Do you want reliable trained behaviors? Do you want your learner to enjoy the experience and crave more learning? Borrow some ideas from the best.
I’ve written about Angry Birds before, but I just touched on the player-friendly mechanics such as the lack of punishment. But I happened across a link to a user-experience (UX) design blog which analyzed the game in more detail, and there’s a lot of good information for trainers and instructors here!
First, of course, we have to see that the ideas we’re about to borrow are worth taking — is Angry Birds successful? In a word, yes. More than 50 million people have downloaded the casual game on phone, tablet, or other device, and while it’s a “casual game,” the users aren’t necessarily casual:
More compelling is the fact that not only do huge numbers download this game, they play it with such focus that the total number of hours consumed by Angry Birds players world-wide is roughly 200 million minutes a DAY, which translates into 1.2 billion hours a year. To compare, all person-hours spent creating and updating Wikipedia totals about 100 million hours over the entire life span of Wikipedia (Neiman Journalism Lab).
These players crave their game time in exactly the way that I want my animals to crave their training time — not for food, or from fear, but for fun. It’s a rewarding experience which is both simple and compelling enough to keep coming back. Awesome! Let’s get started.
What? You haven’t seen Angry Birds? Get a quick video overview here.
The first time a learner is introduced to the learning structure, it needs to be simple — not necessarily too simple, but definitely simple.
Simplification means once users have a relatively brief period of experience with the software, their mental model of how the interface behaves is well formed and fully embedded. This is known technically as schema formation. In truly great user interfaces, this critical bit of skill acquisition takes place during a specific use cycle known as the First User Experience or FUE.
Most of the time, the first behavior I shape when I introduce dogs and clients to clicker training is a target. There are a number of reasons for this:
- It’s very easy to capture in an average dog. I can easily talk to the client while simultaneously training, narrating what I’m doing.
- Most novice clients haven’t tried to train targeting, like they probably have with sit or down. This means that 1) there’s little chance of it being poisoned (and I sure don’t want to start with a poisoned cue), and 2) the client knows it’s from scratch, and then I and this method get full credit for teaching a new behavior in just seconds, helping them past any skepticism.
- The dog learns from the start that the clicker is attached to movement and behavior; it’s not just a random treat generator.
- And, of course, a target is very useful in many other contexts, so it’s handy to introduce early in training.
- Most of all, it’s a very simple discreet behavior, easy to repeat, with a high chance of success.
Put in the UX terms, the FUE with the clicker is simple and successful. That sets a great foundation for building a compelling and addictive training relationship.
How quickly does the user — our learner — get feedback and the next opportunity?
The surprising point that is often misunderstood is that not every aspect of the user interface needs to be or should be as fast as possible.
At first glance, this seems like heresy to the polished clicker trainer. After all, we know that the canine brain links cause and effect more efficiently with a delay of .39 seconds (putting those archaic training books advising to respond “within five seconds” to rest) and that the reinforcement should arrive about a second after the initial marker signal. How can delay then be a good thing?
But our UX experts aren’t recommending a delay in feedback so much as opportunity to observe and self-assess — which is critical for learning and for preparing the next response.
For example, in Angry Birds, it was possible for the programmers to have made the flight of the birds fast – very fast, but they didn’t. Instead they programmed the flight of the angry flock to be leisure pace as they arc across the sky heading for the pigs’ glass houses. This slowed response time, combined with a carefully crafted trajectory trace (the flight path of the bird), solves one huge problem for all user interfaces – error correction. The vast majority of software user interfaces have no consideration for how users can be taught by experience with the system to improve their performance.
If the player of Angry Birds had to launch his birds rapidly, with no chance to assess his choices, it would lead to a lot of wasted repetitions and greater frustration. Assessment of results is vital.
For our animal learners, then, we want to make sure they have a clear opportunity to to self-assess as well. If a dog pokes his nose at the treat hand instead of the target, I don’t need to micro-manage him back to work — I let him realize that it’s a wasted effort. And when he succeeds, I let him dance with joy as I produce his reward. Of course, we don’t want to lost time; repetitions still happen promptly and smoothly. But if the dog hesitates and then self-corrects his course, that’s far better than if I push him to act instantly but he makes a mistake.
I’m training myself new skills right now for a new hobby — more on that in another post — and self-assessment is a huge part of my routine. Speed without improvement is more wasted effort.
Short Term Memory & Training
Generally speaking, a training session works in short-term memory, while a behavior on stimulus control resides in long-term memory.
In computer-speak, human short-term memory is also highly volatile. This means it can be erased instantly, or more importantly, it can be overwritten by other information coming into the human perceptual system. Where things get interesting is the point where poor user interface design impacts the demand placed on [short-term memory].
Obviously, I’m going to make applications outside of the human mind, but I think they’re just as relevant. Taking data from one venue (one screen, in the UX example) to another venue (a second screen) is actually fairly challenging, because it relies wholly on short term memory, which can be easily disrupted. This is why you so often ask a phone number or the time, are distracted, and then need the same information again. But what constitutes a distraction?
This disruptive data flow can be in almost any form, but as a general rule, anything that is engaging, such as conversation, noise, motion, or worst of all, a combination of all three, is likely to totally erase SM.
Ye gads, it’s what we do to our poor animals all the time!
Target, click, treat, target, click, treat, target, click, treat.
“Wow, Ranger’s really doing great!”
“Yes, he is. Do you think he’s ready to move on?”
“He looks fantastic! Yes, try setting your target over there now.”
Ranger stares at the target. Looks at the trainer. Stares at the target.
“Oh, this breed is so stubborn!”
We introduced all three major distractions at once! Yikes! Ranger isn’t stubborn; we just made too much of a demand on his short-term memory.
The subtle, yet powerful concept employed in Angry Birds is to bend short-term memory but not to actually break it. If you do break SM, make sure you give the user a very simple, fast way to accurately reload.
In Angry Birds, that’s the chance to slide the screen for another look at the pigs’ fortress. In training, that’s the easy click for reduced criteria you give when the dog makes a guess and offers an incomplete behavior at the start of a new session. “You clicked for lifting my paw… Oh, right, paw target! Got it!”
I often hear people talk about trainers with “special dogs,” but I think far more often the case is not that the dogs are unusually gifted or eager, but that the trainers have made training a simple and compelling environment in which the dogs enjoy playing. What can you do to addict your dog to training? Tell us in the comments!
Read the entire UX breakdown here: http://www.mauronewmedia.com/blog/2011/02/why-angry-birds-is-so-successful-a-cognitive-teardown-of-the-user-experience/