I almost think of it as this thing that happens when your cognitive capacity has been overwhelmed. You're trying to comprehend something that has a magnitude that you almost don't have enough neurological connections to deal with. And … your brain sort of overflows ...
 
In this summary Michael Frederickson, lead technical director at Pixar Animation is interviewed by Douglas Rushkoff from the Team Human podcast, to talk about this deeply human experience of awe.
​
It’s a conversation that spans the “awful” to the “awesome” and all those ambiguous spaces in-between. Rushkoff and Frederickson dig into questions of technology and storytelling, the narrative arc, and the evolutionary, even empathetic value of having our minds blown.

MAIN POINTS:
1) Awe - inspired by vast and novel scenes - has a narrative function.  The experience of awe encourages people to really take notice of the world around them
2) Awe, also creates a narrative rhythm.  It opens people up to new insights and experience.  Awe is different from spectacle, which can attract, but also overwhelm people.


How did you come upon the notion of awe and what sparked your fascination with it?
 
Awe ended up being the confluence of just about everything I found myself interested in over the years and it almost became like this inevitable thing to look into, especially after some of the work I did on Inside Out a couple of years ago. For me, it was precisely how difficult it is to articulate what the experience of awe feels like, what it is and why it exists that got me interested in it. I think my entry point initially was looking at why a director or any kind of storyteller might be trying to draw out a particular emotion in us. I felt like every time I would ask someone about awe, rather than saying how it made their body feel they'd talk about the experience that caused them to feel awe and I set out to discover what research has been done on this recently.  But also, if you try to apply that to storytelling when and why does it make sense to try to make people feel awe?
 
Right, so if we think back there are these moments in movies where you come upon an awesome sight, like the world of the Avatar people or the roof-tops of a great city like London when Peter Pan flies out at us, or when we first see the giant panorama of a new city in The Game Of Thrones …so there are these moments of awe that storytellers tend to put in to a story?
 
Yeah. And I was curious why we were doing that? Talking to different artists over the years and myself as an artist, I've found people who have different levels of comfort with how much they want to talk about the conscious act of producing art, like some people seem to be made really uncomfortable by talking about having almost mechanical tricks to producing something creative.
 
Because it feels more contrived if you've manufactured it?
 
Yeah, a lot of artists seem to be concerned with creative spontaneity and regard it as this thing that's hard to articulate, and it's almost a little bit ephemeral and hard to grasp. But it seems to me that some directors and storytellers are instinctively using awe at specific times in a story arc, and I’m curious why. The first step in that process for me was understanding what awe is as an emotion and that led to some research that defines it as something you experience when you encounter something that's both vast and novel. That value doesn't have to just be spatial vastness, like with the Grand Canyon or something like that. It could also be the vastness of somebody's talent, the vastness of time – something so vast it exceeds perceptual limit. Like I remember feeling it when I saw the Sagrada Familia in Barcelona, and it was both huge. And also, I was just overwhelmed by the amount of time that went into constructing that human effort. And then also novel, I mean, if you're seeing the Grand Canyon every day, you're probably not going to feel, awe every time.
 
One really helpful thing to do is ask what physiological change you go through when you're experiencing an emotion.  There's not a universally, cross cultural agreed upon expression that one makes when they're experiencing awe. But it tends to be head forward, mouth agape, eyes open, which points to the idea that you're trying to take in as much as possible, you're trying to taste and smell and see as much as you possibly can. Because the thing you're taking in is so enormous that you want to maximize your sensory input.  People describe feeling goose-bumps and there's actually some studies that say that awe can ultimately reduce inflammation, so If you're super inflamed, you know, go check out the Grand Canyon.
 
But the thing I got really interested in was recent work in evolutionary psychology which asks questions like, what might be the purpose of you, why has evolution selected us to continue having this very peculiar feeling and to what advantage is it to look up in the sky, for example?
 
I was also thinking about the root of the word awesome and it's, it's interesting because to be accurate, if something is full of awe, then we could say it is awful. Which is interesting, because it almost speaks to the history of the emotion. Researching emotions there's this term I've come across, which is the valence of the emotion, which refers to whether or not it's subjectively considered positive or negative. So, most people would describe happiness as having a positive valence and sadness at least initially, as a negative feeling. And most Westerners describe awe as being a positive emotion, which is super interesting because in the same way that comedy and horror exploit our experience of surprise, there’s a razor thin line between those two things. I think it really just comes down to the stakes, when awe is positive, you're experiencing something vast and novel, but something that is not mortally threatening to you.  Yet a long time ago, when the world didn’t always make sense awe was a more negative feeling
 
Despite the importance of having emotions that we can’t control, the study of their contribution to our humanity has traditionally been limited to the realm of poetry and philosophy in the arts.  But some of the earlier research pointed to the purpose of awe was to help us organize into groups. So maybe you'd feel awed by the power of a really strong leader, and that would make you feel humbled and part of something bigger, and then maybe that might help you organize and join a community that would then have a greater chance for survival.
 
