What this video is telling me is that our bias not only expects to reap immediate rewards from our decisions in games, but also not NOT expect eventual rewards. Now, this isn't always true--in Ninja Gaiden, I knew that buying the wooden sword and continually upgrading it would be a good thing--but it seems for most cases true. Imagine whether you're asked to save a burning family in a game, or let them die. Think about this. I bet your expected outcome is something along the lines of save them and they'll give you money, let them die and nothing happens. But, what if saving them gives you nothing, since they have nothing, but ends up bringing you riches later in the game when they can pay you better. Now, imagine you let them die: you can steal all their stuff! You discover a shining family gem in the rubble which they'd otherwise, if you'd have saved them, sold and given you half the profit, but now you get it all.

The problem is that gamers never look ahead five moves. They want immediate gratification. They want to have fun now, not later in the game. So, not only are gamers (and probably designers) not thinking of eventual happiness and true consequence, but they don't want it. And if they did have try, we designers could always make up some crazy shit and make the gamer wrong.

So, if the gamer can verify--without thinking too much, 'cause I wouldn't expect him/her to--that choice A will be better than choice B, I bet the gamer is going to go with choice B just to see what happens then revert to a save and play with choice A.

If I've been right up till now, then the way to truly make the gamer make non-biased choices is to remove the ability to revert to his/her save and make A and B look the same in terms of rewards. Or, maybe the designers could make it so that there are no immediately foreseeable rewards and make the player realize it will take a while to see the outcome. The player will then look back and may decide not to revert to the save a couple hours back since so much has happened.

Quote:

Original post by Wavinator Coming from someone who used to save and restore just to walk dialog trees and saved and restored to watch all 4 endings of Deus Ex, I don't know that unwinding a choice makes me dissatisfied. What it allows me to do is explore the fullness of the game, and I like that I'm getting full and complete use of the product-- the whole experience. I doubt that I'm alone in this.

You know, whenever I read/watch/listen to a review of a game like fable that asks you take make choices, the reviewers always complain how at some point, you can completely revert to the other path thus making all your previous choices relevant. I want my choices to make long-term impacting changes. I don't want to see all four endings in one sitting, I want to work at them. But, even better, I'd like to play four entirely different stories!

To further my point, I'll admit to having emulated games in the past. Especially extremely old games where my hardware and/or cartridges no longer function properly. When I started, I used savestates (allows you to save the entire state of the game) to such an extent that I realized at one point I couldn't beat some of these games without them. Since then, I've had an appreciation for the patience it takes to get through hard sections without saving in the middle, and I've started accepting the decisions I make. This has increased my overall enjoyment in gaming.

Quote:

Original post by Argus2Most choices are going to be sub-optimal, and *many* players enjoy trying to find the optimal choices. In an FPS I can choose to use all of my ammunition shooting into the sky, which is a sub-optimal decision, but I can't really blame the game for allowing me to shoot the sky even though it would be impossible to win by doing so. Finding optimal choices may involve lots of deaths and failure, which some players will like, and other players will hate, so I'm agreed with Drakostar - it depends on the player.

Agreed. This is why I'd still be playing Ninja Gaiden if I still had my Xbox. The combat was so deep that I was still learning new techniques and optimal strategies a year or two after I bought it.

And, I used to play a racing game (although, I can't remember its name) and I found that the most fun I could have with it is to convince everyone in the room to drive backwards the whole race. It was slow as hell, sub-optimal, but our choice lead to our enjoyment. Racing games in general are a good example of how gamers deal with choice; having to drive without a single mistake through the entire race.

Quote:

Original post by Wai For example, FFXI is an MMORPG with classes. I started the game as game as Thief. When I was at high level, I played again as White Mage. I was happy with both roles and I was happy that I did not need to change my character. Does the knowledge of the research indicate that FFXI should NOT have let me replay as White Mage? Is the result of the research applicable in this context? If not, what context can it be applied to?

Hmm... but your choices still had long-running consequences, yes? Just because you were a high level Thief didn't make you instantly a high level White Mage when you switched, did it?


