adaptations generated by N0, N1, and N2 for one of the
two tasks. Our hypothesis is confirmed if people prefer N2
to N1 (N2>N1) and N2 to N0 (N2>N0).
Twenty-five participants were involved in the study. The
results are summarized in Table 1. All results were put to
one-sided tests on binomial distribution at the significance
level of p < 0.05; asterisks (*) mark significant results. In
group 1, a significant number of participants preferred N2
to N0, but no significance was found about those who
preferred N2 to N1. For plotlines in group 2, a significant
number of participants preferred N2 to both N0 and N1.
Results from group 1 and group 2 should corroborate,
suggesting a hidden independent variable. The N1 plotlines
in both groups contained a dead end. However, the group 1
dead end appeared to be events that were never followed
up, whereas the group 2 dead end directly contradicted the
apparent intentions of other events. It is likely that our
system, using formal definitions, is more sensitive to story
incoherence than human game players. Thus, we believe
that group 2 plotlines, consisting of more disruptive and
noticeable dead ends, are more representative of worst-case
situations. Group 2 results indicate that it may be beneficial
to be cautious, erring on the side of being overly sensitive
to story incoherence. Results of Group 2 validate our
hypothesis, leading us to believe that enforcing narrative
coherence is beneficial and that no harm is done by being
overly sensitive to story incoherence.
Conclusions
As game players possess different motivations, tastes and
needs, a one-size-fits-all approach to game plotlines may
prove to be limiting. We treat adaptation as the
optimization of plotlines based on requirements derived
from a player model employing knowledge about player
preferences and a model of novelty. As such, we find an
offline approach to be beneficial in achieving global
optimization of plotline structure.
The adaptation problem itself is solved by an iterative
improvement search based on partial order planning.
However, in order to start from a complete plotline and
arrive at a variation with different quests, we employ both
additive and subtractive improvement mechanisms. To the
extent that the player model is an approximation of player
preferences, future work may pair our offline adaptation
technique with online interactive storytelling engines.
As the world orients toward greater on-demand and
customized entertainment experiences, overcoming the
content authoring bottleneck will increasingly require
automation on the level of creative production. We believe
that a partnership between human authors and automated
adaptation can scale up our ability to deliver the “right
story to the right person at the right time.”
References
Chen, S., Nelson, M. and Mateas, M. 2009. Evaluating the
Authorial Leverage of Drama Management. In Proc. of the
5th AI and Interactive Digital Entertainment Conf.
Lebowitz, M. 1987. Planning stories. In Proceedings of the
9th Annual Conference of the Cognitive Science Society.
Muñoz-Avila, H. and Cox, M. 2008. Case-Based Plan
Adaptation: An Analysis and Review. IEEE Intelligent
Systems 23(4).
Penberthy, J. and Weld, D. 1992. UCPOP: A Sound,
Complete, Partial-Order Planner for ADL. In Proc. 3rd
Int’l. Conf. on Knowledge Representation and Reasoning.
Porteous, P. and Cavazza, M. 2009. Controlling Narrative
Generation with Planning Trajectories: The Role of
Constraints. In Proc. of the 2nd Int’l. Conf. on Interactive
Digital Storytelling.
Riedl, M.O., Stern, A., Dini, D., and Alderman, J. 2008.
Dynamic Experience Management in Virtual Worlds for
Entertainment, Education, and Training. International
Transactions on System Science and Applications 3(1).
Riedl, M.O. and Young, R.M. In press. Narrative Planning:
Balancing Plot and Character. Journal of AI Research.
Roberts, D.L and Isbell, C.L. 2008. A Survey and
Qualitative Analysis of Recent Advances in Drama
Management. International Transactions on Systems
Science and Applications 3(1).
Rollings, A. and Adams, E. 2003. Andrew Rollings and
Ernest Adams on Game Design. New Riders.
Saunders, R. and Gero, J. 2004. Curious Agents and
Situated Design Evaluations. AI for Engineering, Design,
Analysis, and Manufacturing 18(2): 153-161.
Seif El-Nasr, M. 2007. Interaction, Narrative, and Drama
Creating an Adaptive Interactive Narrative using
Performance Arts Theories. Interaction Studies 8(2).
Sharma, M., Ontañón, S. Mehta, M., and Ram, A. 2010.
Drama Management and Player Modeling for Interactive
Fiction Games. Computational Intelligence 26(2):183-211.
Thue, D., Bulitko, V., Spetch, M., and Wasylishen, E.
2007. Interactive storytelling: A player modeling approach
Proc. 3rd AI and Interactive Digital Entertainment Conf.
Trabasso, T. and van den Broek, P. 1985. Causal Thinking
and the Representation of Narrative Events. Journal of
Memory and Language 24: 612-630.
Yee, N. 2006. The Demographics, Motivations, and
Derived Experiences of Users of Massively-Multiuser
Online Graphical Environments. PRESENCE:
Teleoperators and Virtual Environments 15(3): 309-329.
Young, R.M. 1999. Notes on the Use of Plan Structures in
the Creation of Interactive Plot. Working Notes of the AAAI
Fall Symposium on Narrative Intelligence.
Young, R. M. and Pollack, M.E 1994. Decomposition and
causality in partial-order planning. In Proc. of the 2
nd
Int’l.
Conf. on AI and Planning Systems.
Table 1. Results of the second study
N2>N1 N2>N0 N1>N0
Plot group 1 52%
88%*
88%*
Plot group 2 76%*
100%*
60%