When you use an example like that all I can think of is Hitler at the Nuremberg rallies, you know, shooting aircraft lights in the sky.
 
I think with any emotional response like that, you can either use it for good or evil. Yeah, I've often thought our proclivity to recognize lots of different things as faces is one of the reasons we can deal with animation and the abstraction of animation. It's because you can buy into that being a character, when it is not at all close to the real thing.  This reflects our need to accommodate something within our model of the world. This is one of the reasons that magic is entertaining, because you're kind of exploiting the fact that when something levitates you so desperately and instinctively need to square that up with your understanding that things don't float, and you're like, almost instinctively driven to explain how is that happening? And that's one of the things that sort of makes magic entertaining.
 
On a certain level, we're talking about a kind of emotional cognitive overload that throws me into a different state. Now, if it's a dinosaur popping out through the woods I'm going to go into fight or flight and overflow into terror. Whereas if my wife or my child encourages me to see a sunset in a way I've never seen it before, I start connecting things and I can ease into a state of awe.
 
So then there's the question of why. Take a very simple assumption that people might have about the world like what might be on the table at a romantic dinner: A candle, wine, two pairs of hands, a rose.
Most people assume that candles would be on the table. In one study, the researchers exposed three different groups to emotional stimuli. So they showed the awe group that Powers of Ten video from the 70s, where they sort of zoom out as far as possible into the cosmos and then down into the human skin. Another group saw somebody unexpectedly winning a gold medal. They were supposed to feel happiness or joy. And then another group, a neutral group just watched someone assemble a cinderblock wall.
They then told all the groups the same five minute story about a romantic dinner, where they very specifically did not describe candles on the table and gave them a distracting task after that, and then asked some follow up questions, one of which was, hey, were there candles on the table?  And after the study was done, they noticed that the awe group consistently did better than any other group at remembering that there were not candles on the table.  So one of the potential conclusions that came out of the study was that maybe by making you temporarily feel humbled, in relationship to something vast and huge, maybe awe has this temporary effect of stunning you and reminding you that you don't know everything, so you should look at information, new information for what it is and not apply a kind of bumper sticker reaction to it.  So, when things are going fine in your life, you might hear about a romantic dinner and go, Oh, I know all about romantic dinners, there's going to be candles on the table. But if maybe you've just been made to feel small and humbled you decide to really listen closely to the story, and to be alert to what’s actually there rather than apply pre-existing assumptions.
 
That's really interesting, because my senses are assembling movies all the time and a lot of the times we assemble stuff that isn't even there, we just assume that it is. So awe almost wipes the slate clean, it's almost like a rebirth.
 
Because you've been cognitively overwhelmed, you've effectively blown your mind, so maybe now your mind is kind of clear for a few minutes and can take in new information afresh.
 
I found this some this quote in an interview with Kubrick about The Shining, where he seemed almost a little bit stressed hoping that people didn't just go see his films and take away what they already believed based on their pre-existing models of the world. And he said, You know, I wonder how often people are fundamentally changed by a piece of art? He said, if you're 15 or 16 years old, you might be ready to do this. And I loved it because he was talking about his use of the subconscious in The Shining, he does all sorts of tricks like messing with the spatial layout of the hotel and all these things to kind of, say, have a new emotional experience. And so as soon as I heard of this study of awe I thought, maybe this is part of the answer. Maybe one answer for why we might want to go through the emotion of awe is to prime us to internalize the lesson of the story.  In other words, maybe all these TV shows and movies that inform resolution on us, even the ones that are supposedly strange like a Westworld or Memento or Inception still suggest that there's a right way to understand what happened at the end. Whereas with a Kubrick movie it's like, wait a minute, what happened here?
 
Maybe real art makes you question the world you're in and keeps you in an alert, open state?
 
