« Accelerating Change 2005 | Main | Magic »

Sep 23, 2005



I feel like crying when I see industry research with thousand-dollar pricetags, and think to myself how I'll never get at the good stuff cuz academics are too broke.


I haven't formed a strong opinion on this, but I think we are still in search of the emotional power of theater, movies, TV, etc.

Visuals can invoke emotions, but I’m leaning on a combination of movement and audio, especially music.

Music, as many asserts, is universal and the same musical patterns invoke the same emotions universally. And as audio can be processed more passively than visuals, and musical structures are more defined than movement in the popular media formats (TV, movies, but not the stage), music is an important area to study.

I need to read up on the impact of music on games and MMO, but one master at bring the music element to any production is http://members.shaw.ca/ykdb/index.html”>Yoko Kanno.

Gravity Co. recently recruited her to create 90 songs for Ragnarok Online 2. This is an equivalent of ~9 albums.

The press release states:
David Yoon, CEO of Gravity stated "It is a great pleasure to work with Kanno Yoko on Ragnarok Online 2. We continue to strive to provide gamers with excellent content and a great gaming experience. Compelling music can make a great title extraordinary. Kanno's background music has been effective in animations and games with strong fantasy elements. Therefore, her music which will express the medieval European fantasy world in Ragnarok Online 2 is expected to appeal not only to the game users but also to the general public."

In the same way John William’s score has lifted George Luca’s Star Wars, Yoko Kanno’s score has lifted http://www.animedream.com/music/cowboy-bebop-ost1/”>Cowboy Bebop, http://www.ex.org/1.5/35-esca3_cd.html”>Visions of Escaflowne among others.

Now going back to the first claim that caught Nate’s eye:
1.) MMOs are smack middle-of-the-pack of the game genres ranked in terms of emotional impact. At 32% MMOs rank above Racing games but below Sports games and way below First Person Shooters (52%) and Role Playing Games (78%).

That MMOs rank below Sports games and way below First Person Shooter seems odd and absurd. But, I can also see how players can get emotionally attached to their team and FPS players can get emotionally engaged by the visuals and the immersion.

So, if the ranking is accurate, then we’ll need to make more effort to make MMOs more emotionally engaging and music is an area that I am studying.

Listening to the Yoko Kanno’s composition to Visions of Escaflowne I am emotionally engage by the mix of uplifting choral, pop, love songs, stomps, and other invocative music. But when you http://www.escaflowneonline.com/music/guide.shtml”>connect the music to the scene used, you get a double sense of joy.

And in the mainstream media, the combination of visual and audio is best seen in today’s crop of TV shows for teens such as http://www.themusicedge.com/moxie/news/girlsrock/alexandra-patsavas-the-oc.shtml”>The O.C. and founded in the current era by John Hughes.

So I do like a good cry, but haven’t felt this emotion swell in a video game yet.

Not a practitioner, so I’ll just wait for that day or roll up my sleeve and create :)



$999 is a lot to pay for nothing.


"frustration and wanting to overcome it"

Why don't people say fiero when they mean it? People will learn the word if we start using it. Although, to be fair, 'frustration and wanting to overcome it' is really the counterpart to fiero which would be better characterised as 'triumph over adversity'. Still, the list doesn't include fiero and should - that's the emotional payoff, not the frustration, which is more of a side effect.

Just my 2 arbitrary units of currency... :)


Chris> fiero

I admit a double-take. From here:

"The challenge focuses attention and
rewards progress to create emotions such as Frustration and Fiero (an Italian word for
personal triumph)."



Nicole Lazzaro (who leads XEODesign, cited by Nate) is a leader in understanding the emotions of why we play games (and has led the popular usage of useful terms like fiero). She has some of the best (and only) actual observational data in this area. Worth reading.


Where's a good source of info to find out more about "fiero"? I wikipedia'd it and ended up with a car. I'm missing something here... thanxies in advance!

(Also how do I do bold and italics within comments?)


Ekman, P. Emotions Revealed. Time Books Henry Hold and Company, LLC, New York, USA 2003. ISBN 0-8050-7275-6.

