New take on Conversation in Video Games

Those aren't the only 2 types of conversation.

3. Text-only games like MUDs and the King's Quest series commonly featured a vocabulary of key words, which the player used as their main method of interacting the game, including talking to NPCs. They were commonly limited to 3-word sentences, because an Eliza-type AI (or CleverBot-type) has more difficulty parsing sentences in any sane way the longer they are. In some cases the player must pick a sentence paradigm, which is like a mad-lib, then fill in the blanks in the sentence, either by typing or from a drop-down menu. The downsides of this system include making the player remember keywords and emphasizing the fact that the NPCs aren't really human.

4. Some RPGs, instead of giving the player specific dialogue choices, give the player a standard array of actions, like: agree, disagree, ignore, intimidate/taunt, charm/joke, buy/sell/bribe, ask for more info, answer truthfully, lie, etc. The game then supplies the player's dialogue to fit the choice and the situation, and understands the player's intent without needing to interpret anything, allowing the NPCs to always react correctly to the player's intent. This system can combine well with a system keeping track of which NPCs (and the player) know a piece of information (including false info). I think the Erasmatron engine was an example of this type.

Well, what you describe is not really new. But it is nice to see another soul that's thinking how to replace the simplistic dialog tree mechanics. So don't get me wrong because most of the following sounds like critique.

a) There are more dialog mechanics besides the both you mentioned. The group of "branching dialogs" consists of the dialog tree but also of the "hub and spoke" variant; there are even planner like mechanics. See e.g. this and this article on Gamasutra.

b) A FSM is by definition a collection of states with conditional directed transition in-between. In an FSM just 1 state is active (as long as you do not mean HFSM). Implementing an FSM with those 6 degrees of freedom would practically become impossible very fast. What you describe seems me to be something different, more like some kind of decision tree.

c) A "state" you've forgotten is the need / goal of the speaker. This is very important, because a person is willing to ignore mood and whatnot in case that it helps to yield a goal. It must even be possible to lie. All this may be important to drive the story, too.

d) Having the possibility to change the topic is nice not only for the player but also for the NPC, because it allows an NPC to drive the dialog into a direction that matches the story. But care must be taken that an NPC does not rapidly hop from one topic to another. When the player des so then the the opponent should probably get angry (i.e. changing its mood). See e.g. Emily Short's articles about that problem. (BTW: Emily has many interesting stuff to say, although she's more related with IF instead of video games.)

e) Besides the mechanics, also the UI plays a big role. The player need to recognize whether a sentence may endear or insult the opponent. Remember that not all players may understand the language in which they are playing good enough to recognize nuances. It is also nicer for the player not to need reading 4 times 5 lines of text just to decide which answer to give. There is an article about dialog UIs on Gamasutra (unfortunately I do not find it ATM). The more influence a dialog system has, the more important is its flawless controlling.

f) Regarding the effect of stats on dialogs and the possibility to change stats due to dialog, you may want to read the PDF about SimDialog available here.

I think a state machine or decision tree are too limited as approaches. If each variable other than topic has 4 states, you're already at 1,024 edges or nodes. Obviously, you wouldn't write out a variation for every single possibility, but even at 1% of variations, you're talking 10 edges and that'll get very complicated very quickly. I think you'd drive yourself crazy managing them.

My idea would be to set the preconditions and results of each snippet. Something like this:

Action: "Lament about your upbringing"

Text: "Let me tell you about...(blah blah blah blah blah)"

Mood: NOT (Angry OR Happy)

Relationship: Friend OR Lover OR coworker

Childhoods: Active

Importance: 80

Repeatable: Never

Result: Mood = Sad, Childhoods += 2, Orphanage += 3, Parents += 1, Profession += 1

Action is the choice you'd display to the player, and text is what would get displayed when it was selected. The next three variables set when this statement makes sense (what's your mood and relationship, and is childhoods an active topic). Importance is for filtering which options to give when many are possible. Repeatable is how soon you could return here, e.g., once per conversation (you'd probably need something a little fancier to keep from retelling the story with a different mood).