And when I hear that I hear you setting up a dichotomy between reason on one side and passion on the other. I really believe that the best stuff is somewhere in the middle. It has a tasteful contrast and balance of both. I think Kubrick, for example, walked a great line of being able to have enough structure that his stories weren't just, you know, experiential nonsense that not many people could relate to.  Trying to keep those things in harmony is a really fun challenge.  I mean, he did screw with us in that movie. I mean, there's the scene where the old guy is sitting in his room thinking about a boy and sometimes you look at him and there's nothing behind him on the wall. And then you cut back to him and there's this velvet painting of a naked woman. Is the picture really there? No, he's doing something to our brain, intentionally. That subtle realization that something's not right adds to a feeling of eeriness. So you're playing with things that are just below that line of conscious recognition.
 
I think there are certain I don't want to say morals, but messages that are almost incompatible with the traditional kind of hero's journey. You know, if you just give people the standard version of the Hero's Journey I don't necessarily think you can deal well with what most of life is like, which is ambiguity and having to wrestle with a lot of unclear, subjective things. So it's made me wonder, are plot structures or games where you go around and explore and you're not following a linear narrative, potentially better suited to messages that aren't really cut and dried all the time?  Indie films where it's very unclear if the main character lives or dies at the end, or what the conclusion is don't tend to do as well commercially, because people like some resolution. Well, in a film, I don't have any interaction with the narrative, I can't affect it, so it feels good to have resolution. But I wonder if more interactive storytelling might be better suited to ambiguity, because at least you have a role in it, and you have time to speculate and think about how things are unclear because it resembles life a little bit more than a clear cut story.
 
I'm intuitively suspicious of virtual reality type entertainment, partly because I understand that it's really two or three corporations that own 99.9% of the VR tech, whether it's via Google Facebook, or Sony I can't imagine that they’re using it to try to open up new avenues of human imagination and intelligence.  They're more oriented towards human manipulation. But, you know, many people in VR worlds are yearning for this verisimilitude, this granular copying of the real. I'm wondering how open ended they'll be willing to make those experiences?
 
Well, this is why I tend ignore their efforts and focus on those people exploring how to make an experience that helps someone empathize more or that puts participants in a situation that they otherwise couldn’t experience, in order to think about it because now they have agency and a different relationship to it.  For now, people need to make lots and lots of VR sketches to say is this mechanic enjoyable? Is this humanizing? Is it dehumanizing? How does it interact with our emotions? Does it get a point across to people.  If it feels good, it'll survive and hopefully, you know, that doesn't get co-opted again to dehumanize us in some way. But even if that does happen, I'm confident that the people who are doing it with good intentions are gonna keep doing that.
 
There's definitely a difference between awe and spectacle.  You can create a spectacle that throws someone into something like a state of awe, but it's really just a powerlessness against you know, the state or the money or Las Vegas or whatever it is you're not doing in Las Vegas to humanize people. What are some of the litmus tests you use to decide if a piece of work is humanizing or dehumanizing?
 
Wow, that is a big question. For me, it's heavily dependent on the context, but some of the things I look at are the amount of vulnerability that the storyteller is showing. I mean, I love pieces of art that come from personal experience. And when somebody is trying to honestly work out something that is difficult to share with a mass audience, and they're trying to say, I want to, you know, through this piece of art through this narrative or this interactive experience, give you an opportunity to experience what I went through, and what my conclusion was so that that might help you ifyou go through something similar. To me that is extremely humanizing, because it involves empathizing and effectively an artist saying, you know, I want to connect with this audience and help them by explaining what I went through. I also ask whether this thing encourages me to interact with other human beings more? Or does it encourage me to interact with them less? Like I said, I think some of the best art walks a careful line between appealing to the conscious and the subconscious and balancing passion and reason. I like pieces of art that don't depict horrific exploitative technological dystopia and don't make us all want to give it all up and become completely agrarian, you know, they, I think we're in need of more and more art that depicts a world where those things are in imbalance. It's kind of stymied me sometimes and thinking about, you know, the art I want to produce. Because if that's the final conclusion that you want to get across, well, it's hard to do that with a big action sequence. But I do believe that the solution isn't to sit around and philosophize about it. It's really to prototype short things and see what works and what doesn't, right. prototype and iterate, prototype and iterate
 
And it doesn’t always have to be the indies.  Pixar is an environment to be creating things in you know, I mean, I would refer you to, to the director of InsideOut, Pete Docter makes some of the most humanizing stories between Monsters Inc, and Up and InsideOut.  He’s a very vulnerable, sensitive, amazing storyteller, and he wants to tell stories to help people feel the same thing.
 