That's the book Nicole references for emotions, including fiero.


saw this when it was released, was waiting for a beard to initiate it...

mmorpg's ultimately have an immersion issue. one might argue, mmorpg's have huge worlds with way more backstory and content than: a sports game, an fps, etc. however, in actualy gameplay, the game mechanics are not immersive.

in a traditional 1P RPG, the player isn't one of thousands who have done the same quest. in an mmorpg, combat is reduced to a simple real-time-turn-based "whose dps is teh win." if you die, you lose some time -- any money and xp lost is merely more time. in a sports game, you lose and that's it. game over. mmorpg's don't even have a game over. absolutely nothing of value (assuming time is "free") is lost -- or won...?

it's strange that in mmorpgs, where one can spend the most time customizing your avatar, multifold more than in any other game genre, players connect so much less to their experience emotionally. in stand-alone RPGs with absolutely no avatar customization, players feel the most emotional connection to the characters.

i want to call it the doll phenomenon. ultimately, i truly believe all mmorpgs revolve around a giant game of playing dress-up with dolls. with GI Joes and Barbies, kids play "let's cyber" or "let's PK" all the time. it's experimentation with pretend people. the poor anthropomorphic qualities of dolls allows kids to enact out situations while clearly knowing that the results aren't real. "i didn't do it. megatron did."

anyone out there intending to make more worldy, immersive mmorpgs will have to deal with this hurdle. it maybe true that inevitably, people continue to subscribe to mmorpgs over time because of the connections they've made to other people, and not to the game itself. however, this would probably be equal across all persistent games. mmorpgs IMO are still a juvenile art/commercial form, because despite all their $20 million bells and whistles, they still have yet to evoke the Aerith response...


I've actually gotten an emotional rise out of City of Heroes a couple of times. But the problem is, the rise comes from a plot twist - and when with a second character you're essentially replaying the twist, that moment of shock, resolve, amusement, whatever, is completely gone.

Reading a spoiler site for quest details is another way to totally circumvent any emotional impact that was supposed to show up.

Though I do agree that music has something to do with it - CoH is definitely _very_ shy in that department, to the point where I just play random MP3s and turn down the scarce in-game stuff. A _big_ part of the Aerith response was her theme, perfectly timed to complement the FMV.

If they can find a way to put good dynamic music into a MMO, that would help a lot with emotional appeal.



I wish people would stop using Aeris like some kind of gold standard in emotional response to games. Why people praise such poor writing and one-dimensional character development escapes me; she wouldn't cut the mustard in a movie, so why do call it a great event in gaming? She's just a basic naive/innocent white mage archetype that the Japanese go crazy over, introduced only to be summarily executed a short while later. How people developed a bond to her in that period of time (with no character development on her part) completely escapes me. As far as I'm concerned, Aeris was merely a easy cop-out for the writers to make Cloud more moody and angsty, and killing her without developing her character cheapens, not deepens, her character.


hikaru> Aerith response


Here is some context.


Interesting point from the above citation with implications for "replayability" and mmmorpgs:

his game was designed to play like a movie, so the scripts were really good.


Good scripts are of course important to emotional response, although I will contest that any Final Fantasy game had a good script.

I think it's also worthwhile to seperate certain categories of emotional response - not as in, happy/sad/angry/etc, but rather the method through which they are inspired. The sense of fiero from besting a competitor in Street Fighter is generated differently than response to actual gameplay scripting/storytelling.


hikaru> the doll phenomenon

Interesting idea. For discussion, let me play with it in this way. Is it true that players for the most part are experimenting? And is that cynical (with respect to the world)?

I'd contrast this with a view that speaks more towards the conflation of socialization and play. I sort of hint at it in comment in this thread:


Which is more true: are the participants earnest, or are they cynical? Or it is it neither?


"mmorpgs IMO are still a juvenile art/commercial form, because despite all their $20 million bells and whistles, they still have yet to evoke the Aerith response..."

I agree. There are a variety of factors which in my mind work against immersion/emotional investment in MMOG's. One of the things which Years ago some fans of writer Harlan Ellison who worked for Lucas Arts got him a copy of the "Empire Strikes Back" video game to review. He was horrified that the game went on and on and on. Where was the ending?

Now of course it's tough to find a single player game that doesn't "wrap up" in some fashion. Even a modern side scrolling arcade style shooter typically brings a player to a final level, with a final boss. MMOG's in contrast are stuck where single player games were decades ago--they go and and on and on. When I tried to introduce some game playing friends to MMOG's that's one of the things that they found surprising and disappointing--you help out NPC X by wiping out Monster Y and fifteen minutes later NPC X still needs help and Monster Y is back. Can there really be a narrative when nothing changes?


James O wrote:

"I wish people would stop using Aeris like some kind of gold standard in emotional response to games. Why people praise such poor writing and one-dimensional character development escapes me; she wouldn't cut the mustard in a movie, so why do call it a great event in gaming?"


"Good scripts are of course important to emotional response, although I will contest that any Final Fantasy game had a good script."