Results would change variables. Some set the appropriate tone: the states you discussed. The others set reasonable transitions: if the conversation is about being an orphan, the other conversant could talk about their childhood or parents, or talk about what you've since become. Basically any side road off the given text. I included values for these, on the idea that a topic decays over the course of a conversation. You can double back to talk about the orphanage more up until a point, and then we've moved on.

This way you can just write content without getting too caught up in mapping all the ways it could flow together. Maybe there's a murdered_parents variable, if this variable is activated (getting them to admit it, someone else tells you, putting together clues), you can go accuse them without worrying too much about how you got there or exactly when you do the accusation (maybe you wait for just the right moment). This feels more natural to me as a writer: choose a scene, write it out, move on (rather than writing variant after variant and constantly interupting to map flow).

That said, this is a rather ad hoc method, and it would be very easy to leave dangling story lines or branches that just can't be reached or conversations that tend to peter out. I think you'd need some code doing smart analysis to help you find what content needs writing, and a good content generation tool to keep track of variables.

Anyways, just a brain storm, hope it helps.

sunandshadow, I'm afraid I have to disagree with your examples, to me both your examples appear to be dialogue trees where the choices are merely obfuscated by the system. The text only games simply let you add a couple words here or there or let you choose your dialogue option by typing it in, while the standard choice of those RPGs you mention simply connect standard options to specific dialogue options. Neither would work all that well for what I'm thinking of, unfortunately.

The differance is that most, if not all of the dialogue choices would be scripted specifically for that situation, while I want a system that dynamically allocates phrases and words from a bank in such a way that it seems consistent and logical without having to script it out specifically.

I don't want to design a slightly different phrase depending on what mood someone is in when they deliver a message. I want that phrase/word to be chosen by the system itself.

haegarr, I like some of the ideas you put fourth here. A goal is a nice touch, and topic changing segues as well as topics the NPC would like to avoid and such. Topics could be highlighted in such a way that indicates what your character THINKS would happen if they said it. This would not always be entirely correct, as the player/character will not always have all the information, just like in real life.

Polama, you approach is interesting, and believe me, I know the difficulty of writing all the possible content. But notch didn't design each and every block, each vista and ocean when he made minecraft, did he? What I propose to be done is to write words and phrases and let the system choose how they fit together based on the rules we give it.

I propose there be 2 databases. 1 database for tags, and 1 for actual words and phrases.

Each word and phrase will have tags associating them. So, say, KILL would be tagged with attack, anger, enemy and HELP would tagged with assist, (positive emotion), friend/ally. These tags and the system that reads them would have to be refined, but you get the idea.

Then the phrase, "%person% %question% %action% %character%." could be informed by Relationship: Boss->Subordinate, Disposition: Giving order, Mood: calm, topic: attack, enemy. this would make the phrase "I order you to kill Caren."

But who would make all the content? Who would tag all the words? Well, people, of course. People are willing to help other people most of the time, especially if helping is a game. So over time a huge database of words and tags could be created and plugged into the system, which would be optimized to do the sort of word choosing quickly.

Thoughts?

There's a company, Narrative Science, working on this sort of thing. You can feed them a log of a baseball game and they generate an article about it. It's a hard problem, though. It's hard not to get lots of very similiar, simple sentences. "Caren stole my crown. I want my crown back. I order you to kill Caren. I order you to retrieve my crown." Fine, acceptable prose.

But as a human, I'd write something like "My investigators have been scowering the town, and an adventurer named Caren left the night of the awful theft. My crown, god, the symbol of my power! There's enough unrest as it is. I've got everyone I can fanning out; I want you to head West towards Adventureshire and see if you can learn anything of this 'Caren'. Get me my crown back. I want this kept as quiet as possible so don't bother bringing her back, just eliminate her. And I know those rebels have been spreading dirty words about my generosity, but I promise, if you succeed you'll have more treasure than you'll know what to do with."

