Categories
psychology

real fake emotions

LA Noire is the new game from Rockstar Games, the notorious publisher of Grand Theft Auto. This guardian article describes it as “a new era for interactive entertainment.”, where the gameplay is not about is not about hand-eye coordination but about emotional perception, being able to judge body language and facial “tells”.

The thing is, what will the “true” meaning of the facial expressions in the game be based on? Will the correct judgements be based on the game-designers’ intepretation of what different facial expressions mean? If so, how can we trust that they have the correct intepretation? It isn’t straightforward to read meaning from expressions. Even people who think they are experts at it can be wrong, and many of the clues popularity associated with deception, such as gaze aversion, don’t truly help you tell truth from lies (see the work of Aldert Vrij, at Portsmouth).

It might be that we end up playing a game where you learn to intepret what Rockstar games believes about what people look like when they lie, rather than practice any real world emotional perception

9 replies on “real fake emotions”

Hello Tom,

this is kind of interesting. Apart from the obvious line they are walking across the uncanny valley, I think it will ultimately fall into fairly standard gameplay.

Rather than typical muscle memory that makes people able to beat Super Mario Bros in four minutes and perfect sports sims, this will develop a sort of emotional memory. As you say, you will recognise the things that signal to you that that is what the game believes, not what you necessarily believe. In-game signifiers are more often dominant than the actual form. In Portal, for example, there are only certain textures that you can create Portals on; first impulse is to find those textures, then decipher the puzzle. That’ll be the same with developed behaviours.

Nice post.

I’m interested to know what you think novels can teach us about emotional perception.

To defend Team Bondi (the developers of the game) a little, they haven’t used animators to program the face and body movements of the characters. Actors were recorded using a motion-capture system, along with their voices. This data will fill 3 discs on the xbox 360 console, making the game larger in data terms than 99% of 360 games.
2nd: You’re not playing in a real-life world, you’re playing in an explicit film noir setting, where gaze aversion really does mean you are hiding something. These are the specific emotional cues for this specific noir environment. They probably aren’t realistic as you say, because that would provide awful gameplay characteristics. You aren’t supposed to be learning how to be a real detective or that guy from ‘Lie To Me’ – you are supposed to feel like a film noir tough guy detective from 1940s LA.

@peter – i take your second point completely, about this being a genre thing. On the first – them using motion capture with actors – I think this will mean that we still get in the game what actors (edited by Rockstar) *think* deception looks like, rather than real deception

@hubert – i don’t have very developed opinions about novels, but in as much as we enjoy novels for the simulation they give of reality they are a source of experience (and the evidence is that older people are ’emotionally’ smarter, on average, at least in the ways that emotional smartness can be measured). That said, they will tend to give us experience which is in line with our preference for novels, so there’s a sort of “self-bias” rather than “world-bias” in the way they are likely to educate us. Speaking personally, novels I’ve read can act as touchstones for particular situations or emotions – which means you can recognise them or communicate about them more easily

I think this is a very interesting development in game play (but I would say that given the @mr_fante work :)).

To what extent do you think that the medium here is critical to the interpretation? I presume that expressions have been used in game play for some time just not as explicitly, and people have got used to some general emotional cues.

Will people playing believe that any analysis of the players’ faces will translate into other settings, e.g. when meeting friends, or buying from a shop? Seems to me that there are contextual differences that mean generic cues (like the one’s used in this game) are seem almost as parodies.

With this in mind I’d be interested to know if the pattern recognition in the cues people give are consistent across cultures, peer groups (friend, family, colleagues). This article (albeit a lay article) points to emotional cues being culturally distinct:
http://www.msnbc.msn.com/id/18614086/ns/health-livescience/t/happy-or-sad-emotional-cues-vary-culture/

Tom, reviews are starting to come out, and I am becoming more likely to purchase the game. I wonder if there is a resource you can point me to, or if you can just tell us a few real-life deception cues. [I found one article on Vrij’s website (http://www.port.ac.uk/departments/academic/psychology/staff/downloads/filetodownload,89132,en.pdf) but he doesn’t seem to have too many articles on his website.]
Then, once we get some more info about the game, we’ll be in a better place to judge how real to life the deception cues portrayed are.

“The game is all about reading body language. Did you study the subject during development?
BM: Well, Jeronimo plays a lot of blackjack, so he can probably explain all about tells!

JB: Obviously we did a lot of research, read all the books, but you’ve got to remember, we’re using actors, and they are inherently lying all the time. We directed their performances toward what people are used to seeing on film or on TV, when you watch liars on a cop show, versus really good liars in real life.

BM: The other side of that is, actors will say they’re never lying because they’re truly in the moment. We weren’t consciously looking for micro-tells, as you won’t be seeing this game on 14-foot-high screen like you are in the cinema. If you say to an actor, “on this line I really want you to lie”, it’s amazing how that comes across.

JB: And all the time you’re investigating crime scenes, so you have to keep data in your head about what’s happening, what’s going on. Part of the game is to catch people, to have the evidence available so that you can accuse them when they’re lying. It’s about keeping track of the story, the whole performance, rather than necessarily looking for an “answer”.”

The Guardian broke the embargo on reviews yesterday, I guess, but they have this interview up which speaks very precisely to the issue we were discussing. (http://www.guardian.co.uk/technology/gamesblog/2011/may/11/la-noire-exclusive-interview)

Leave a Reply

Your email address will not be published. Required fields are marked *