Code & Go Robot Mouse

A small plastic mouse, a maze made of green tiles, and the first real test of whether your child wants to code or just likes the idea of it.

Your child has seen other children playing with robots on a screen somewhere, maybe a classroom video, maybe a friend's house, and asked if they can have one too. You typed "coding toy" into a search bar and immediately drowned in a sea of primary-coloured boxes with STEM written on them in capital letters. Most of them cost more than you'd like, and none of them tell you what actually happens on the first afternoon.

The Code & Go Robot Mouse from Learning Resources is one of the simplest and cheapest ways to find out whether your child has genuine interest in programming or is just chasing the novelty of a thing that moves. It's been around since 2016, which in children's tech terms makes it practically ancient, but it's still in production, still in schools, and still doing the one thing it was designed to do quite well.

The First Session

Here's what actually happens. You open the box, snap together sixteen green plastic grid squares into a flat board roughly 50 cm across, and slot some purple walls in to create a simple maze. You put the cheese wedge at the end. You put the mouse, called Colby, blue, about 10 cm long, with a slightly manic expression, at the start.

Then your child presses the coloured buttons on Colby's back. Blue arrow for forward. Orange for left. Purple for right. Green to go. The mouse moves in five-inch steps and turns in ninety-degree increments. If they've pressed the right sequence, Colby reaches the cheese and his nose lights up. If they haven't, he bumps into a wall or rolls off the edge of the board, and they have to clear the memory and start again.

A five-year-old will need you beside them for the first session, full stop. Not because the buttons are hard (they're colour-matched to the coding cards and perfectly intuitive) but because understanding that you plan the whole route before pressing Go is a genuinely new concept for a young child. They will want to press a button and watch the mouse move immediately. You'll need to explain that the mouse stores the instructions and runs them all at once. Once that clicks, most children are away.

The first session takes about twenty minutes of real engagement. The coding cards, which progress from dead-simple two-step paths to longer sequences with turns, are the best thing in the box. They give you something to hand to a child and say "try this one" without having to invent challenges yourself.

What It's Genuinely Good At

The mouse teaches sequencing, the idea that order matters, that step three depends on step two, that a left turn after two forward moves is not the same as a left turn after three. A child who spends half an hour navigating Colby through a maze they built themselves has practised the same kind of thinking that underpins every line of code ever written, and they've done it without anyone mentioning the word "algorithm."

The physical maze is the key. Because the child builds the walls themselves, they're doing two kinds of thinking at once: spatial reasoning about the layout and sequential reasoning about the instructions. When the mouse fails, when it bumps a wall or overshoots a turn, the child has to debug, even though they don't know that's what they're doing. They retrace the route on the coding cards, find the wrong step, fix it, and try again. That cycle of plan, test, fail, revise is precisely what programming is, dressed up as a mouse chasing cheese.

It is also entirely screen-free, which matters if you're deliberately looking for things that teach tech concepts without adding more screen time. Everything happens on a physical board, with physical cards, pressed by physical fingers. For a four- or five-year-old, this tactile approach is far more appropriate than dragging blocks around an iPad.

Before You Open the Box

Colby is not a precision instrument. On longer sequences, anything above about eight or nine steps, the mouse can drift slightly off course, particularly on turns. This is the single most common complaint from parents who've used it for a while. On a short maze, it doesn't matter. On a complex one, the mouse may clip a wall or miss the cheese by a couple of centimetres, and a young child will not distinguish between "my code was wrong" and "the mouse went wrong." If the mouse starts drifting, the first thing to check is the batteries: weak power is almost always the cause. Fresh AAAs fix it more often than not. Learning Resources now sell a rechargeable version with USB-C charging, which solves the battery problem entirely and is worth the slight premium if you plan to use this regularly.

There's another thing you should know about batteries. The compartment requires a Phillips screwdriver to open, which means you'll need one handy on the first day. It's not a design flaw, it's a safety feature, but it does mean an excited child cannot start playing the moment they unwrap it unless you've already done the prep. Have the batteries and screwdriver ready before the big reveal.

The maze board is well-made and the walls slot in firmly, but the box itself is useless for storage. You'll want a plastic tub or ziplock bag for the walls, tunnels, and cards within the first week.

The ceiling on this toy is real and arrives faster than you might expect. A bright six-year-old will exhaust the ten activity cards within a few sessions. After that, the longevity depends entirely on whether your child enjoys inventing their own mazes, and some do, enthusiastically, while some lose interest once the structured challenges run out. Learning Resources offer free bonus activity sheets on their website, which buys you a few more weeks. But by the time a child is seven or eight, the forty-step limit and the simple forward-turn-forward vocabulary will feel limiting. That's not a failing of the toy; it's a sign the child is ready for something more capable. Botley 2.0, from the same company, picks up roughly where this leaves off, with a separate remote programmer, object detection, and loop commands.

One more thing: you cannot correct a single step in a sequence. If your child presses the wrong button halfway through programming a twelve-step route, the only option is to clear everything and start over. For young children learning the basics, this is fine, routes are short. But as the mazes get more complex, it becomes genuinely frustrating. Teaching your child to lay out the coding cards in order before touching the mouse is the workaround, and it's actually better practice anyway.

How Long Does It Last?

At month one, most children are still working through the activity cards and building their own simple mazes. At month three, the ones who love it are inventing obstacle courses on the kitchen floor using books and pencil pots instead of the maze board, because the mouse works on any smooth, flat surface. At month six, you'll know. Either the child has moved on and is ready for something with more depth, or the mouse has quietly joined the toy rotation and still comes out once a week.

This is not a toy with a long growth path. It teaches one set of concepts, sequencing, direction, debugging, and it teaches them well. When your child outgrows it, that's a success, not a failure. The secondhand market is strong precisely because the toy is durable and doesn't go obsolete, so you can recoup some of the cost when you're done.

The Verdict

If your child is four, five, or six and you want to find out whether coding holds any real interest for them without spending seventy pounds on a kit that might gather dust, the Code & Go Robot Mouse is a smart first purchase. The activity set, which includes the maze board, is the one to get; the mouse on its own is less useful because you'll end up improvising mazes out of books and cereal boxes, which is fine but less structured for a child just starting out.

If your child is already seven or eight and has used Scratch or any other coding tool, this will feel too simple. Go straight to Botley 2.0 or an Ozobot instead. You'll spend more, but you'll skip the frustration of a toy that's already beneath them.

If your child wants a robot that does impressive things on its own, walks around, talks, performs tricks, this is not that. Colby sits still until your child tells him exactly what to do, step by step. That's the whole point, and it's the reason the toy works as a learning tool. But if your child wants spectacle rather than problem-solving, buy a remote-control car. Cheaper, simpler, and nobody has to debug anything.

At £40–£70 for the full set, or from about £15 secondhand, it earns its place. Not because it will last forever, but because it answers the question every parent asking about coding toys actually wants answered: does my child enjoy this kind of thinking? If the answer is yes, you'll know within a week. And if the answer is no, you haven't bet the house to find out.

© 2026 Topical Blogs · Product reviews for parents, not marketers