In the video, he notes that people in Harvard would rather be in the course where they'd have the ability to swap, even though statistics show they will be less happy with their choice. I seems some of us would rather take that course because we refuse to believe that consequences should matter in games.

It's also worth reiterating that for our experience to be conducive of synthetic happiness, we must first be disatisfied with the experience we have had. There's no analysis in having two good experiences of equal measure, such as replaying a game as two different characters and getting equal amounts of enjoyment or keeping two pictures one likes equally as much.

In the video, all of the test subjects had to give up something they liked in order to maintain a status quo. They had to lose something. In that respect, in order to enduce this synthetic happiness, the testers exposed their subjects to bad feelings; feelings of loss. BUT bad feelings can only lead to 'regret' (post-meditation unhapiness) if there is another option that is potentially as good or better.

In games, we usually see the simple choice of fight or die. We are relying on the game to satisfy our understanding that fighting is better than dying. If we are presented with a more complex choice (perhaps of taking the bow over the sword), a choice which may have no obvious repercussions, we are relying on the game to play out in a way that satisfies our choice. If it doesn't, then the game has deceived and betrayed us by offering us a 'death in disguise' scenario. The hooker in Hitman springs to mind here - damn her.

That same betrayal is communicated by forcing us to reconsider our choice. If we are later asked again if we want to keep our bow rather than swap it with the sword, again assuming there is no obvious consequence, then we will indeed feel as if we have made some sort of 'wrong' decision because we were relying on the game to make our decision right.

I think mere frustration and indignation is more a result of this than simple unhappiness. Frustration at the fact that our decision meant nothing in the long run. Others have already said it - decisions need to matter. If they don't, they'll become a destructive element of the design.

Re: Splinter of Chaos

In my example, the fact that I was high level was my justification that I was satisfied with the choices given hat those choices are not permanent. If I wasn't satisfied I would not have stayed playing till getting high level. (I also tried other classes but I didn't like them and did not continue.) Does the fact that I was happy contradict with the research? If not, what is the difference in the context?

I think that the difference is that in the FFXI example, I was allowed to choose BOTH. The choice to choose BOTH was not covered in the talk. I think that was the difference.

One setting where the research could apply is as follows:

Suppose the server has a problem, and each player must choose one character class they want to get, and the experience of the other class would be wiped out. So in that case, I would need to choose between keeping my role as Thief or my role as White Mage. A policy maker that knows the result of the research might be tempted to offer the players only one choice, knowing that this is the only choice that the players will be ultimately satisfied. However, this doesn't affect the fact that the majority of players would dislike THE SITUATION that they could not keep both.


I actually also do question some interpretation of the result of the experiments.

In the first photo experiment, the students did not know that they must choose one photo until they had developed two that they liked. In the second experiment, the students were asked before they develop those photos. So those students in fact have a semester worth of time to prepare for the fact that they could only keep one of them. They could possibly put in different effort to the photos. So you cannot infer that 66% of them will be dissatisfied.

A second question I have about the interpretation is that perhaps the difference in happiness is not the fact that the choice is reversible or irreversible, but whether the phrasing of the choice puts the observer in an appreciation or fault-finding mode. I mean that when the student was told that he could exchange the picture if they feel that they like the other picture more, the researcher puts the student in a 4-day, time-limited quest to find faults in their own photo, fault so great that would lead them to the "right" choice. This pressure did not happen for the students that were asked to simply choose between the two. I think this could make a drastic difference in the next 4 days before the deadline to swap.

I agree with MachiniMax's technique where if a true choice is presented, the game should amplify the positive effect of that choice (perhaps adjusting the contents accordingly so that the choice will be a good choice). So that the game can let the player play the style that they have just chosen. In another thread I was defining happiness this way:

Happiness: Participant receives an event with value matching the status of the participant.

So MachiniMax was referring to the case where the player declares a status by selecting the weapon (choose bow expecting ranged combat, choose sword expecting melee combat). So to deliever happiness, the game needs to create events matching the gameplay mode the player has decided to be suitable for his character.


