Teaching Little Kids to Cube

cubetimer

Regular
Joined
Aug 31, 2007
Messages
58
Location
San Diego, CA
WCA ID
2007DAIL01
YouTube
Visit Channel
In another thread I mentioned that I've been teaching my 5 year old son how to cube and someone asked for tips on how I'm doing so. I thought it would be better off as a new post instead of hijacking an existing thread with a rather length explanation...

I've been breaking it down into really small, easy to manage steps. Basic outline of what I'm teaching him:

- A few basic terms. What is a corner, edge, center, and face.
- I'm staying away from words like clockwise and counter clockwise. I describe all the moves in terms of up, down, left, and right. For example something like R U as I'm looking at the front would be 'up', 'left'.
- I'm teaching algs that are as short as possible, and I'm not teaching mirrors of those algs. I always have him do it on the same side, and multiple times if needed. We'll get to optimizing later.

So my basic steps are this (always keeping yellow center on top):

Cross:
1. Get all the white edges around the yellow center, paying no attention to their permutation, only orientation
2. One edge at a time, turn the top until the edge's non-white color is lined up with the correct center around the outside, then flip that side 180 degrees to put the edge in it's correct position around the white center

First layer:
1. Find a white corner and get it in the top layer. If it's in the bottom, hold the cube so that it's in the FDR, then execute R U R' (we call it up left down)
2. Position the corner over the correct 'slot' by turning U, then hold the cube so that the corner is now in FUR position. Note that everything we do is always working with the right side for consistency.
3. If the white sticker on our target corner isn't pointing to the right (isn't in the R face), execute R U' R (this one is called up RIGHT down - almost identical to the previous move). The corner will need to be put over the correct slot again (repeat step 2). If the sticker still isn't pointing to the right, do it again.
4. R U R' (up left down). The nice thing here is that the moves in step 1, 2, and 4 are basically all the same move, only adjusting which direction you turn the top. The move is short, too, and easy to learn. Basically the top layer goes left when you're moving it from the bottom to top or top to bottom, and it goes right if you want to keep it in the top but twist it.

Second layer:
1. Find an edge in the top without a yellow sticker. Turn the top until the outer sticker matches the center just beneath it. This is now your Front.
2. Execute F U F U F - U' F' U' F' or it's mirror. To teach this move, I've taught my son to look at the right and left side to find which one has the center that matches the edge we're about to place. This determines if our first move is clockwise or counter clockwise. We'll just turn the front in whatever direction puts the edge into it's desired place. Now we have to decide which direction to move the top. One of the two front top corners will have a white sticker pointing to either the right or left. Whichever direction it's pointing, that's the direction the top will go (right or left). I describe the move to my son by finding the first two turns in this manner, then telling him to repeat them alternately 5 times, then backwards 4 times.
3. If, at step one, there are no edges in the top without yellow stickers then we need to break one out of the second layer. To do this we identify an edge we want to break out, then identify any edge currently in the top that has a yellow sticker. Use the moves just described to put that yellow edge into the place currently occupied by the edge we want to break out.

Last layer
We haven't gotten this far yet, but I'm planning on teaching him to use F U R U' R' F' to orient edges. So long as the cube is held such that the UL edge is already yellow side up when you begin (unless all 4 edges need orienting, then it doesn't matter), it's just a matter of repeating the move until all edges are upright

For orienting corners, I plan on teaching him to position a 'bad' corner in the UFR position, doing R' D' R D until that corner is yellow side up, turning the top to put another bad corner in that position, and repeating until all corners are orienting.

Then I think I'll teach him to permute corners and edges with two separate algs. That's going to take a little work at memorizing, but I'm finding he's got a pretty good memory. I'm open to suggestions on the absolute easiest algs possible for memorizing, or any modifications at all to my method at approaching the last layer.

Hope this helps someone else with teaching a kid. Honestly, I've been using this method to teach adults, too. It's broken down into very easy steps and scales well if they want to learn to get faster by optimizing certain steps. It's always memorizing the permutation algs for the last layer that throws people.
 
