Breaking the Chains of Analysis Paralysis – The Black Box (Part 1 of 7)

chain-icon.png

If you’re here, reading a game designer blog discussing analysis paralysis, then I’m pretty sure you probably have some idea of what I mean by the words “Analysis Paralysis” (hereafter referred to as AP). If you don’t, don’t fret! The “What is AP?” section below and the included links should give you a solid primer on the subject. If you feel you have a solid grasp on the concept, feel free to skip over that section.

This is the first article in a series of 7 that I will be putting up in the coming weeks. The purpose of these articles is to highlight primary causes of AP among people who play games, and discuss strategies that you can apply as a designer to reduce the potential for AP if it is a problem. (Emphasis mine).

Each article in this series will go into depth on one potential cause of AP, examples of situations where it can be problematic, and present some solutions that can improve the situation.

  1. The Black Box – this article
  2. The Paradox of Choice – Dec 13
  3. The Prisoner’s Dilemma – Dec 20
  4. The Maze to Victory – Dec 27
  5. Relationship Status: It’s Complicated – Jan 3
  6. Sophie’s Choice – Jan 10
  7. Bigger vs Better – Jan 17

If you have any suggestions, examples, questions, or want to point out something I’ve missed, please do so in the comments. This is not a scientifically rigorous examination of AP, and I would be happy to include any contributions you have!

Before we dig in, let me be clear that games are all about making choices, making decisions, taking action. Any ‘game’ that does not have these elements are either an activity or gambling. Candyland is a valuable and useful tool for child development, but it is an activity. There is no choice, no decisions to be made, no way to influence the outcome, only pure chance. Any choice or decision is an opportunity for AP – a truly indecisive person may be paralyzed from the beginning.

The purpose of this article is not to eliminate choice, eliminate decision making, eliminate strategy from games. It is aimed at providing tools you can use to help make decision-making easier and less stressful for your players without eliminating the need to make decisions or the impact that those decisions have on the game. It is up to you, the designer, to test, observe, and determine if your players are showing signs of problematic levels of AP and whether or not it needs to be addressed.

What is AP?

Many people have written about AP far better than I can, so here are a few quick links if you want to know more:

Paraphrasing Wikipedia, AP is a term for being in a situation where you are over-analyzing or over-thinking a situation or a decision to an extent that you cannot decide what to do and as a result… do nothing. You are paralyzed with indecision because you are seeking the perfect choice, or are afraid of making a poor choice.

AP is not unique to games, it is something that can impact any aspect of life where decisions need to be made. Games simply provide a unique view into the phenomenon where (ideally) every decision is well defined, consequential to the outcome, and many decisions need to be made in a relatively short period of time. The results of AP are also much more apparent in a game than in life, long turns, inability to choose a course of action, frustration, confusion can all be hallmarks of game-based AP.

Why is it a problem as a designer?

There are two main sources of AP as a designer, the paralysis of an open design space, and the paralysis gamers may feel when playing your design. The first is the same kind of AP that you see across all aspects of life. The infinite possibilities of a blank sheet of canvas. A quick google search will turn up a multitude of strategies to help you overcome it, so here are a couple quick links and then I’m going to skip along to the second.

Player AP is a problem for your game because – well… it’s not fun. That’s the long and short of it. It slows down a game, it increases the time players must wait between interactions, and it can be frustrating and disheartening for the player who is having trouble.

Games that are prone to AP are more likely to result in fewer positive experiences for the players, and a less favourable overall impression of the game.

Now that is not to say that everyone feels this way. There are many gamers that enjoy the challenge of making tough decisions, seeing situations and opportunities that their opponents don’t see, and coming up with the perfect solution. These games are the brain-burners, and they will tax your ability to analyze a situation and make the best choices. But even if that’s the path you are aiming to pursue, there are ways to reduce AP in your game without impacting the burn.

The primary goal of reducing AP in your game, is not to reduce the challenge of decision making, but to reduce unnecessary delay caused by roadblocks that get in the way of decision making.

Why does it happen?

The cause of AP is very simple at its core. AP happens when the benefit of a choice is difficult to know. The more difficult it is to understand the benefits of one choice compared to another, the more difficult it is to choose.

That’s it. Very simple. It’s when we start drilling into the multitude of ways that the benefit of a choice can be obscured or be difficult to grasp is when things get complicated. But in general, these ways fall into three categories, the 3 Q’s: Quantity, Quality, and Clarity.