So, yes, there are rare pockets where you can do both well, you know, make something commercial that’s also meaningful to people.
The difficulty of any storytelling is you can't just sit there at the end and go, Hey, in case you didn't notice, sadness is important. A lot of people refer to the end of the second act when they lose Bing Bong, Riley's imaginary friend, as being a devastating part of the movie to watch because they're really emotional about that loss, but I think that sadness almost physiologically primes people to receive the message of the third act. You’re not going to see somebody say, hey, you know, sadness is really important. Still, after you feel sad about something, you often have this peak afterwards, like you feel relieved physiologically after you cry, right? So, so you're now in the state watching the third act feeling a little better, because you've just gone through being sad, and you didn't get to choose to feel that way. I don't think that anyone sat down consciously and said we were going to do this, but it's my pet theory that one of the reasons the message of that film works is because you go through that feeling….and that's kind of what I'm getting at. You mentioned the dichotomy or, you know, the difference between just raw spectacle and awe, I think, you know, big action movies and stuff like that, you're always going to try to have some spectacle in the third act because that’s almost a tradition.  Or stuff like, you know, the first reveal of the dinosaur in the first act of Jurassic Park.  That might be more inspiration than spectacle, because it’s signalling that you're about to be in that environment where these characters are going to go through a major journey and they're going to change, so you relate to them a little bit more, which helps you invest in their story. That scene would be a lot less successful if they didn't show you the dinosaur or they just showed them coming back to the hotel saying, holy shit, did you see that?
 
The fact that the visuals make the audience actually experientially go through that as well is good filmmaking.  You're not communicating it just with the dialogue or lecturing, you have to give people that feeling. And by virtue of doing that, they understand because they go through it emotionally.
 
I think it’s enough that we have this capacity. When we go through emotions, or we fall in love, or eat. We like these things. And that's enough, just being human is enough. And that's why I'm not excited about emulating our consciousness with AI. I'm much more interested in how to use technology and narrative to make us feel emotions, because to me, and at least in 2017, that's the thing that feels the most human because we experience emotions and we don't have a whole lot of control over them. So, learning how to deal with them, using them to tell stories and help us make sense of human experience. That's the thing I care the most about.

.Julius Kuschke, Chief Product officer at RTC creates tools for game writers.  RTC makes Artists Draft, a software to create branching stories.  In this talk he reviews highlights from the ways that previous games have ensured what Julius sees as the four secrets of a good dialogue:
1) Pacing
2) Authenticity
3) Conflict
4) Agency

1. Pacing.

Make it sound good, make it sound like music. It needs rhythm.
 
Previously, in a hub dialogue system, you could choose all the options offered in turn,. And that's very unnatural. It's also very much like an interrogation, because only the player is driving the whole dialogue.  This turns NPCs into information wending machines, where the player feels obligated to punch every button, just to get each available tidbit of data. 
 
So, this is why I think waterfall structures are so popular today because they feel so much more natural. Players understand that each option not chosen is gone forever, they don't have a second chance. The conversation will move on.

 
Just as an example for that, let's take a look at Assassin's Creed Odyssey.
“Tell me mysterious. Did you learn anything worthwhile in your dealings with the world?”
“I'm done talking, I should tell you where you stand.”
“They said you'd be different. And blood is blood, I suppose.”
 So, even though it's clearly a waterfall structure, without reoccurring hops, the dialogue often feels slow and artificial. Whenever there's a choice, everything stands still and the game waits for your input.

The most simple solution for this problem is to have time decisions and Telltale Games in particular made that system very popular.
Timers are great to create tension, and they reduce that weird awkward silence in between choices. They can be very difficult for non-native speakers or just slow readers, however. So most games offer a short preview of texts that hopefully can be understood instantly – but that can create more problems again because unclear options making the player character behave in unwanted ways or options leading to the exact same line of dialogue are frustrating, so frustrating even that one of the most downloaded mods for Fallout 4 replaced these short preview texts with a complete line of dialogue.
 
So, just to keep in mind, if you shorten your choice texts to be able to do something like time decisions, clarity is always more important than reading speed. 

• Bioware’s Dragon Age use iTunes and symbols to make the options more clear.

 

• Another option we have is just to be very consistent in how we present our choices. e.g. Bioware positions dialogue options consistently. For example, agreeable options are always in the upper right corner. If we disagree with what the NPC said, they're, etc. 