I think generative content is an interesting problem to attack, but there's so many ways to construct a sentence, so much below the surface weaving together sentences. And so much knowledge! What can be held? Lit on fire? Comforted? Consider "an adventurer named Caren". We could do "a beautiful woman named Caren", "an orphan named Caren", "a level 3 wizard named Caren", "some jerk named Caren", "Henry's sister, Caren,". Each would be appropriate for some piece of text, not all would fit well or make sense in the boss's command.

So at the end of the day, I wouldn't expect anything better than poorly written text. That would be pretty amazing, though. It's a very cool problem to make progress on.

One thing to try is using deep learning. I have no idea how or why, but that seems to generate human like results when it works =)

Another interesting point, Polma! What if we systematized paragraphs as well as full conversations on top of individual sentences, with lots of back checking. Keep doing this, reiterate on the problem fast enough, and you can get something that works acceptably with enough testing. Especially when the testers have the ability to directly affect the script using the controls.

But what tests would we do that could be specific enough for a computer to do quickly, but vague enough that it works on a language problem?

Using some sort of grammar for paragraph layout should work. There's a continuum, where at one extreme you're writing extremely generic patterns and working very hard at them and your representation of the world and artistic quality, and hoping they'll work for anything. At the other extreme, you're writing out the content directly. In the middle you have madlib style patterns and small modular pieces that go together, etc.

A very generic paragraph might be: What? Who? When? Where? How?. "You must retrieve my crown. Caren has it. Do this now! She was last seen in Adventureshire. Go there and kill Caren."

You can also insert "tell me more" sentences anywhere. "You must retrieve my crown. It is gold with a ruby inset. Ruby is a red gem exported by Dwarvia. Dwarvia is an unfortunately named elven settlement."

But for higher quality content, you'd want more specificity. A love proclamation might be [statement about the current situation. Statement that they like a positive quality about you in the current situation. Positive shared memory. Positive shared memory. "I've been meaning to tell you, but reason I haven't." Proclamation of love.] Even more specifically, you might have different grammars to handle telling a person being kidnapped you love them, telling a person after a big battle you love them, telling a person on a romantic date you love them...

For content, you can have a generic "Oh no, I've (lost my [object] | my [object] was [destroyed])!. Oh no, I've lost my sword! Oh no, my cat was killed! Oh no, my hair was burned off!" Or you can have a separate sentence or two for a lost pet versus a lost weapon. And for more specificity, you can have content for a murdered beloved pet of a gruff warrior, or a dead wizard's familiar that was brand new to the job. But the nice thing is that you could fill in the levels of detail over time: if there's nothing else written, it's just "oh no, I've lost my eyeball". If there's a long rant written up for eyeball-eaten-by-spider, you can pull that.

For testing, you could have something surreptitious for marking errors within the game, just right click on a sentence and choose "ungrammatical, awkward, inappropriate, etc." Have it send that info back to a server to collect. When examining the errors, I'd suggest having the engine generate a bunch of results with small tweaks, so you can say which other things display the issue. Was it a bad pattern and all this is broken? Is it specifically that a dragon liver is not a good wedding present? Would it be a good birthday present though? No, not a present at all?"

Anyways, fun stuff to think about. I do think it is plausible, but I do think you'll need lots and lots of fairly specialized logic and content.

Cool. Reminds me somewhat of the dialogue system in Glass Rose

When I first read this topic, I immediately recalled a video that I found so many years ago (but forgot the name of it). Today, I happened to come across it again. Check this out:

[media]https:

[/media]

Edit:

I like the visible word bank approach. It alleviates much of the frustration that is often involved in "guess the verb" systems.

First we start with an absolutely huge cache of words, phrases, and paragraphs.

Soo, the AI would need to "find it's way" through this ?

It would come closer to how humans think then most algorithms, so +1 for that,

I see however two problems:
1) It would take a real big database, probably something like a 20-dimensions big matrix.

2) Without action tied to the conversation, it'd be just another cleverbot.

@Onigiri Flash:Awesome vid.

@Onigiri: That's intriguing, and quite impressive. I'll admit that it looks somewhat awkward to interact with in the video, but it may be rather more intuitive when one is actually using it. (I find the changes in the sets of words available as one builds a sentence to be a little disorienting, but this may be much like watching someone move quickly through a menu system: the fact that one is not controlling the actions makes it easy to become a little lost, even if one could move just as quickly when controlling the menu oneself.)