In the experiment, the students who only had one chance to choose end up happy because the brain adopts the status that he is only entitled to one of the photos he developed. On the other hand, the students who were suggested that they could swap the photo enters a mode where they would not be satisfied untill they could unquestionably reject one of their photos. Since they initially liked both, this quest is futile. The student end up unhappy because they cannot find such unquestinable evidence to favor one photo over the other.


On the other hand I agree with the overestimation of the impact an event has to happiness. I also agree with Adam Smith's quote where this overestimation could be dangerous.

[Edited by - Wai on February 14, 2009 8:19:13 PM]

Yes, the game should match the player's expectations, but I wouldn't condone something like...say adaptive gameplay. I'm a purist when it comes to everyone playing the same game with the same rules and being as fair about it as possible.

Quote:

Original post by Wai In the first photo experiment, the students did not know that they must choose one photo until they had developed two that they liked. In the second experiment, the students were asked before they develop those photos. So those students in fact have a semester worth of time to prepare for the fact that they could only keep one of them. They could possibly put in different effort to the photos. So you cannot infer that 66% of them will be dissatisfied.

That's a really good point; I cannot. However, my point, I believe, is still valid. The player does not know until asked to make a decision that s/he will have to decide.

One possible reason why it might not be possible to induce a long run rise in happiness through choice restrictions with regards to games is that we are probably not able to remove a key choice : players choosing to play another game.

In the experiment, majority of the students chose the class that offered them a choice. In this context, the majority of the players might choose the game that offered them more choices - at the expense of their long term happiness.

[Edited by - Girsanov on February 15, 2009 8:45:25 AM]

Re: Splinter of Chaos
That is what I meant by true choice (it has no right or wrong answer), as oppose to a tactical decision. RPG sometimes allows true choice by letting you recruit different party members to support your role. So the game let you adjust the content so that you can play your role.

Re: Girsanov
I agree with you. I am further supporting your point in reminding you that, according to the first half of the talk, there is no long term expense in happiness. Paralyzed people are just as happy in the long term. So the first impression would kill the inflexible design, while any difference in happiness would disappear in 3 months.

Re: Opwiz
I actually really like TED, so thank you for bringing this up. It does feel really unsatisfying that the information appears unapplicable.

[Edited by - Wai on February 15, 2009 2:58:39 AM]

Interesting, but I'm not quite grasping how this can be directly applied to games. Maybe I'm missing some key detail, but arn't video games (when you boil them doen) just experience synthesizers not unlike the function the lecturer attributes to the frontal lobe?

I work in manufactureing engineering. Basicly means I'm tasked with generateing and improveing the process by which widgets are produced...I design and improve upon the widget assembly line, if you will. The least enjoyable aspect of my job involves multitasking and micromanagement. Most enjoyable is problem solving.

So if given the choice of playing StarCraft or God of War. My frontal lobe kicks in trying to analyse the more preferable of my two choices. Both games feature problem solving (development of strategy in StarCraft, and the more direct overcomeing obsticals in God of War). However the process of problem solving in both games is different so I choose to play the game that lacks the multitasking and micromanagement bits that I find unenjoyable.

But according to the findings of the lecturer in the TED video, my choice is likely wrong, and I'm synthesizing my happiness in playing God of War? Basicly tricking myself into enjoying it more than I prolly really do?

Maybe, but I have great doubts that is the case. There just isn't enough data in the situation I outlined above to warrent such a conclusion. (What else do/don't I enjoy?)

One of the tests in the TED video involved the choice of keeping one of two pictures the students took. In one situation the students were given some time to change thier mind, in the other they wern't. Then in another test the students were given the option of decideing beforehand which of the two situations they would most like to pursue.

He shows B(no chance to change mind = happy) is more happy than A(time to change mind = less happy) in the first test and shows that more people choose D(participation with chance to change mind) than C(particpation with no chance to change mind). But the lecturer concludes that student choice D is automaticly going to result in situation A...Thats jumping to a unsupportable conclusion. A = already have the photos and time to make a choice, D = hasn't even taken the phots yet to have time to choose between. A!=D

Quote:

Original post by MSW But according to the findings of the lecturer in the TED video, my choice is likely wrong, and I'm synthesizing my happiness in playing God of War? Basicly tricking myself into enjoying it more than I prolly really do?