The first step towards fixing an AP problem in your game design is understanding the reason for it. So let’s start by looking at cause #1.

The Black Box

The problem of information obscurity.

Category: Clarity

Player Symptoms:

  1. Repeated examination of the same game elements/components
  2. Consistently referencing the manual to decode symbols and their meaning
  3. Referencing the manual regularly during play to look up information, even when playing with experienced players
  4. Cross-referencing information in three or more locations to evaluate a single action
  5. Regular use of a table that has more than 3 columns and rows.

Cause of Problem:

There are two primary causes of information obscurity induced AP, the first is physical and the second is behavioural.

Problem 1: Memory

This particular problem may be near and dear to my heart as an overtired parent of 4, but it is well known and has been well established that human memory is limited and fallible. The most famous academic result is the Rule of Seven by George Miller [Miller, G. A. (1956). “The magical number seven, plus or minus two: Some limits on our capacity for processing information” http://psychclassics.yorku.ca/Miller/] which places bounds on the typical performance of short-term working memory. Further studies have provided much better nuance and analysis, but for our purposes the general ‘rule of thumb’ for memory is more than sufficient.

The physical reason that information obscurity causes AP is simply memory. Every time a player is required to lookup a piece of information, consult another location to reference a value, or lookup a number in a table is another component that they need to store in their short-term working memory.

Consider a situation where a player is attempting to make a decision by evaluating the outcome of two different courses of action. If a player needs to reference three different sources to evaluate the outcome of a single action, we can reasonably expect the average person to be able to judge two outcomes and determine which one they prefer.

However, as soon as we consider a situation where a player needs to evaluate three options, we start getting into trouble. When evaluating an outcome a player needs to remember which choice they are processing, relevant information from two other locations, plus an assessment as to how ‘good’ that choice is. Even if we assume the other two choices have already been processed, that adds an additional two choices plus their assessment to remember – giving us a total of 8 pieces of information to remember at the same time.

Now this may not seem like much, the example is pretty generic, so let’s try something a little more concrete.

Zombie Chase

Consider that we are playing a zombie game, and we have 3 characters that each need to complete a task and we are trying to decide which moves first. Movement is determined by a table in the manual that tells us how far each character can move depending on who they are and how much equipment they have.

First we need to look at the character card to get the name, we have: Alice, Bob, and Carol.

Then we need to look at the board to determine that character’s position, both in terms of distance from the zombies, but also the distance from their objective. Alice is 3 spaces from the Zombies, Bob is 5, and Carol is 8. Alice is also 2 spaces from her objective, Bob is 10, and Carol is 8.

Searching the rule book tells us that Alice’s speed is 4, Bob’s is 3, Carol’s is 3, and the Zombies move 2. We can conclude that if the characters don’t move, Alice will be attacked in 2 turns, Bob in 3, and Carol in 4. Alice can reach her objective in 1 turn, Bob can reach his in 4, and Carol in 3.

Who moves first, and in what order do they need to move to avoid the Zombies? Confused yet? What if we also have to consider that Bob and Carol’s objectives will protect them from the Zombies if they reach it, but Alice’s won’t – she will have to continue moving to keep ahead of them?

The point is, the more sources of information that need to be considered to evaluate a course of action, the greater the chance that a player will lose track of some of that information and need to spend time revisiting a previously considered choice. Retreading ground is a form of analysis paralysis because it increases the amount of time and effort required to form a conclusion, it increases the difficulty in evaluating options and reaching a decision, and it causes frustration and delay in being able to do so.

Problem 2: Causal Calculus

The basic (simplified) idea of causal calculus is the ability of a person to infer that if they take a specific action, they will accomplish a particular goal. The strength of this inference depends on the distance between them, how obscure or indirect the path is, and any confounding factors that may interfere.

The harder it is for a player to perceive the effect their choice of actions will have on achieving their goal, the more difficult it is to choose between them.

This problem also comes with a corollary – given two equal choices, the one that is easier to perceive how it will lead to a goal appears to be more valuable than the one that is harder to perceive.