Just goes to show how much tastes differ. I think Final Fantasy VII is the only game I've ever played which evoked in me an emotional response comparable to that from a novel or movie. Yes, a lot of the writing was juvenile but at the same time a lot of what I saw impressed me. I credit the writers for allowing the ending of the game to be genuinely ambiguous. The final animated sequence doesn't even attempt to wrap things up in anything like a concrete fashion and for that reason, among others, struck me at least as simultaneously evocative and beautiful.

As for Aeris what happens to her is at least as important as any character development she goes through for the purposes of advancing the story. There is an element of sacrifice in the way that she dies. Clearly she wouldn't mind being able to hang around Cloud for awhile and yet she gives that up for the greater good.

There is also the manner in which the other characters react to her death. When was the last time that you played a game where someone actually grieved? I can only think of a handful. Considering the role that violence and death play in video games that's probably symbolic of something.


Before despairing over the score for MMO's consumers of research statistics should consider the methodology that supports the claim. Surveys are not the best method to research emotion and games because players can't always tell you what they are feeling.

First consider the validity and reliability of the study methodology. A test is valid if it measures what it claims to measure, like pointing a telescope at the right nebula. A test is reliable if it does it's job repeated like a good set of tires. Every time I push the button the telescope turns on and collects data from what's in front of it. A valid test points the scope in the right direction. A reliable test performs the same function every time. A study can meet both, one, or neither of these qualifications. Using large sample sizes improves the reliability of the results, but increasing sample size does little to increase validity. Even with thousands of users, the test could still point to the wrong star.

A quick overview on validity and reliability: http://www.georgetown.edu/departments/psychology/researchmethods/researchanddesign/validityandreliability.htm

I have not paid to read the full paper, but online surveys generally suffer from validity bias (in addition to self-selection bias covered elsewhere). To understand why we play games requires first determining what kind of telescope(s) to build and choosing a direction to clearly observe the emotions in these player experiences.

Online surveys map frequency shifts in attitudes for understood phenomena. They work when the right kind of telescope points in the right direction and the researcher wants to understand what's there today versus yesterday. A large sample size gives online surveys statistical reliability. However, the study may look at the moon when it really should study the sun; and it may completely miss the new comet.

Because they collect frequency of predetermined responses surveys require writing the answers in advance. For example, a question on racing games from the sample survey on this researcher's site constrain the player's answer to 3 responses speed, graphics, or style of car. If a player prefers developing a character overtime (RPG/Career Mode) or the feeling of freedom to drive anywhere (GTA) then this survey misses that data completely. Worse it forces the user to fib and say that speed is the most important thing. This obscures the true percentage of players that really choose this as their answer.

Another source of validity bias shows up in the "emotion" ranking of genres. It probably (and those who have read the full paper jump in here) came from a question(s) asked like this:

"Which of the following types of game are the most emotionally powerful?"

What this tells us is how participants when asked this type of question rank these game genres. It is not a valid method for comparing the = experiences = of playing each type of game.

What does the researcher mean by "emotionally powerful"? Should this cover amplitude, frequency, and variety of emotional moments? Or simply "memorability"? The researcher was "surprised" that RPGs scored better than MMO's. Hopefully the survey used example games and not these genres titles - many gamers label FPSs as RPGs, and many MMO's truly are RPGs. Such ambiguities need to be clarified when designing survey questions.

Surveys also require the player to retain and filter through a great deal of data to come up with such non-answers as "RPG games are the most emotional." In my experience over the past 15 years many many many of those who curse the controls during play later say on a survey that they found the game easy to use. These players either deny the emotions (fib a little), or genuinely forget that they experienced any frustration.

Games are complex experiences that happen over time. Emotions are so fleeting they often occur before thought. Counting on players to answer written questions on emotions is a bit like hoping that players can describe the horses after they have left the barn. Once they're gone, they're gone, and new horses take their place. Even if the player remembered them go flying by, he or she may not have the words to describe the breed, the gait, or the markings. I have the same complaint of most focus groups. To get more reliable results (bigger sample with less researcher time) they ask players to examine themselves with their own telescopes. A more direct method of data collection is what we do at XEODesign: watch players play and measure their emotional responses through multiple modalities.

In my book, it's the researcher's job to observe the phenomena directly. For validity it's a professional who need to sift through the video tapes, transcripts, interviews, and surveys to get at what's going on during play. It's important to listen to what players have to say about their play experiences, AND it's also important to watch them at play. As those who have read XEODesign's research know, by doing just this we have found other ways games create emotions.

In short, to be truly surprised (as in the feeling from "a sudden, unexpected event" -- Paul Ekman) we must in a sense be unprepared. It's often better to leave the survey behind and go right to the source by watching players play. Otherwise, regrettably, we find exactly what we are looking for.

The comments to this entry are closed.