I do wonder how they provide information to the AIs--where is it specified that there is a thing called an "axe", that it costs "3 gold", and that NPC #2 wants it?

I'm reminded of a video that I saw regarding Shroud of the Avatar; I don't seem to be finding the video that I have in mind, but this one seems to show the system in action (starting at around 11:40):

The result looks quite similar, although I wouldn't be surprised if the mechanic used in Shroud of the Avatar was a little less adaptable.

I've given some thought to conversation mechanics myself: The "adventure" genre (by which I mean the genre that includes games like Gabriel Knight or Myst, not games like Zelda) a favourite of mine, but I've found that on occasion conversation can be somewhat underwhelming and low on interactivity, at times becoming little more than a barely-interactive cutscene: one just clicks through until all options have been visited.

An idea that I've given some thought to is based on the conversation system used in certain old games, such as the Worlds of Ultima games, Savage Empire and Martian Dreams. In this system, the player typed keywords (or short phrases) in order to elicit responses, and important keywords in NPC responses were often highlighted. The result is quite simple, if these days a little archaic. It also perhaps lacks a little in gameplay depth: one can easily enough spot keywords, note them down, and exhaustively go through the list with every character (if one has the patience for it).

What I have in mind, then, is this: The player has two lists of keywords/key-phrases, corresponding to their character's short- and long- term memories. Long-term memory would persist between conversations, while short-term memory would be empty at the start of each new dialogue. During conversation, the player selects entries from either short- or long- term memory, eliciting some response from the NPC, much as in the old games described above. These entries are generally taken in turn from NPC responses: the player picks words or phrases (via mouse) from the most recent response, adding them to their character's short-term memory. Furthermore, both short- and long- term memory would have limited capacities (with long-term memory likely storing more than short-term), prompting the player to decide which topics they feel to be worth pursuing, and which to leave behind as the conversation (and game) progresses. I would imagine that "saving" items from short-term to long-term memory would be done at the end of a conversation.

I see three main problems with this approach, however:

• It would presumably call for quite a bit of additional work during development, tagging each phrase or word in each dialogue. Conversely, if only important words can be picked, then they may become somewhat easy to find.
• Brief responses from NPCs may make it a little too easy to pick out keywords, and it might be tricky--not to mention obnoxious--to have every NPC speak at length on even the most trivial topic.
• Finally, when the dialogues themselves are brief, the restriction on memory-capacity might become irrelevant unless made ridiculously low.

I do think that this system might work in some sort of mystery game--although I haven't tested it.

I enjoyed The Elder Scrolls: Morrowind's system, of having a list of keywords gained in conversation and via the story, unlocking new branches of dialog for NPCs.

Inspired by Morrowind, I've put some casual thought into it. I think I'd want group-specific topics. "Group" can include NPCs belonging to a specific organization like the imperial guard, or it can be geographical like everyone in a specific village. NPCs should be able to belong to multiple groups. e.g. Bob is a soldier of the imperial guard, and stationed in such-and-such village. He's also a PazaakPlayer (since he's a fan of KOTOR), and part of the underground black-market.

Each group he belongs to adds some dialog options to the available options. The player has to have "unlocked" that dialog choice before it shows up even if the NPC is from the right groups. Some dialog options might even be conditional on other characteristics - like the player's current standing with that group.

What I have in mind, then, is this: The player has two lists of keywords/key-phrases, corresponding to their character's short- and long- term memories. Long-term memory would persist between conversations, while short-term memory would be empty at the start of each new dialogue. During conversation, the player selects entries from either short- or long- term memory, eliciting some response from the NPC, much as in the old games described above. These entries are generally taken in turn from NPC responses: the player picks words or phrases (via mouse) from the most recent response, adding them to their character's short-term memory. Furthermore, both short- and long- term memory would have limited capacities (with long-term memory likely storing more than short-term), prompting the player to decide which topics they feel to be worth pursuing, and which to leave behind as the conversation (and game) progresses. I would imagine that "saving" items from short-term to long-term memory would be done at the end of a conversation.

