Categories
Game Design

Multi-level decision making.

At any moment during a game players are liable to be thinking about events in multiple timeframes at once. Performing tasks that are over in seconds, in order to achieve goals that are over in minutes as a means of completing missions that may take hours.

The lowest level of actions occur on the Immediate layer, these are the second to second decisions made in the heat of combat, during a conversation, or while climbing a wall. When and where to shoot, which dialogue line to select, which handhold to reach for. These events are the Encounters, over in seconds and repeated dozens of times during the course of the game. The narrative strength of actions in this layer is best served through directly embedded content. Animation cycles, dialogue lines, and the options available to the player all serves as vectors for narrative meaning.

Above this there is the Tactical layer, the longer term minute to minute decisions made in the execution of plans. Which particular enemies to engage, which NPCs to talk to, which wall to scale. These are the Objectives, and can be defined either explicitly by the game, or implicitly by the players themselves. Variation in these Objectives and the levels in which they take place can be used to provide narrative.

Operating above both of these there is often, but not always,  a Strategic layer, actions on this layer occur over a much longer term, possibly hours. They include, which missions to accept, which character upgrades to select, which tools to equip. They can be either explicitly defined as Quests chains, or often they are not defined at all the goal of the Strategic layer simply being to reach the end of the game. This layer is best used to define the narrative context for actions that occur on the lower layers.

  • The Immediate layer is Reactive.
  • The Strategic layer is Proactive.
  • The Tactical layer is both.

Though some traits can be associated with each layer,  the boundaries between them are fairly permeable. The goals of the Immediate and Tactical layer are often elements of those defined on the Strategic layer. Strategic goals lead to the creation of multiple Tactical goals, and multiple Immediate goals will be needed to fulfil a specific Tactical goal.

If the Strategic goal is to get to a specific location, it might require engaging in combat with several groups of enemies. This leads to the creation of Tactical goals concerned with how to deal with each enemy group and in what order. These Tactical goals in turn lead to the creation of Immediate goals, when to shoot, where to move. Successful completion of the Strategic goal requires successful completion of the Tactical and Immediate goals that stem from it.

In this way it can be seen that actions on the Strategic layer directly influence those on the lower layers.

Layers

Plans trickle down from higher layers to lower ones. Immediate actions are defined, their scope is limited by decisions made on the Tactical layer. Where you are and which tools you have at your disposal are based on decisions made at the Tactical layer, which in turn are influenced by decisions made on the Strategic layer. If a particular character upgrade has not been obtained on it will not be available.

This flow of influence does not only occur in one directions. Actions and their consequences trickle upwards. Events that occur in the Immediate layer change the Tactical status of the world, new routes are located, items are found. Events on the Tactical layer in turn affect the options available in the Strategic layer. Meaningful actions are ones that send ripples out beyond the layer in which they occur and affect decisions made on all layers: actions on the Immediate layer that leads to consequences on the Tactical and Strategic layers.

In ludic terms each layer has some degree of repetition, as there are only so many valid actions that can be performed at any given time. The repetition is mitigated most on the Strategic layer because the goals are long term, any repetition that does exist occurs over the course of several hours making it difficult to ascertain any patterns in the type of actions being performed.

On the Immediate layer the sense of repetition can be the strongest, as often the core mechanics of a game only allow for a few options. However at this layer the direct connection between action and outcome serves to lesson the impact of the repetition, as the consequences of actions on this layer are often the most directly stimulating, the blood spurts of a successful headshot, the ding of a loot pickup, the fluid animation of a character clambering over a ledge. Each one a little endorphin kick that keeps us engaged; if anybody is in doubt I point you to the immediate feedback presented in a game like Diablo.

The biggest problem with repetition comes on the Tactical layer. Action games suffer the most on this layer. Consider Far Cry 2, the actual combat mechanics and the options available to players in combat can be quite engaging (The Immediate layer is well designed). The ability to select which missions to attempt and in which order lessens the restrictive sense of repetition on the Strategic layer. However regardless of which mission the player selects and for whom, the short term goals required to complete each are usually very similar, if not identical: go here kill, these people\find this item, get back to here. The execution of these individual Tactical goals on the Immediate layer might be entertaining but that does little to cover up the fact that players are basically doing exactly the same thing during each mission. This is not helped by a lack of narrative feedback regarding the overarching consequences of actions. Assassinating a Police chief might be contextualised differently to the assassination of a Warlord but the narrative feedback from each event is not differentiated enough to mask the underlying repetition.

Because Tactical goals can take minutes to an hour to complete they occur over too short a timeframe for their patterns to be lost in the noise of all the other decisions, yet at too long a timeframe for that endorphin kick to keep players engaged. It’s here that a strong narrative context can keep players engaged in performing what are mechanic very similar actions.

Halo: Combat Evolved is another prime example of a game that suffers on the Immediate layer. Those “thirty seconds of fun” are, at least for me, some of the most pleasurable in gaming, but there can be no denying that on the Tactical layer the game is little more than a sequence of goals of the form: “Kill these hostiles.”

With their focus on Immediate and Tactical actions, action games are geared to a shorter play session, that serves to mitigate their repetitive nature. Plans are often completed within seconds or minutes, so players are given more points at which they are “free” to quit because they have no plans remaining to complete. Under these circumstances it’s little surprise that action game stories are fairly perfunctory, serving only to cover up the core mechanical repetition and provide a loose context for who, where and why.

In comparison a high level strategy game (An accurate genre name if ever there was one) like Civilization IV relies almost entirely on actions playing out on the Tactical and Strategic layers. This leads to a long term investment as players keep playing in order to see the consequences of actions, the beloved\cursed “one more turn” syndrome. Goals at these layers are well served by a more “hands off” approach to narrative, as players will be less likely to baulk at the lack of direct feedback on the Immediate layer, when they have played a greater part in the selection of the Tactical goals that led to those Immediate actions.

3 replies on “Multi-level decision making.”

“The narrative strength of actions in this layer is best served through directly embedded content. Animation cycles, dialogue lines, and the options available to the player all serves as vectors for narrative meaning.”

Love that quote. It’s is a good analysis of what you described as layers. This is what you were worried about making point on? If so, I thought it was pretty damn clear (though this is me we’re talking about). The only thing I’d question would be for you to keep digging in to the points you described here. I haven’t combed through the backogs of your posts though and only glanced at your others this one links to. It might be hard though, and the only thing I can suggest would be applying your logic here to a singular game as a sort of paradigm post.

~sLs~

Liked the note about strategy games and the strategic layer at the end. Have you tried Dawn of War II? They do this sick shit where they’ve got the form of a topdown strategy game but it’s really a team-based shooter–so it’s emphasizing the immediate a lot more. I absolutely cannot grok it without spending numerous hours end-to-end. Miserable and engaging at the same time.

I like to separate layers even more in order of time spent, to help combat the type of problems you describe. So decision loops made in 1-5 seconds, 5-10, 10-30, 30 to 2 min, maybe 5min and 10min if the game really allows.

It’s most interesting to take a really successful open world game like earlier GTAs and break down just where the decisions are, because if you don’t analyze it that way, it’s a mess. You’re driving, you stop to do a random mission, meanwhile you’re making your way across town for something else – there’s a lot of feedback between decision/time layers that you’re not aware of otherwise, and is essential to making those games work with such loose mechanics and hodge podge of gameplay, your attention is entertained by shifting rapidly between layers.

Leave a Reply

Your email address will not be published. Required fields are marked *