Actually, I believe this video doesn't apply to "which game should I play" because there are no long-term consequences to that decision. And, you aren't going to be playing for long enough for this to affect your overall happiness.

Quote:

He shows B(no chance to change mind = happy) is more happy than A(time to change mind = less happy) in the first test and shows that more people choose D(participation with chance to change mind) than C(particpation with no chance to change mind). But the lecturer concludes that student choice D is automaticly going to result in situation A...Thats jumping to a unsupportable conclusion. A = already have the photos and time to make a choice, D = hasn't even taken the phots yet to have time to choose between. A!=D

I believe the point of that test was not to prove that A==D (although, yes, he incorrectly implied this), but to show that we would rather have more choice because we think that makes us happy, even thought it appears it does not. We don't know what makes us happy!

In the experiment of the patients with memory problem, we see a case where the existance of a choice makes the mind justisfy the choice after the fact by polarizing the individual's preference toward the selected and against the rejected.

According to this, the satisfaction for both group of students should be the same: both group should increasingly like the photo they initially picked and increasing reject the other. The difference must lie in the fact that one of the group could reverse the decision within 4 days.

According to the graph, once the decision period is over, synthetic happiness would make the student like the photo they chose again. The dissastisfaction effect is transient.

What happens if the number of days for that decision varies? Will the dissatisfaction drop just as fast or faster? What if the students can take forever to make that decision? Will they be dissastisfied forever? Is the dominating factor the presence of an option to swap, or the pressure to decide whether to swap? What if they could change the photo anytime, any number of times? This last scenar shows a case where the presence of the option is forever, but there is no pressure to decide. What would happen to these students?

For a game design, it seems that a comparison with the last scenario is most applicable. The original alternative in the study is not applicable because we don't normally have a situation where we give the player a deadline to revert a decision once.

Quote:

Original post by Splinter of Chaos Actually, I believe this video doesn't apply to "which game should I play" because there are no long-term consequences to that decision. And, you aren't going to be playing for long enough for this to affect your overall happiness.

And the degree of happiness in keeping a certain FREE Monet print or photograph illustraits a long-term consequence of choice? How can the test results even be considered valid without the subjects assestment of the value of said choice?

If some subjects take thier Monet print home and hang it on a wall, while others just toss it in a corner forgetting about it; then how could such a test be considered a valid representation of long-term choice consequence? Even those that toss the print into the corner can still report they are honestly pleased with thier choice(in that moment), which supports the test result, but not the conclusion.


If I walk down the street and some hot girl steps up handing out some flier about some rock band playing at a club that night. I might flirt with her, I might even try to get her number and promise to attend the show...and in that moment I could honestly inform some tester that I'm happy.

But it doesn't mean I'll keep the flier or actualy go to the show, or even call the girl for a date. And without further data there really isn't a conclusion to be drawn from my honest statement of happiness in that moment.

Quote:

I believe the point of that test was not to prove that A==D (although, yes, he incorrectly implied this), but to show that we would rather have more choice because we think that makes us happy, even thought it appears it does not. We don't know what makes us happy!

If that is a supportable conclusion, then why resort to an implication?

Most of the tests depicted in the TED video are quite suspect, the photgraph one hints at a trick con artist perform.

Con Artist walks up to someone and hands them $50 inviteing them to play three card montie or some such. Then they take back $30 of the $50 given (subject left with $20) allowing them to take whats left and go or bet it to get the $30 back. Sense the subject had all $50 in hand at one point, they almost always try to win the $30 back...eventualy they loose the $20 given to them and start betting thier own money. Within minutes the con artist has his $50 back plus possably $100s of the subjects money.

Forget what psychologist call this (tactile reenforcement?) but if student had both photos in hand and had to give one up with the option of changing thier mind later. Well the test results might just reflect this phenomenon more than a true value of happiness.

Besides, through experience most of us develop at least some idea of what makes us happy.

[Edited by - MSW on February 15, 2009 8:40:26 PM]