 

• In Star Wars, The Old Republic response options were arranged differently. according to the players character class.  The smuggler class, for example, always had the most Han Solo option appear at the top of the list. so lazy players could just pick the first option to experience a very classic interpretation of that character. And that's great, I think, because at least I know what to expect when I choose that.

 

• But as with any pattern based system, it can be boring for the player and also feels more artificial. The difficulty is to create a consistent system with enough flexibility to make it feel not to systemic.

 

• Another game that uses consistency to speed up the decision making process is Alpha Protocol. Their stem system placed answers at certain positions based on their attitude, for example, professional or aggressive. But what makes the pacing in Alpha Protocol really special is that the system has no pauses.  So you pick your answers while the NPCs are still talking and that feels extremely natural and fluent. And much more recent example for that is Firewatch. And it's really a great example when when we talk about pacing, because they combined a lot of the things I just mentioned.  They have very short but precise options. They have a timer, and they have no pauses at all. 

 

• Another way of keeping players engaged even though there are pauses in the system is Heavy Rains’ moving thoughts system. By simply displaying options more dynamically, it urges the player to respond quickly, and it keeps up tension.  n.b. moving texts can be very hard to read for a lot of people, so if you're doing something like that, be careful, maybe at least offer an option to disable these animations.

 

•  If actions do not appear in predictable patterns, players are more inclined to stay alert at all times. 

 
 
2. AUTHENTICITY

If we don't think that the characters are believable, chances are very low that we are interested in what they're saying.

Oftentimes when we talk about games, it's more about the illusion of choice. It's not about real choice all the time. …It doesn't have to be real agency either as long as it is believable.

I don't think that we need a completely advanced AI. We just need to give NPC the illusion of agency by creating more believable characters.

One way to do that is to give them their own agenda. Basically, to give them a life of their own and great example for that are the NPC conversations in the camp in Red Dead Redemption 2.
“Come on, Jack.”
“I'm hungry Mama.”
“We're all hungry son. Just try reading later.”
“Okay,”
“Gotta get some to eat Arthur.”
“Sure”
 
So this scene would also happen without the players standing so close. But the last sentence is only triggered if player's are positioned close by and that really impacts.

The feeling that the NPCs are aware of my presence, they know that I'm there. So I as a player make a difference. But what is more important is that I'm not in the center of everything.  The NPCs they have their own worries, they have their own thoughts and feelings, and they just talk to each other.
 
In real life, many things happen at once. You can't be everywhere or experience everything that's just not possible. In The Last Express a game released back in 1997 by Jordan Magna, the creator of Prince of Persia things happen whether you're there or not.  The NPCs have their own schedule, they sleep, they get up, they go to the dining carriage, and they have conversations there. And if you don't go there, they still do that. It still happens, the story progresses anyway. And this really elevates NPCs from being marionettes only reacting to the player to believable acting personalities.
 
A simpler approach to make NPCs feel more alive is to just let them actively seek interaction with a player. And I would call that mixed initiative, with just a simple timer in the background paired with a few variable triggers, it can seem as if the characters have a will of their own because it was not predictable when they will do that.
 
So if you always ignore them, if you always decide against what they are proposing, they can also just leave you. They can tell you at some point, Okay, I'm done with you. If you don't listen to me, I’ll go my own way.
 
But to be honest, there was one thing that could lead to really awkward situations and that was because the characters didn't have a really good knowledge about the game world, or what has happened so far. …but a game that does that so much better is again Firewatch because they have a system where the characters know quite a lot about the game world. They know everything that has happened so far. And that really influences how they behave and what dialogue choices they have. In Firewatch dialogue lines are selected by a system that tries to find the line with the most matching requirements, so they don't have a traditional dialogue tree, they have a completely different system. And how that works is, I think best explained with an example. On day two of the game, the NPC Delilah would start a conversation with a player. But it makes a huge difference how much the player already told her about his wife Julia. If Delilah knows nothing about Julia, you get a pretty generic statement about relationships. But if the player had spoken about Julia and also told them that Delilah wasn’t feeling well  they heard a more fitting response. “What does she have?”  And the magic about the system is that players probably won't even notice its complexity. It's very invisible in a way. But still, it really makes me feel that all the choices I make matter, and they have consequences. And the NPC respond to me in a very natural way. 