Then I think I'll teach him to permute corners and edges with two separate algs. That's going to take a little work at memorizing, but I'm finding he's got a pretty good memory. I'm open to suggestions on the absolute easiest algs possible for memorizing, or any modifications at all to my method at approaching the last layer.

You could try to teach him an understandable way of permutating the last layer.

1. Take the UFR corner out with R' D2 R (D2 is important to avoid confusion)
2. Turn the top layer so that the corner you just put into the D layer belongs to UFR
3. Insert the old corner to UFR (R' D2 R).
4. Repeat it until all corners are solved.

If he's familiar with orienting corners with R' D' R D he probably won't have problems with the "turn the U layer" part.

The same applies to edges, just with M D2 M'.

btw. most people i've teached solving a cube had big problems with the corner orientation part (the commutator way). They kept up rotating the whole cube instead of just turning the U layer.
 
Last edited:
Then I think I'll teach him to permute corners and edges with two separate algs. That's going to take a little work at memorizing, but I'm finding he's got a pretty good memory. I'm open to suggestions on the absolute easiest algs possible for memorizing, or any modifications at all to my method at approaching the last layer.

You could try to teach him an understandable way of permutating the last layer.

1. Take the UFR corner out with R' D2 R (D2 is important to avoid confusion)
2. Turn the top layer so that the corner you just put into the D layer belongs to UFR
3. Insert the old corner to UFR (R' D2 R).
4. Repeat it until all corners are solved.

If he's familiar with orienting corners with R' D' R D he probably won't have problems with the "turn the U layer" part.

The same applies to edges, just with M D2 M'.

btw. most people i've teached solving a cube had big problems with the corner orientation part (the commutator way). They kept up rotating the whole cube instead of just turning the U layer.

I posted a thread about teaching young kids how to solve here:
http://games.groups.yahoo.com/group/speedsolvingrubikscube/message/40708

For my job we run week-long track out camps for year-round students, and part of the camp is Rubik's Cube half hour each day where I teach them to solve. When we first started this I used to just start teaching them how to solve, but I found that even simple concepts such as knowing they have the ability to turn the F face at any given moment were difficult for them to grasp.

I've honed my method over about 2 years of teaching now, and I always tell kids that they can learn how to solve, but I have 9 puzzles they must do first, or they don't get to learn how to solve the cube. This gives them a goal to try to achieve, and also the puzzles are instructional and teach them some basic ideas about the cube, which I call "cube sense" like "number sense." The puzzles are described in the thread linked to above, with the only exception being that puzzles number 4 and 5 are switched now since I wrote that guide.

Basically I teach them the checkerboard pattern, then the spot pattern. Then I teach them (R2U2)*3. After that every single puzzle uses (R2U2)*3 in some way shape or form to solve. For example U' (R2U2)*3 U or B (R2U2)*3 B' etc.

This is teaching them the concept of memorizing an algorithm, and learning to see how you can use the effects of that algorithm to solve other types of situations. After they complete all 9 puzzles I then began to teach them a solution. I think your solution for the first two layers is fine, it makes sense and it is easy to do without having to learn any code or lingo for your son, which will mean he will intuitively understand what he is doing.

For the last layer, I teach just like Tim does, I show them F R U R' U' F' to orient edges, and use commutators to flip corners (R' D2 R F D2 F') and the same exact ones Tim describes for permutation. To avoid the mistake of them turning the cube instead of the U layer I describe the U-layer as one of those pin-wheel fans for kids where you blow into it to spin it really fast. I tell them that they have to spin the pin-wheel to get the next corner to the UFR spot. I describe the move R' D2 R (for permuting) as a baseball glove move. I show them how the UR edge and the UBR corner together stay connected, and I call that the "baseball glove". The UFR corner is the baseball, and it gets thrown to the back with the D2 move.

I've found that kids who have done all the 9 puzzles I have first, developing a sort of basic cube sense, can learn to solve it usually within 1 week after that, which is the equivalent of 2.5 hours of instruction spread over a week.

