14
Sep

Spoiled for choice: The psychology of choice overload in games, and how to avoid it

TL;DR version:

  • Too much choice can lead to dissatisfaction with or avoidance of choices
  • Excessive choice should be particularly avoided with game and genre newcomers
  • Choice overload seems to start kicking in when there are more than 7 options to pick from
  • Experienced players may not suffer from choice overload, and having more choice could even be beneficial for them
  • Avoid choice overload by reducing the options available, or by otherwise easing the decision making process for newcomers
  • Dividing options into categories can also help improve choice satisfaction among players unfamiliar with the choice domain

Games are full of decisions. Players routinely choose between sets of options, such as items in a store, heroes in a MOBA, dialogue branches in an RPG, or tactics for taking down an outpost in an open world. The processes of decision making are a core subject of cognitive psychology, and there’s a wealth of literature relevant to those wishing to present choices to players in ways that support the designers’ intent. This post will focus in on a particular avenue of research around the psychological implications of giving players many options to pick from, and consider the instances in which it can be beneficial (or not) to give players an abundance of choice.

Overloading new players with too much choice

It’s been argued that the more options you can give people to choose from, the better, since more choice means more opportunities to satisfy a range of individual preferences (e.g., Mas-Colell, Whinston, & Green, 1995). This perspective also has commonsense value – I mean, why wouldn’t you want more options to choose from? Research, however, tells a different story: one really can be spoiled for choice – something referred to in the literature as choice overload (e.g., Iyengar & Lepper, 2000), or the excess choice effect (e.g., Hafner, White, & Handley, 2016).

Iyengar and Lepper (2000) first demonstrated choice overload with an experiment in which they observed supermarket customers as they perused displays of either 6 or 24 varieties of jam. They found that customers exposed to only 6 jams were much more likely to make a purchase than those who saw all 24 jams. They concluded that a set of 6 options was more psychologically manageable, whereas a choice between 24 jams was overwhelming, leading people to defer a decision altogether. These shoppers had other options. They didn’t need to make a choice, so when confronted with an array of 24 subtly different styles of a non-essential food, they just moved on and went about finding other groceries, and making decisions that likely required less effort. How might these observations in a grocery store apply to players at home, making (or not making) choices in games?

Most simply, the excess choice effect suggests that if you give your players too many options to choose from, you might be discouraging them from picking one at all. If a player goes into an in-game store and finds there are 30 types of armour to choose from, they could defer this choice by just moving on and progressing with the game, or by quitting the game entirely. Avoiding a choice by quitting the game is perhaps most likely to occur during someone’s early experiences, when they have little investment in the game. Imagine you’ve just downloaded a mobile free-to-play game and, right off the bat, you’re confronted with a mandatory choice between 20 playable characters, all with different stat profiles. Consider how easy it would be to just put your phone back in your pocket and go about your day. Conversely, if presented with the same choice after investing some time into the game, you might feel more motivated to make the choice and keep playing. Indeed, it’s been shown elsewhere that people are more likely to keep investing in something if they’ve already sunk resources into it (Arkes & Blumer, 1985). On the other side of the coin, premium games paid for up-front may be less susceptible to players immediately quitting when confronted with choice overload, since they’ll have already sunk money into buying the game itself. However, in these circumstances choice deference could take another form, with players simply disengaging with parts of the game that give them excessive choice, but which might otherwise enrich their experience. Take Fallout 4, for example. If a new player spends several hours questing and looting, then checks out her crafting bench for the first time and finds she’s got over 40 items available for crafting, it would be very easy for her to abandon a decision on which item to craft and instead pursue other activities. While getting distracted by the multitude of things to do might be a core part of the Fallout experience, it’s unlikely that its designers actually intend for their players to avoid a part of the game due to choice overload. So, following this line of reasoning, designers should be especially careful with giving excessive choice to players during their early experiences with the game. Further on we’ll see more reasons why presenting large numbers of options is more risky with new, rather than experienced, players.

 

New Fallout 4 players may be overwhelmed by the crafting choices available to them early in the game, and therefore defer crafting in favour of other activities [Image credit: deadgames.org | Game: Fallout 4]

 

Besides deferring choices, Iyengar and Lepper (2000) also looked at what happens when people don’t defer choices, from both large and small option sets. Interestingly, it was found that people were less satisfied with, and less committed to choices made from excessively large sets of options. It seems, therefore, that while having more choice might sound appealing, it can have potentially unintended psychological consequences beyond simple choice deferral. Here designers must think carefully about how they want their players to feel about a choice, after it’s been made. If someone makes a relatively inconsequential decision from a set of options they’ll be able to try all of eventually (e.g., cars in a racing game, or special abilities in an action game), mild feelings of dissatisfaction or a lack of commitment might not be a big problem. In fact, these feelings could potentially enrich the experience by enhancing motivation to keep playing and unlock all of the options. However, if the decision is final, such as a choice of character class, or between dialogue options, or game endings, lingering feelings of dissatisfaction and lack of commitment could damage the player experience – especially so if the player only becomes aware of the decision’s finality retrospectively. The implication here is that special care should be taken with all decisions between large numbers of simultaneous options, and particularly when those decisions are final.

