Learning in great video games
As Linderoth pointed it, video games may be designed to minimize the risk of players giving up because they failed. A game which is too easy would also be boring; a balance between boredom and anxiety has to be found to reach "optimal experience" (Csíkszentmihályi, 1991).
In my PhD research, I study how learning works in "great video games." Great games should take "one minute to learn and a lifetime to master" (Kunkel, 2003). This particular learning is what makes great games great. If a game is able to maintain this balance between boredom and anxiety, then it remains interesting for a long period of time even though the player improves her or his skill level. Linderoth (2010, p 2) defines learning "in ecological approach [as] becoming attuned to perceiving and being able to utilize specific sets of affordances that belong in specific practices." An affordance of learning would be a possibility of action offered by a game to a player which results in a modification of the player's abilities to perceive or use this game's affordances. These abilities may be physical as well as mental. Learning would be an action improving the player's ability to perform actions. Thus, great games have properties allowing this type of learning.
These great games are not necessarily Massively Multiplayer Online Role Playing Games, in which there are far too many possible things to do. Great games may be arcade games you can finish within one hour. These games have to fit the definition of great games because of the context in which they are meant to be played. Originally players had to pay every game they played. In order to earn as much money as possible, an arcade game has to be easy to understand so that the player may quickly have fun but also hard to master so that the player keeps playing (and paying).
Shoot'em-up games are a very good example of arcade game. Once the player has understood how to move and fire, she or he can play them, but the games remain difficult. "MON" is a Japanese supergamer on shoot'em-up games. More precisely, he plays danmaku games. The Japanese word danmaku literally means "bullet curtain" ("bullet hell" in English) and it refers to a shoot'em-up game in which the screen is practically covered with enemy bullets. MON estimates that in order to finish one of the most difficult mode of DoDonPachi Daioujou (Cave, 2003), called "Death Label" mode, a complete beginner would have to play 4 hours a day... for ten years (Kemp, 2011, p 166).
However, as difficult as it may be, finishing an arcade game does not necessarily mean having mastered it. The danmaku called Ikaruga (Treasure, 2001) was first an arcade game; on the Gamecube version (Ikaruga, Treasure, 2003), the player is given a new credit for every hour of play. Every credit increases her or his chances to finish the game, after seven hours of play the "infinite credits" mode is unlocked. As every level is timed, it is not necessary to achieve something to get to the level after. If the player does not defeat the boss before the time is up, he just goes away and the next level begins. No matter how bad you are, you just have play long enough in order to finish the game, but there is a quantum leap between finishing it and having a decent high score. The score is reset to zero every time a credit is used, so it is necessary to finish the game with one credit before even thinking about getting a high score. Michael Molinari (2009) reviewed 83,279 high scores of Xbox Live Ikaruga players:
Knowing how brutally the learning curve treats players, the top score of 34.4 million points is quickly cut in thirds to 10.3 million by the 500th player. [...] At 10,000 players, the score is at 1.3 million. From there, it has a steady decline in scores until around the 83,000th player, who has 11,100 points. [...] And speaking of low scores, I understand that the game is a tough cookie to play, let alone master, but many of these scores are just horrible. [...] If you just hold the fire button down and stare at the screen, you get a game over with 40,000 points, a score that over 4,000 people failed to acquire.
Ikaruga is known for its difficulty. Molinari considers that "everyone dies very early upon first playthrough, [and] the next 50 to 500." There are really bad players and really good players, but bad players are far more numerous. However even top players were once beginners and had to learn how to reach this level; this learning did not work for everybody.
Tetris (Bullet-Proof Software, 1989) is typically a game that is easy to learn and hard to master. But at very high speed, the original game is impossible to play because it was not created to be played very fast. The Tetris: The Grand Master (Arika, 1998) series gathers versions of the original Tetris that are specifically designed to be played at a very high speed. At the maximum speed, a tetramino takes one frame to go from the top to the bottom of the screen. In Tetris: The Grand Master 2 (Arika, 2000), obtaining the highest rank, "Grand Master," requires one to get the "Master" rank, which is already hard to get, and then to survive one minute at the highest speed without seeing the tetraminos. According to "Amnesia," a French supergamer on this game, this minute of invisible play represents a real gap and requires an additional year of training after getting the "Master" rank (Pilot, 2009). The mechanisms of the game have been changed in order to increase the learning possibilities of the original game concept.
"DamDam," (Pilot, 2008) a French supergamer on musical games, talks about "infinite superplay" on Pop'n Music (Konami, 1998). Pop'n Music is a rhythm game which may be compared to Guitar Hero (Harmonix, 2005). Instead of five buttons aligned on the handle of a guitar, there are nine big buttons in two rows on a one-meter board. Sometimes, the player has to press five buttons at the same time. The game presents an almost unreachable challenge, because even the best player in the world has not reached the highest score on the most difficult songs. In DamDam's opinion, Guitar Hero is too easily beaten because it is possible "to finish the most difficult song with the maximum score within six months," which is far too short (Falcoz, 2011).
Clearly, this type of learning does not work for every player. Progression in these cases is not about learning how to perform a new task, rather it is about learning how to perform the same task better despite the fact that the game remains exactly the same. So, a great game offers, to the same player and at the same time, several different affordances to complete the same task. These affordances require more or fewer skills and more or less knowledge. The more difficult an affordance is, the more performing it rewards the player. Great games have properties that offer a fast and yet long learning. The purpose of this article is to study the properties of a video game and the capacities of the player affording this kind of learning. To do so, we will carefully study one particular mechanism in one specific arcade video game. Thus, we will highlight how a mechanism may be quick to learn and yet, take a long time to master.