Learning curves

/

Alto's Adventure [iOS] is a beautiful little game. You start as Alto, a snowboarding boy who is trying to catch runaway llamas. As you progress, you can choose to play as some of Alto's friends (my husband's favorite is the quick-flipping Maya; I tend to prefer Izel, who has apparently attached rockets to her snowboard). The rules of the game are deceptively simple: don't fall, and don't get caught by a disapproving elder. For much of the game you have only one control: tap on the screen to jump (if you keep your finger pressed, the jump turns into a flip). By timing your jumps well, you can land on and slide along bunting lines and tiled roofs, or you can bounce off of rocks and campfires while catching llamas along the way down the snowy mountainside. You can also shoot over chasms or perform tricks as you fall through the air. When you land on your feet, you get a speed boost that can smash through rocks and help you escape pursuit. If you land on your head, get your snowboard stuck in the snow, faceplant into a rock, or tumble down a chasm, your score (from performing tricks, catching llamas and successfully escaping elders) is tallied, and you start back up at the top of the mountain. That's it.

Well, that's the basics, anyway. Each time you start down the hill, you get a list of three goals. Complete all three goals, and you level up. Get enough levels, and you unlock a new character. Your first three goals are:

  1. Land a backflip
  2. Grind a bunting line
  3. Jump across a chasm

The first time I tried to have Alto backflip, I landed the poor kid on his face. I'm pretty sure I did that the second time, too. And possibly the third.

I've played this game for a few weeks now, and I've gotten well past Level 1. I can land triple backflips and flow from one trick to the next. But each new level has new goals, and I often fail at them the first few (dozen) tries.

Playing Alto's Adventure last night, it struck me that I'm learning. In many kinds of games (e.g. RPGs1) the more you level up, the more advanced your gear becomes. There are a few upgrades possible in Alto's Adventure, but really you are leveling up, not your character or your gear. By playing the game, you get better at it, practicing tricks over and over again until you get them right. It doesn't always feel like learning, though. And it certainly doesn't feel like failure most of the time. (Except when you finish a stylish and well-executed string of flips and grinds with a faceplant. That still feels like failure.) You just take Alto back up to the top of the mountain and try again.

I have two trains of thought on this:

  1. I am the kind of student (and academic) I am because I'm the kind of person who is doggedly persistent when something doesn't work the first (or second or twenty-seventh) time. PhD = persistence to a high degree, right?

  2. Many of my students are not persistent. At least, they're not persistent when it comes to chemistry. They try something once (and some of them try only half-heartedly), and if they don't get it to work right away, they get discouraged. This is a problem if I want them to master chemistry. (I do!) I suspect, though, that many of these same students would happily play Alto's Adventure and perfect the difficult tricks.

I think there are some lessons in teaching we might take from the game design of Alto's Adventure. First, start simple. The first three goals are the simplest actions beyond just jumping. Before you get complicated tricks (e.g. backflip and then bounce off of a rock), you start with the fundamentals. Second, provide clear, concrete (and concise!) goals. The first goal isn't "Backflips are the simplest trick, worth 10 points," it's an actionable task: "Land a backflip." The success and fail states are pretty clear, too: if you land it, you keep going, and if you don't land it, you get stuck in a snowbank (or fall into a chasm). Third, list just a few goals at a time. You get three goals per level, and you have to master all three before you can move on to the next set. You can't skip it and come back: you just have to practice. Fourth, give timely feedback. When you achieve a goal, a message shows up briefly on the screen. The next time you fail, the goal is crossed off the list. If you've completed all the goals for a level, you get the next list of goals. Fifth, make failure normal. You are going to make mistakes. As you learn, you will make new and different mistakes, but you will never stop making mistakes. When you plow into a rock, the game says "Ouch! Try again." When you drop Alto into a chasm, the game says "Better luck next time." And you go back to the top of the mountain. There is no other penalty. You just get to try again.

I've heard about attempts to "gamify" courses, and I think some of the ideas I'm currently mulling over are related to standards-based grading.2 Somewhere in here, though, may be another way to encourage less persistent students to keep trying, or another way to quantify mastery. What if, instead of course learning objectives, each student had their own list of goals? They could "level up" their chemistry as they achieved the goals. (Yeah, I know, it sounds like a lot of work, and to be useful for a semester-long course the goals would probably need to be broad enough to apply across topics. Dream with me for a bit here.) What if there was a leaderboard for each topic? (Not grades because FERPA, etc., but maybe "mastery points" or something??) Different students have different strengths, even within a general chemistry course. It might foster better peer instruction.

As is probably clear, my thoughts on this topic are half-formed and disorganized. I have a bunch of (possibly crazy) teaching ideas swirling around my head, but I think it's also worth mentioning some observations I have made and a few things I have already tried.

Despite having taught college freshmen at both UM and MSU for a few years, I am still surprised by the number of students with weak algebra skills and poor reading comprehension. I am not going to wag my finger at high school teachers about this. I've seen my mother's elementary classes; I know what kinds of challenges there are. It just surprises me that, with the amount of effort teachers are expending and the sheer amount of time college-bound kids spend in school, more learning doesn't stick. I understand if a student has rarely used logarithms and needs a refresher, but the number of students who don't really know how to solve for unknowns, yet plan to get A's in chemistry, is boggling. I don't expect fully polished treatises for answers to exam questions, but I find the frequency with which students fail to answer the question posed, and their propensity to write responses to "Why does X do Y?" in the form "X does Y because it is X" a little alarming. We need to start courses with their fundamentals, but the fundamentals of chemistry are words, numbers and unknowns.

For several years I have created lists of learning objectives to help me focus my teaching. I treat it like a checklist: Today the students should accomplish these few things; I need to make sure they have the necessary background and practice to do so. This semester, on the advice of a colleague, I did something simple, yet helpful: I shared the list with the students. I started every lecture with "Today's Plan," the list of the day's learning objectives. It was an attempt to give the students clear, concrete goals for the material they were learning. For example, instead of "kinetics" or "rate laws" I'd say "Write a rate law expression for a given reaction." I think it helped. (Can't quantify it, though, so anecdata only.) Admittedly, we didn't always achieve every objective the first day it was listed, but that also tended to happen when I didn't narrow down the list enough. (See lesson #3 from Alto's Adventure: list just a few goals at a time.)

Lastly, I am trying to make my students comfortable with making mistakes. One thing I try to do is be okay with my own mistakes. If I can be forgiving with myself and be gracious when I am corrected, I can be a good example to my students. I want them to see that the professor isn't perfect.3 Another thing I do, particularly at my office hours, is present questions as puzzles or challenges. Last semester, when we did a unit on basic organic chemistry, I gave a student a trickier-than-usual structure to name so he could get the hang of the naming conventions. He'd take a stab at it, and then I'd have him show me which part of the name corresponded with which part of the structure. If it didn't match, he'd try again. It was more like a brainteaser than a chemistry assignment. He didn't mind failing a few times before he finally got it. He even appeared to enjoy the practice. I want to find ways to scale that experience up to the whole class.

I'm going to keep chewing on these teaching ideas. In the mean time, I highly recommend Alto's Adventure as a quick pick-up-and-play game (or a where-did-the-last-2-hours-go game… ). Enjoy.

1: RPGs = role-playing games

2: I'm not sure if I have the right terms, since I'm not very familiar with the literature yet. Be nice to me; I'm new to this field.

3: I do worry a little bit that a new, young, female professor making mistakes can be seen as "incompetent," rather than "human," but I can't really win that battle…