It’s important to think here about what constitutes excessive choice. Why should smaller sets of options be more psychologically manageable, and what’s the threshold for choice overload? Interestingly, Chernev (2003b) has noted that the findings in the literature are consistent with previous research showing that people perform poorly at tasks that require them to hold many more than 7 items in working memory (Miller, 1956). So, while a boundary between manageable and unmanageable numbers of options has not been systematically investigated, a useful rule of thumb to keep in mind is that the excess choice effect tends to manifest when the set of options is larger than 7.

 

What counts as excessive choice for newcomers might be fine for veterans

The excess choice effect has been examined in more detail and it’s been found to manifest differently between people who are experienced and those who are inexperienced in the choice domain. More specifically, it’s been shown that people who have expertise in the choice domain (e.g., a veteran player of a game or genre) – who already have a clear set of attribute preferences and values – are more likely than domain-novices to make, rather than defer, choices from large option sets (Chernev, 2003a). Additionally, people who’ve already accrued experience, and developed a preference for a particular set of attributes also show greater preference for the options they pick from large sets (Chernev, 2003b). Consider an expert League of Legends player trying out Dota 2 for the first time and being faced with a colossal set of 100+ heroes to pick from. He’s never seen Dota 2’s heroes before, but knows all about League’s champions and their attributes and, if asked, is able to clearly articulate his stat preferences, such as high attack damage, coupled with medium defence and low ability power. Since League of Legends and Dota 2 are in many senses very similar, he can simply compare each Dota 2 hero with his pre-existing preferences from League, and select one that matches. Compared to a MOBA novice, the domain expert is able to rely on this decision making shortcut when choosing from a large set of options, thereby making it easier to actually make a decision, and more likely that he’ll find something that matches his preferences. Interestingly, Chernev (2003b) also showed that domain experts, with clear attribute preferences, actually prefer large over small option sets, presumably because they afford them more opportunities for satisfying their ideal attributes.

 

An experienced League of Legends player might find it easier than a MOBA newcomer to make a Hero choice when starting Dota 2 for the first time, since they’re likely to have a preexisting set of attribute preferences [Image credit: PCInvasion | Game: Dota 2]

 

Chernev’s (2003a, 2003b) research suggests that large selections of options can be a good thing for experienced players, but, as highlighted earlier, new players are likely to struggle with such choices, which could contribute to them giving up on the game altogether. Consequently, if you’re designing choices that will require players to pick from large option sets, think carefully about where, when, and how you present these choices. Ideally, do not incorporate such decisions into the early stages of the game, when players are unlikely to have developed a clear set of attribute preferences that would enable them to benefit from such a large range of alternatives.

Occasionally, you may have no alternative but to include choices from large option sets in the early game. Let’s say you’re working under the design constraints of an existing game or an inflexible genre standard. In this case it’s important to acknowledge that there’s a good chance that the choice might negatively impact the early experience of your game, and may lose you some players. However, there are still measures you can take. Essentially, anything that eases the decision process for new players should help. You could, for example, reduce the number of available items temporarily, introduce the attributes to help players form an idea of their ideal set, narrow down the options that are most suited to novices, or provide a mechanism that facilitates easy comparisons between options. League addresses this issue by making a selection of 10 champions free to try each week. New players’ choice set is therefore practically reduced to a much more manageable number. Epic’s MOBA, Paragon, similarly focuses its set of heroes available to new players with a small set of recommended starter heroes. This is particularly good choice design for new players since it makes their decision easier by both reducing the number of options, and presenting those options that are most broadly appropriate for newcomers.

 


League of Legends narrows down the set of Champions for new players with a free Champion rotation, thereby easing the decision making process and avoiding choice overload. [Game: League of Legends]

 


Paragon makes it easier for novice players to decide on a hero by recommending a small set of starter heroes most appropriate to newcomers. [Game: Paragon]

 

Categorisation of options has also been shown to help improve choice satisfaction (Mogilner, Rudnick, & Iyengar, 2008). Specifically, it has been demonstrated that dividing up a set of options under category labels not only helps convey important information about the types of distinctions between options, but can also imply a greater variety of options, and therefore lead to a greater sense of self-determination and satisfaction after a choice is made (Mogilner et al., 2008). In fact, more categories tends to suggest more variety – especially to people unfamiliar with the choice domain – and can therefore lead to greater choice satisfaction (Mogilner et al., 2008). Consider, then, whether your options can be categorised in ways that support these effects. If you have a very large number of items to present to players, categorise them under informative labels, and don’t be afraid to use lots of categories if necessary (though it’s crucially important that these categories are also easy to understand and navigate). Thus, you may be able to offset some of the dissatisfaction that comes from excessive choice by leveraging the increase in satisfaction that comes from categorising options.

 

Conclusion

People find it hard to make choices from large sets of options, and when they make these choices they tend to be less satisfied with and less committed to their choice. It’s been argued that this is likely to be especially problematic for free-to-play games, where new players might feel less initial psychological investment in the game, which could in turn make it easier for them to avoid difficult choices by quitting the game altogether. Importantly, choice overload appears only to be a problem for domain novices, that is, people who don’t know what it is they should be looking for when comparing options, such as game or genre newcomers. While the number of choices that count as excessive for new players is likely a function of many factors, a useful rule of thumb is to avoid giving them many more than 7 things to pick from early in their experience with the game, especially if the choice is final, since this is likely to lead to choice deferral and/or dissatisfaction. If there is no alternative but to present new players with a large choice set, however, it’s suggested that measures be taken to ease the decision making process.

 

Written by Bob Tilford