I see three main problems with this approach, however:
It would presumably call for quite a bit of additional work during development, tagging each phrase or word in each dialogue. Conversely, if only important words can be picked, then they may become somewhat easy to find.
Brief responses from NPCs may make it a little too easy to pick out keywords, and it might be tricky--not to mention obnoxious--to have every NPC speak at length on even the most trivial topic.
Finally, when the dialogues themselves are brief, the restriction on memory-capacity might become irrelevant unless made ridiculously low.

That's an interesting game mechanic, for sure (choosing which information to store in long-term memory).

To address the problems that you mentioned with your idea, I think it would work well if you went with the latter--tagging only important phrases or phrases that can spark specific conversation topics. I wouldn't worry too much about making them too easy to find. I tend to find the most enjoyment in dialogue bits that provide clear options, but invoke a sense of strategy and contemplation in deciding which ones are most valuable to pursue (without the opportunity to replay the entire conversation).

In Dangan Ronpa (speaking of mystery games), there's a semi-interactive conversation system. Some bits of dialogue resemble hyperlink fiction (click on a highlighted word or phrase in the text). However, the one thing that I feel the dialogue was missing (outside of trials) is a sense of urgency. Players can sort of "mine" the conversation until all routes have been taken in the dialogue.

Great content in the linked video.

Some dialog options might even be conditional on other characteristics - like the player's current standing with that group.

I've also considered implementing this type of dialogue (with faction-influenced interactions). As food for thought, Emily Short has written several great articles on the topic of NPC conversation.

I'd just like to be the vague voice of reason that says: Videogames (on the whole) lack good writers. Compare even your favorite games to your favorite movies or books and you'll get an idea. Now imagine the difficulty in crafting multi-variable conversations to craft an enhanced story.

Frankly I think you'll end up trading the semblence of interactivity (if you succeed) at the cost of good writing. Not saying you shouldn't try, as what you want to do may be the holy grail of game design, I am just trying to point out trading interactivity for entertainment is probably unwise.

I'd have to agree with DanglinBob - the whole idea and a lot of the examples given make valiant attempts to make conversation more interesting, but in the end, it never is. I read somewhere a conversation about dialogue in literature. It said something like "conversation is always stereotypical and dull - so avoid it unless it furthers the story". So the idea of making conversation more interactive in games is like trying to make digging a hole more realistic and interactive.

The Morrowind system was interesting, but in the end you often just ended up going down the list with every keyword trying to find the info you need. ShadowRun for SNES did something very similar with similar results.

There are some chat-bots online that can simulate conversion somewhat, but often end up having issues themselves. There are probably entire websites and forums dedicated to the hobby of making chat bots. Despite their relative realism, they still would probably be unwieldy in a game as well as rather difficult to program - especially if each had it's own personality, history, lines i needs to say, etc.

All that said, there was a game released quite a while ago that was suppose to have an interesting take on conversation. I haven't played it, but your topic made me think of it. It may be worth checking out. However, i think the situation in the game is rather controlled with limited characters and a specific context for the conversation, so it probably wouldn't work in a broader game.

It was called Facade

https://en.wikipedia.org/wiki/Façade_(interactive_story)

Don't forget about one thing, the goal is not to make a conversation system (you don't have conversations with NPCs, that's not the point). The goal is to provide quests and information by NPCs.

(the same goes for books BTW, characters in books have no conversations, their talking is just a tool to move the action forward or to provide information about characters involved)

Unless you make a chat simulator, not a game :)

I'd just like to be the vague voice of reason that says: Videogames (on the whole) lack good writers. Compare even your favorite games to your favorite movies or books and you'll get an idea. Now imagine the difficulty in crafting multi-variable conversations to craft an enhanced story.

Frankly I think you'll end up trading the semblence of interactivity (if you succeed) at the cost of good writing. Not saying you shouldn't try, as what you want to do may be the holy grail of game design, I am just trying to point out trading interactivity for entertainment is probably unwise.