3. CONFLICT
Without conflict, there is no drama.

• But how could we create a dialogue system that really enhances that feeling of conflict? There are very few examples I found that I think are interesting. One is the Council, which has some kind of resource management system built into the dialogue system. and it certainly makes you think, do I want to spend my five points now to get that piece of information? Or do I want to save to maybe convince someone later on. So it creates a conflict for the player.
• And in Blindfolds we try to do something kind of similar, so every line of dialogue basically costs time, and there's nothing I can do about it. Everything I say, will move time forward. So I have to be really careful to choose which character I want to talk to and when I want to switch a topic.
• Interrogation are game-like.  You can actually win or lose a conversation. And players probably think twice before choosing one of the options. And they are much more focused on what is being said because in the end, they have to judge between truth and lie.
• If we look at game mechanics to integrate into our dialogues, another great example is just having puzzles inside of the dialogues.
• The deeper and more emotional the relationship between two characters, the more we care about the conflicts arising between them. Probably one of the best examples for a game that uses relationships to create conflict is The Sims, and that's kind of sad, because the in character interactions in that game cannot be really called a dialogue. The romance systems in Bioware games are more concrete. And they usually have some kind of value to track a relationship between an NPC and player character. And just to have the player aware of that value, really strengthens the feeling that my choices in a conversation matter because saying the wrong thing might damage a relationship I care about. But this can also feel artificial and shallow e.g.  the approval system in Dragon Age Origins. Here, the conflict  is undermined by giving players the option to simply increase the approval rating by making gifts to the NPCs. So I could just install an NPC and never listen to them, but give them a kitchen and hey, we're friends now, right?
• In real life relationships, the challenge is to truly listen to people and to remember what someone cares about in order to say something that they appreciate. In Tokyo Mickey Memorial for example, neglected infrequently dated characters would eventually become angry and gossip to their friends. severely reducing love meters across the board. In the middle of the game, when the number of known love interests was high, these bombs became the primary concern,  forcing careful planning and strategic round robin dating. I would like to see more jealousy systems in games, because it creates conflict…but forcing the player as a solution to undertake this round robin dating strategy is a bit silly, but if that system can offer meaningful choices that could be really, really interesting.
• Of course, there are several other ways to integrate game mechanics into dialogue systems and even simple things such as a skill check in an RPG can enhance player engagement because it makes that choice somehow stick out, it makes it feel more valuable, more rewarding. BUT the greater the sense that a player can win a conversation, the more we engage the analytical calculating parts of the brain, which presents a barrier to empathy. If we want to tell a good story, we have to be careful what we're doing there. If it's too systemic, if it's too artificial, then players will not really care about what's going on anymore.
• So we have to find a good middle ground. Adding elements of conflict to a dialogue system can work, but only if it really fits the setting, like the interrogation example in a crime story,

 

4. AGENCY
Agency is the level of control the player has over the game world. It's the feeling that my choices matter. Usually NPCs have no agency. They respond in pre-programmed ways, and this makes them feel like marionettes, which destroys the illusion of interacting with another character.
 

• So I think the most prominent example to challenge this is probably hinting at consequences as done by Telltale. “Clementine will remember”, this phrase almost became iconic for story driven games.
• ​Hinting at consequences also shows that you need to deliver on the agency you're promising. And I think that was kind of the problem, at least for me in the later Telltale Games. And a lot of people in the press, as I think saw that equally. For example, Game Informer said at one time, I can't help but feel a bit deceived. The Walking Dead had me believing that I could save characters from death. After multiple playthroughs I realized characters fates are etched in stone. 
• Allowing the player to choose the topics they're interested in can enhance the players feeling of agency and that's why we try to combine dialogue systems. If you initiate a conversation with a character, first of all, you get a very traditional app based UI where you can just choose the topic you would like to talk about. And as soon as you choose that, it switches completely to a very natural waterfall system.
•  Agency is also about being in control. Again, Firewatch is a good example.  You can always move and look around even hear dialogue. It even allows you to interrupt an ongoing conversation by switching between topic, … So you wander around, you're talking to Delilah. And you come across that tree. And you could just say, Oh, I found some claw marks. And that really changes the topic of the dialogue.  Whenever a conversation turns to a new topic, a bookmark is set. If the dialogue is interrupted, now the system will try to resume the dialogue at a later point. And there is a specific entry point for each bookmark. This entry point comes with a special line of dialogue used to pick up a topic in a more natural way
• 
  

Truly engaging dialogue systems are still the exception. But as these examples show the design of a dialogue system has a significant impact on pacing, authenticity, conflict and agency and probably a whole lot of other good things as well.

Kuschke, Julius. 2019. Narrascope 2019 - Julius Kuschke - How Dialogue Systems Make or Break Player Engagement. U.S.: Narrascope 2019.