I think you have really good techniques for teaching your son, and definitely consider a commutator last layer so that he won't just learn it via short term memory and possible forget 1 year from now, but will understand intuitively what he is doing.

Let us know when he solves it, it sounds like he is already close!

Chris
 
That sounds almost exactly like the method I used to teach my then-8-year-old and then-6-year-old daughters to solve the cube. They are now both (barely) sub-minute solvers, and the then-6-year-old (who is now 7) can solve a 4x4x4.

I taught them the commutator for orienting corners, but they've both now dropped it, and that helped their times about as much as anything.
 
I think little kids have great memory, but lack in understanding. They could easily recognize cases and apply algorithms, but probably suck at FMC. They can memorize many algs very easily! Learning Fridrich F2L is probably more difficult than a 2LLL.
 
I think little kids have great memory, but lack in understanding. They could easily recognize cases and apply algorithms, but probably suck at FMC. They can memorize many algs very easily! Learning Fridrich F2L is probably more difficult than a 2LLL.

*Nudge**Nudge* I'm 13 and I don't suck at FMC. Plus, my LL sucks more than my F2L.
 
I think little kids have great memory, but lack in understanding. They could easily recognize cases and apply algorithms, but probably suck at FMC. They can memorize many algs very easily! Learning Fridrich F2L is probably more difficult than a 2LLL.

*Nudge**Nudge* I'm 13 and I don't suck at FMC. Plus, my LL sucks more than my F2L.

Then perhaps you aren't as little as you think you are. :D
http://en.wikipedia.org/wiki/Jean_Piaget#The_stages_of_cognitive_development

By little kids, I guess it would be stages three or two.
 
Wow, great ideas guys. I wasn't even aware myself of how to use some of those concepts. When I solve LL myself I'm doing it with memorized algs but don't really understand what I'm doing. Those moves made a ton of sense though. I look forward to passing this stuff on to my son. He's totally excited about the idea of going to a competition with me.
 
how about teaching zbll lol

No, they wouldn't understand how to recognize the cases and will grow bored of learning algs.

So would we.

Because we're still kids on the inside, right?

dude that's deep

Pun intended?

Or was the pun the only content of the message in the first place, and by posting this I'm making a fool of myself?
 
Thought I'd come back on post an update - My kid solved his first Rubik's cube, start to finish, Sunday June 27. It was a big moment :)

I tried several times over the last couple years, but he honestly wasn't interested enough to get past his frustrations until recently - and then it all clicked pretty quick. I used the moves Tim mentioned for LL. The great thing about those is that permuting corners and edges are almost identical, which made teaching that much simpler. Thanks Tim!

He's already asking about bigger cubes. I want him to be more comfortable with the 3x3 first. We'll tackle those a big later.

In the meantime, he's working on speed. There's a certain website he likes to use for timing his solves. What was it called....
 
Congratulations!

I am currently teaching my 5 yo daughter and reading the thread it seems like I may be in for quite a journey.

I recently started a thread for kids accomplishments where parents / syblings can post the progress of the youngest cubers.
Feel free to post your son's progress to inspire others.
http://www.speedsolving.com/forum/showthread.php?t=22183

Now that it has been bumped we could use this thread for kids teaching discussion.
 
I'm looking forward to teaching my nephew (now almost 2 years old), but I'll be teaching him Petrus. I'm afraid I'll have to have 3 or so years of patience before teaching him, though.
 
Last edited:
Thought I'd come back on post an update - My kid solved his first Rubik's cube, start to finish, Sunday June 27. It was a big moment :)

I tried several times over the last couple years, but he honestly wasn't interested enough to get past his frustrations until recently - and then it all clicked pretty quick. I used the moves Tim mentioned for LL. The great thing about those is that permuting corners and edges are almost identical, which made teaching that much simpler. Thanks Tim!

He's already asking about bigger cubes. I want him to be more comfortable with the 3x3 first. We'll tackle those a big later.

In the meantime, he's working on speed. There's a certain website he likes to use for timing his solves. What was it called....

Cubetimer?
 
Back
Top