Tile Taking

Let’s consider a game where you have two stacks of tiles that are randomly coloured red, yellow, green, and black and you can see all the tiles in both stacks. You and your opponent take turns drawing a tile from the top of either stack. A Red tile is worth 1 point, Yellow is worth 2, Green is worth 3, and Black is worth -1. Your goal is to have the most points at the end.

The strategy for playing this game is pretty simple. Take higher value tiles, and force your opponent to choose between lower value ones.

Now consider the game where you only get points if you have the most of a specific colour of tile at the end of the game. You have to determine whether or not another Green tile is worthwhile or overkill if it costs you the majority in Yellow and Red.

Now consider the situation where you are stacking the tiles you draw in the order that you draw them, and there are 5 tile patterns on cards that are stacked and hidden from view. You know what the patterns are, but not the order that they are in and at the end of the game the top card is worth 5 points, the next 4, and so on. The player with the most instances of each pattern gets the points.

Each layer of abstraction, indirection, or hidden information that prevents a player from translating an decision into a desired outcome is a source of analysis paralysis. The more logical steps a player needs to make in their head, the greater number of variables they must consider, the harder it is to evaluate whether or not an action is beneficial or detrimental to their objective.

Solutions:

If your players are struggling with information obscurity, there are a few simple ways to make their decision-making process easier.

Consolidation:

If players are required to regularly look in multiple places for information, try combining those pieces of information into a single place – even if it gets duplicated as a result. Identify the key choices that are being made (are they choosing a card from their hand? a stack on the board? an action from a list?) and include as much of the necessary information as possible on the component. If players are choosing from a list of options, consider creating components to represent each option or a reference card with the relevant information as an aid.

In the Zombie Chase example, the board already provides a great visualization of respective positions of players and zombies. Adding Speed/Movement information to the character cards would eliminate looking up those values in the rulebook. Other ideas might include using standees that match character artwork, and creating a card or token of each character's objective - together these link the character, their objective, and provide visual cues as to their relative locations on the board.

Clarification:

Grant Rodiek of Hyperbole Games wrote an article on his journey through the process of refining the graphic design on his cards that is worth a read. (Link: http://hyperbolegames.com/blog/re-shaping-shapely-cards). If players are searching to clarify the meaning of a piece of text or symbology, try modifying it. Add clarifying text, rework your symbology and graphic design to provide a more consistent and intuitive interpretation. Avoid using symbols in a way that run counter to your player’s experience (ie. Green for Stop, Red for Go). Try to pull symbols, colours, and meaning from everyday life that people are already familiar with to reduce the burden of learning that your players require to play.

In the Zombie Chase example, using a foot or a traffic symbol "Walk" symbol for speed. The Tile Taking example already uses a familiar scale of Red/Yellow/Green which, although not colour-blindness friendly, is understood that Greed = good (most points), and Red = bad (fewest points).

Revelation:

Hidden information is often necessary in a game, but try to avoid hiding information that is required for a player to evaluate the choices they need to make during the game. Provide a clear, simple, and visible way for players to assess the relative value of the decisions they are making. If this is proving to be a significant challenge for you players, you may need to simplify your game’s scoring model, change hidden objectives into visible objectives, make hidden scores both visible and countable by other players.

Everyone has a story of the Monopoly player who hides a few 500’s beneath the table so that other player’s think they are doing worse than they actually are. Rules that state a player must answer how much money they have when asked is a way to increase information visibility to make player decisions easier.

In the Tile Taking example, playing with the pattern objectives face-up instead of face-down changes the game significantly. With the objectives being face-down, each player needs to consider all possible tile patterns, the number of each that they and their opponent has, and try to decide which patterns they want to try and have more of than their opponent... all without knowing with certainty which patterns are good! The difficulty of trying to consider this many variables would make it impossible to choose effectively, and a player prone to AP may be be paralyzed with indecision. By revealing the objectives, each player can conclusively determine at any time who is leading and which tile selection is optimal.

A variation that would fall somewhere in the middle, could be to play with pattern objective face-up, but have each player hide their tile stacks from each other.

Many games have a wide design space available for modifying information visibility and adjusting the perception of the path between action and result. However doing so can have an non-proportional impact on the nature of the game. A bluffing game without hidden information is not a bluffing game. Even minor adjustments can drastically affect the way players approach it.

Next Week

If you have any comments, suggestions, or examples that you would like to share about this week’s topic, please tell us about them in the comments.

Next week we will be looking at the Paradox of Choice.