So ... I just thought I’d take on the small task of analysing what it means to be human and using the results to model NPC motivation and behaviour.
Hmm.
When I’m exploring a new area, I normally like to be as systematic and exhaustive as possible to make sure every aspect is included. But with human nature, this becomes a bit tricky. Where do you get a list of things that are relevant to humanity?
In this series we are reflecting on ways to make game characters
more human and believable. Sometimes autonomous NPCs are assigned a set of
scripted actions to perform, which can certainly make them seem more
real and involved in the world – but scripts are not very
adaptable to unexpected events. NPCs could also wander around doing random
actions, but characters with no real motivation are ultimately unconvincing.
In real life, everything people do is for a reason, even if they are not consciously aware of it: motivation is the source of behaviour. If we can give NPCs a reasonable set of motivations, they can act in consistent ways that are more flexible than scripts and more believable than random activity.
So ... why does anybody do anything, ever? To answer this we will turn to a deeper question: what does it mean to be human?
The International Conference on Interactive Digital Storytelling (ICIDS)
has been held annually since 2008, or 2001 if you count the two conferences it was formed from (ICVS and TIDSE).
The papers from each conference have been collected into a set of PDF (and physical) books. These are quite expensive to download, but working at a university means I have free access to them at the moment.
Topics include things unrelated to IF such as animation and virtual interfaces, but there are many subjects related to IF: narrative theory, story generation, agency, player emotions, drama management, autonomous characters, emergent narrative and natural language.
I haven’t been to an ICIDS conference, but you can read a first hand account from Juhana Leinonen who attended in 2015 and Emily Short who was invited to talk about Versu in 2014.
Here are the papers I’ve come across that are relevant to this blog, along with links where I could find them.
love > like > neutral > dislike > hate
This doesn’t leave much room for subtlety. A courtier might be openly friendly but inwardly envious of another member of the court, unwilling to betray their secrets because their rival could just as easily give away theirs, grudgingly admiring their drive and ambition while despising their family pedigree. Summarising this relationship as “dislike” loses a lot of interesting detail about how the characters relate to one other.
In this follow up to NPC Emotions part one and two, we take a look at how relationships form, what influences their quality and what might be possible in an advanced IF system.
“People are much more difficult to simulate than rocks and trees, not just because of relative complexity, but because we’re more wired to scrutinize our fellow humans” (Artificial Intelligence In The Uncanny Valley, 2007).
In the original context this was talking about graphics, but it applies equally to text based interactions with NPCs. On thing we are wired to look for is emotion; a character who never reacts to events or varies their emotional state seems a lot less real.
Continuing from part one, here are some more ways emotions work in real life that have potential for IF.
This is part of a series about potential future developments
in IF, particularly in the area of believable characters.
One way to make NPCs more human is to model their emotions.
This area has been explored quite a bit in the context of
Artificial Intelligence.
A model of NPC emotions is particularly useful in the area of autonomous NPCs, where interactions can happen that are difficult to predict. Using general rules rather than scripting reactions to every possible combination of events is a good way to manage this kind of complexity.
Conversation
... conversation parsing; creating dialog content; making the input manageable.
Autonomous NPCs
... modelling a human being; motivation, personality and emotions.
Drama management
... dynamic story generation; plot building blocks; player freedom.
Interactivity
... extending the parser; how much to simulate; emergence and unexpected solutions.
Artificial Intelligence
... knowledge representation; AI techniques for IF.
It would take a thesis (or more) to do justice to each of these topics, and I’m no expert on any of them, so I hope to learn a lot on the way.
There is nothing particularly wrong with the current state of IF, but it’s fun to explore what might be possible in the future and what it would take to get there.
