Lighting in video games affect play maze performance (Knez & Niedenthal, 2008)

Yahtzee often complained about the current video games’ use of “realism” colours which are basically shades of grey and brown. This may be based on artistic taste, but this could relate to psychology when considering lighting in our environment, whether it’s in the real world or in the virtual world. It may seem trivial at first glance, but it has some real applications.

Abstract

As a means of extending the significance of findings in experimental psychology and nonvisual psychological lighting research to digital game research, the present study was designed to investigate the impact of warm (reddish) and cool (bluish) simulated illumination  in digital game worlds on game users’ affect and play performance. In line with some previous findings, we predicted that lighting in a digital game world might, as in the real world, differently influence the nonvisual psychological mechanisms of affect, which in turn might enhance or impair the players’ performance. It was shown that the players performed best and fastest in a game world lit with a warm (reddish) as compared to a cool (bluish) lighting. The former color of lighting also induced the highest level of pleasantness in game users. A regression analysis indicated tentatively that it was the level of pleasantness induced by the warm lighting that enhanced the players’ better performance in that digital game world. It was also shown that high- as opposed to medium- or low-skilled players engage almost 2.5 times more per week in game playing. Given their skill, they performed significantly faster and felt significantly calmer and more relaxed in doing so.

This reminds me of another study on how players in red team have a slightly better chance in winning than players in blue team.

Unfortunately, the experiment is a maze task and not an actual video game task. Nevertheless, this study provides a stepping stone into virtual lighting and how it may affect gameplay and game design as well.

Method

Participants: 38 (14 women and 24 men), average age 22 (age range 15 to 32, SD= 4.4). Their gaming skills are in a bell curve with most participants in the medium range. However, it is noted that it is a self-reported, actual skills not tested.

Measures

Game skills: self-reported measure where participants answer whether they are in the low-, medium- or high- skills range. They were also asked the hours per week of gaming and their favourite gaming genre.

Play performance: Time is used to measure on how fast they complete their task.

Affect: To measure their mood, they are asked how they feel in relation to each of the 48 adjectives on a 5-point scale.

Lighting evaluation: 10 items on a 5-point scale. They evaluated the maze’s lighting based on how ugly, glaring, cold, dim, intense, warm, bright, soft, annoying and beautiful.

Game used: Half-Life 2. They created three mazes in their experiment with different lighting conditions, greyish as a neutral lighting, bluish as cool lighting, reddish as warm lighting. The mazes are different, but they are of similar complexity and size. They have screenshots of the mazes, but they’re not in colour.

Procedure

Participants go through a tutorial of the game. They answered the gaming skills questionnaire. They play through all three mazes, the order they went through are counter-balanced (i.e. randomized). A minor note, the experiment runs two participants of the same gender (if possible) at the same time. After that, they completed the affect and lighting evaluation measures.

Results

They used ANOVAs for their analyses. They tested if age and gender as any effect.

Lighting evaluation: no significance difference in evaluations were found, except for the expected results in that warm lighting is warmer and cool lighting is colder.

Affect: Main effects were found. A statistical trend (i.e. close to statistical significance) was found that participants felt happier and gladder in warm mazes. A statistical significance was found in that they felt enthusiastic and peppy in warm mazes than in cool mazes. The other main effect is that high- and medium- skill participants felt more calm and relaxed than low -skill players regardless of lighting.

Performance: Two main effects are reported, first participants complete the warm maze the fastest. Second, high-skill participants complete the maze fastest regardless of lighting. Too bad there’s no interaction effect, so expert players aren’t affected by lighting colours in terms of performance. But it doesn’t mean in a general way.

The authors made some regressional analyses and found some links between performance and affect. Importantly, the “pleased” mood was significantly correlated with performance under warm lighting. So, they felt more satisfied after completing a warm maze than a cool maze. Quite interesting, I wonder how they would have felt if they failed.

Discussion

The authors noted that this experiment measured only time it took to complete a task and not the cognitive mechanisms related to maze performance.  The authors suggested that lighting affects our mood which in turn affects our performance, although the data is unclear about this relationship since the analyses are correlationary. In any case, this could be something for game designers, who may already about this, to consider when designing levels aesthetically or to drive a storyline. IMO, I find this study quite interesting and something to talk about over the watercooler.

Knez, I. & Niedenthal, S. (2008) Lighting in digital game worlds: Effects on affect and play performance. Cyberpsychology & Behavior, 11 (2), 129-137.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s