This would be true...if it's never been done before (this is the core gameplay in visual novels, for instance). Good writing is an issue; but it honestly has nothing to do with the dialogue system. It's only a matter of finding more creative writers, I believe, and bringing them into the project at its inception (not towards the end).

Don't forget about one thing, the goal is not to make a conversation system (you don't have conversations with NPCs, that's not the point).

Not exactly. I think EarthOtherwise is trying to find a unique method of handling dialogue for general use (ex. types of interactive dialogue include keyword, hyperlink, branching, etc.), to shift from passive to active conversations in video games.

This would allow us to make conversation a part of the game rather than something that happens in cut scenes or scripted dialogue scenes. We tell the player about the relationship, we give them the ability to change their disposition and mood within reason, we let their perception stats determine how they perceive the mood and disposition of the other character, and give them a few topics they can cycle through based on their knowledge and the character.

That's an interesting game mechanic, for sure (choosing which information to store in long-term memory).



To address the problems that you mentioned with your idea, I think it would work well if you went with the latter--tagging only important phrases or phrases that can spark specific conversation topics. I wouldn't worry too much about making them too easy to find. I tend to find the most enjoyment in dialogue bits that provide clear options, but invoke a sense of strategy and contemplation in deciding which ones are most valuable to pursue (without the opportunity to replay the entire conversation).

Thank you! ^_^

Hmm... You make a reasonable case for tagging only important words/phrases--and I'll confess that it's attractive in that it incurs rather less work!

One of my reasons for wanting to obscure the important words/phrases was, as I recall, to encourage players to engage with the dialogue, not simply skimming over it--but then I suppose that if the player isn't already interested in the dialogue, then requiring them to pay attention to it might be more frustrating than engaging.

I still haven't found a game in which this mechanic fits, but I do hope to try it out at some stage.

In Dangan Ronpa (speaking of mystery games), there's a semi-interactive conversation system. Some bits of dialogue resemble hyperlink fiction (click on a highlighted word or phrase in the text). However, the one thing that I feel the dialogue was missing (outside of trials) is a sense of urgency. Players can sort of "mine" the conversation until all routes have been taken in the dialogue.

I think that I've seen an approach like that, and it can be quite interesting--and perhaps not entirely unlike a proper conversation, if done well. A sufficiently dense thicket of hyperlinks could be quite easy to get lost in--or at least to lose previous branches in, I suspect.

I enjoyed The Elder Scrolls: Morrowind's system, of having a list of keywords gained in conversation and via the story, unlocking new branches of dialog for NPCs.

Ah, I'd forgotten about Morrowind's system! That's essentially a point between the old system that I described (the one used in games like Savage Empire or Martian Dreams) and my idea.

I'd have to agree with DanglinBob - the whole idea and a lot of the examples given make valiant attempts to make conversation more interesting, but in the end, it never is. I read somewhere a conversation about dialogue in literature. It said something like "conversation is always stereotypical and dull - so avoid it unless it furthers the story". So the idea of making conversation more interactive in games is like trying to make digging a hole more realistic and interactive.

I really don't agree, regarding either conversation in written media or in games. I've seen more general advice to avoid elements--whether dialogue or otherwise--that don't further the story, but then I do think that there's space for such things as "narrative filligree" and "garnishing the story". At the least there's the point that a conversation can further characterisation, even when it doesn't move the plot. I'm also not convinced that conversation is "always stereotypical and dull".

Don't forget about one thing, the goal is not to make a conversation system (you don't have conversations with NPCs, that's not the point). The goal is to provide quests and information by NPCs.

(the same goes for books BTW, characters in books have no conversations, their talking is just a tool to move the action forward or to provide information about characters involved)

That's a rather... mechanical approach, I feel, especially with regards to writing. The advice that believe that I've seen regarding characters in writing is to write people, not narrative devices or collections of tropes. Writing characters as devices will tend, I suspect, to result in flat, uninteresting characters--and likely flat, uninteresting dialogue, too.

To quote Spiderweb Software (Whose games have over 200,000 lines of dialogue in them). Never force your player to read :)

Sad but true!