Scaffolding

In this month’s application of educational theory to gaming, let’s talk about scaffolding. Scaffolding is providing successive levels of support and difficulty to develop capacity and mastery. This should be a familiar concept for gaming, where we have literal levels that usually move from a simple tutorial to greater complexity. Games are increasingly being used as models for education because of this common and successful way of introducing new complexity and skills while remaining engaging. Indeed, our standard A Theory of Fun model of gaming fun is basically learning. But games also provide many counterexamples, where scaffolding is intentionally or unintentionally missing.

Last time, I compared games to assessments, where the game is basically a challenge to be overcome. Games can be better or worse at this, just like you have sat both reasonable and B.S. exams. Scaffolding is part of the connection between the material and the assessment. It is the difference between teaching and lecturing plus tests.

Portal is a great game for demonstrating scaffolding. Portal has gotten some crap for this over the years, but it demonstrates exactly how teaching is supposed to work. Start simple. Introduce new things one at a time. Explain new things. Provide assistance in the face of frustration. Gradually combine simple things to build complexity. Try to find the edge of the player/learner’s competency and ride there, increasing difficulty along with growing competency.

No game is going to do this perfectly for everyone. People have different learning curves, and different curves for different sorts of learning, while a fixed game has only so much wiggle room for that. Different difficulty levels in games are usually changes to the total difficulty, not the rate at which difficulty increases.

Psychonauts is a game like Portal that is known for being mostly tutorials the whole way, followed by a huge difficulty spike at the end. That is a gap in the scaffolding. The game is a bit less tutorial-like than it may seem, but difficulty does tend to remain rather low. Where Portal gradually accumulates complexity, Psychonauts is closer to treating each simple piece individually, then throwing them all together at once with three new things, good luck! Very good scaffolding at the bottom, but it needs to build up. Long after release, there was an update that reduced that late difficulty spike, but it did so by reducing the height of the spike, not providing more scaffolding on the way up there. It is still a spike, just a smaller one. It is a matter of degree, not kind.

Super Mario Brothers is a classic game that does this pretty well. The first level is very simple, but an engaging challenge for a first-time player. It contains most of the building blocks of the full game, but in small pieces and low stakes. It has simple enemies and a chance to learn the basics of the game. It has just a dash of the hidden complexities of the game, with things like invisible blocks and underground paths. Later levels will gradually introduce more elements and then ramp them up. Witness the castle fights, with Bowser gradually getting hammers and fire. Witness levels that are essentially the same thing with a new element, like adding a spinning line of fire to a castle and then a really long line of fire. In terms of scaffolding, many recent games still have much to learn from the oldest Nintendo games.

Many complaints about games are not about the game content but rather the way the content is introduced. Often, the content is not introduced, just thrown at the player. That can be intentional, because some people like being thrown into games and made to figure out what is going on, but a large (and loud) number of players do not enjoy that. Advertising “this game has no hand-holding” can reduce the complaints, but the lack of scaffolding is still there. If fumbling around is the goal, there you go, that’s easy to create. If part of the goal is developing player mastery of the game, it might help to look at what helps develop player mastery rather than treating the existence of the challenge as significant.

For example, Monster Slayers does a better job than Slay the Spire of gradually introducing small units of complexity and explaining them as it goes. It has less total complexity, and it does not explain all of it, but you are more likely to get introductions and explanations. Slay the Spire has explanations, but you are expected to pick them up along the way and notice them on your own. Things are clearly labeled, but you need to look for the labels, and you may be dealing with a few new things very quickly.

The penalty for failure factors in here. Permadeath is a frustrating mechanic significantly because of the need to go back through earlier levels. The annoyance is time lost repeating exactly the same thing that you already know, trying to sprint through that scaffolding. This presents the strange paradox that a lack of scaffolding combined with permadeath is in one sense awful, in that you are starting over frequently as you repeatedly run into weakly introduced new things with a high chance of failure, but at least there is a not a lot of scaffolding to run through the second time. Monster Slayers introduces things better, but the first half of the game is faceroll easy once you have learned it and done the initial unlock grind; Slay the Spire ramps up difficulty more quickly and consistently, making the early game a meaningful challenge on every playthrough.

I think many complaints about games, their difficulty, their difficulty curves, etc. come down to poor scaffolding. If we wanted to be thrown into chaotic and difficult situations with few useful explanations in advance, we have our daily lives. Games are crafted experiences, and we expect better. Dropping someone in the deep end and expecting them to swim is lazy design.

You can overcorrect. Some tutorials are awful because of their length and the slowness with which new elements are introduced. You don’t want to treat people like they’re stupid. But you also have very different tutorials needed between genre veterans and someone who might never have played a game like this. The best introductions often explicitly ask give you entry points like, “I have never played tower defense before,” “I am a veteran and just need to know what you’re doing differently here,” and “let me get to it immediately, and give me hardcore mode.”

I think of novels by analogy. Classic sci fi is famous for exposition dumps, laying out at great length the physics, settings, etc. More recent sci fi often goes in the opposite direction, trying to provide a sense of “this is normal here” worldbuilding and either never explicitly explaining or intentionally pushing the exposition later so it is in smaller pieces or less obvious.

Either approach can be done well or badly, and different people will have different thresholds for what they consider “enough” and “too much” explanation. “Not enough” is easy to do in games, and many games treat it as a feature rather than a bug.

: Zubon

The link I wanted for “If all stories were written like science fiction stories” by Mark Rosenfelder (at Shrove Tuesday Observed) seems to have been bought by a spam site, or at least that is the case as I write this, and the Wayback Machine says, “Page cannot be displayed due to robots.txt.” (I suspect that changed since a previous archiving, given the exact link there.) I did find this fun discussion of the story and what it is parodying while looking for links.