the stone.
Being heavy is a feature of the stone.

A feature however is a shorthand for an interaction, which is a set of relations.

Being gray describes relations (i.e., an interaction) between light waves bouncing off the stone and into my eye that are interpreted by my brain. So, "gray" is shorthand for relations among light waves, the stone, my eye, and my brain.

Being heavy describes relations between the mass of the stone, the strength of gravity, and the strength of the muscles in my hand and arm. So, "heavy" is shorthand for relations among the stone's mass, gravity, and my hand/arm.

Go to the post: "Where Do NEW Features Come From?"

Try This Process

You don't have to actually try this out. You can just read along.

I made a game for kids called "Mad Scientist."

You have two decks of cards: a deck containing common objects (hammer, shoe, basketball, etc.) and a deck of uses (transporting water, slicing bread, fastening paper together, painting, jacking up a car, etc.).

Shuffle each deck separately and draw the top card from each deck. Suppose you get "basketball" and "transporting water." How can you use a basketball to transport water? Everyone, except the judge for this round, creates a solution. If the judge picks your solution as the best, you win a point for that round. Switch judges and draw two more cards to start a new round.


Well, you could cut open the basketball and carry water inside. Did you have "transport water" on your list of uses for a basketball? A basketball can carry water because it is hollow inside. Did you have "hollow" as a feature of a basketball?

You could also deflate the basketball into a bowl-shape and then carry water. Did you have "can become bowl-shaped" and "can be deflated" as features?

The point is that a random process like shuffling a deck of uses can help you create new uses and features that you would not normally notice.

Why might this random process help find uses? Without it, you are stuck following your own association patterns which will tend to stay close to those things closely associated with basketball. A random process can help you get out of the neighborhood close to a basketball to other more distant neighborhoods.

Why might this random process help find features? The short answer is: Because a feature is the result of an interaction with something else. Interact the basketball with something it has never interacted with before and most likely you will notice a new feature. For a longer answer, go to the next post "Where Do Features Come From?"

Unusual Uses for a Basketball

Make a list of all the things you can use a basketball for.

As you make your list of uses, also make a list of features of the basketball (e.g., round).

When you are done making your lists, go to the next blog entry "Try this process."

Two Books on a Bookshelf

Whenever you compare two or more objects, there is so much more there than meets the eye.

Imagine a bookshelf with just two books on it.

1) Make a list of all the ways that the two books could be similar (e.g., "both are written by authors from the same country," "both covers are a similar color," etc.). How long can you make your list? Can you make your list become infinitely long?

2) One book is to the left of the other on the shelf. Make a list of all the reasons that the two books are in the order that they are in (e.g., "they are in alphabetical order by the author's last name," "they are in order by the number of pages in the books," etc.). How long can you make your list? Can you make your list become infinitely long?

Solution to Two Rings Problem

Because of the heaviness of the rings, melting the wax to form a bond around the rings will not be secure enough.

The candle's wick is a string. Free the wick by scraping away the wax on the edges of the cube of steel. Tie the rings together with the string.

Because the wick is not normally seen as a string, this problem is considered an insight problem that requires seeing more than is normally noticed.

Why are puzzles puzzling?

Why are puzzles puzzling?

Here is my basic answer. It will become more refined as the semester goes on.

From a psychological standpoint, our set of normal associations about the elements of the puzzle omit the key association(s) needed to solve the puzzle. But the key associations are there, for those who have the eyes to see. (See the "Two Rings Problem" below for an example.)

From a theoretical standpoint, there is potentially an infinite number of relations among the puzzle's elements, so it can be difficult at times to narrow down to the key ones. (See the post "Two Books on a Bookshelf" for an example.)

Psychological Example: "Two Rings Problem" (created by Andrew Cohen)
Suppose you had to fasten together two steel rings (each eight inches in diameter and each weighing eight pounds) by using only a long candle, a strike-anywhere match, and a cube of steel (two inches on each side). The rings need to be securely fastened so that when one ring is picked up the other ring will follow.

When you want to know the answer, go to the next post "Solution to Two Rings Problem."