Contextualized Conversational Responses

From gdp3
Revision as of 21:44, 31 July 2014 by Staffan Björk (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

The ability of agent to gives responses in a conversation depending on all relevant game states.

Conversations in games may be constructed so they are generic and function in many different settings. When they instead are specific for the current context in which they are taking place, they provide Contextualized Conversational Responses.

Examples

Although only being weak examples of the pattern, non-player characters in Fallout New Vegas and Skyrim have different dialogues when they have moved between places as part of player's progress in quests. While this is really because the players have gotten further in the quests, referencing to the current locations can provide an illusion that the non-player characters say different things depending on the context.

Anti-Examples

Although not much of a conversation, the quotes when selecting individual units in Warcraft III are the same for all units of the same type. The default response when talking to any guard in Skyrim is "I was an adventurer like you, then I took an arrow in the knee".

Using the pattern

The primary choice when creating Contextualized Conversational Responses is whether to do it through creating rules in Algorithmic Agents or providing specific Dialogues. In the latter case, the pattern can emerge through the use of either Character-Specific Dialogues or Location-Specific Dialogues.

Diegetic Aspects

Since the use of the pattern makes Dialogues fit not a specific context of a game's diegesis, it naturally follows that it also creates Thematically Consistent Dialogues.

Consequences

Using Contextualized Conversational Responses provides Context Dependent Dialogues and in turn Thematically Consistent Dialogues. By providing Algorithmic Agents with the possibility of expressing the proper emotions regarding a specific context, the pattern can also help show that they have their Own Agenda.

Relations

Can Instantiate

Context Dependent Dialogues, Own Agenda, Thematically Consistent Dialogues

Can Modulate

Algorithmic Agents, Dialogues

Can Be Instantiated By

Character-Specific Dialogues Location-Specific Dialogues

Can Be Modulated By

-

Possible Closure Effects

-

Potentially Conflicting With

-

History

A rewrite of a pattern that was part of the original collection in the paper Gameplay Design Patterns for Believable Non-Player Characters[1].

References

  1. Lankoski, P. & Björk, S. (2007) Gameplay Design Patterns for Believable Non-Player Characters. Proceedings of DiGRA 2007.

Acknowledgments

-