SceneDriver: The development and evaluation of an interactive narrative system for children.

SceneDriver: The development and evaluation of an interactive narrative system for children.

Project

• The project was supported by EPSRC and DTI

• Open University– Annika Wolff– Paul Mulholland– Zdenek Zdrahal

• Peppers Ghost Production Company• Edinburgh Virtual Environment Centre

Introduction

• Animated Television series are expensive to produce

• Need to produce extra merchandise• Computer games based on the

television series are a popular option.• Some games based on children’s

television series have viewing galleries• Peppers Ghost who produced Tiny

planets have used this approach

Introduction to Tiny Planets

Tiny Planets

Bing and BongThe White SofaSix Tiny Planets

Introduction

• The aim of this project was to develop an interactive narrative system called SceneDriver.

• SceneDriver extends the viewing gallery approach by using narrative principles to enable children to create novel stories whilst using Tiny Planet Content.

• A further aim was to develop a suitably intuitive interface for use by children.

Potential Approaches

Content

Canned Generated

Child Passive

Active

TV Shows

SceneDriver

Dynamically generated TV

Future …..

Related Work

• Interactive Narrative systems fall into two categories– The first use narrative principles to

organise and present a collection of resources in a coherent way • Murtaugh’s Automatist storytelling system

dynamically orders keyword annotated clips • Rocchi and Zancanaro’s work used

directorial techniques for presenting of narratively structured material.

Related Work

– The second the systems create a dramatic experience with a protagonist and plot structure. • Mateas and Stern’s interactive drama uses pre

authored story fragments in conjunction with autonomous animated characters and allows interaction during the course of the story

• Hayes-Roth et al., devised a model of interactive narrative called directed improvisation.

• Gorbet. Orth and Ishi developed a tangible interface for interacting with narrative.

TinyPlanet Analysis

• We devised a plot description both on this analysis and also on narrative theory.

• It is consistent with existing with structuralist theories of narrative that have been used to interpret written stories.

• The analysis suggested that each episode could be viewed from different levels.–  plot level– directorial level

Plot Level Theme

Introduction

Partial Success SuccessFailure

ResolutionAttempt

ConflictIntroduction

Post-completionevent

Directorial Level

• The directorial level is the level at which events occur to provide dramatic effect or entertainment value

SceneDriver: Playing The Game

• A child constructs a narrative by playing a form of dominos.

• The left-hand side has the characters who were in the previous clip.

• The right hand side specifies the characters who will appear in the next scene.

SceneDriver: Playing The Game

• This interface enables the child to manipulate the narrative,

• A scene supervisor module ensures the animation adheres to the principles narrative.

• It also ensures coherent transitions from one scene to another by the use of transitional scenes.

• There are two distinct options– Complete Explicit– Rewrite

SceneDriver: Complete Explicit

• In a complete explicit, the left hand side of the domino has the characters who were present in the previous clip.

• The right hand side then determines which characters are present in the next scene.

Complete Explicit Version

SceneDriver: Rewrite

• In the rewrite rule game the left hand side and the right hand side have a different meaning to that of the complete explicit game.

• In the rewrite rule which ever characters are depicted in the left hand side of the tile are to be removed and then replaced in the next scene with the characters in the right hand side of the tile.

Rewrite Version

Study 1 Aims

• To investigate whether children can use the domino interface to direct a narrative.

• To investigate children understanding of the interactivity narrative.

• To investigate how much the children enjoy the interactive narrative.

• Forty children (20 girls and 20 boys), twenty to each condition. Aged 7-8 years in Year 3

• Two Conditions– Complete Explicit– Rewrite

• Understanding measured by recall.• Enjoyment of the game was measured

using the ‘sticky ladder’ measure and a ‘smiley-face’ scale.

Study 1: Method

Self Report Measures

• Sticky Ladder

Self Report Measures

• Smiley Faces

Errors and Help

• Number of Errors made• Verbal help provided by the

experimenter• Physical help provided by the

experimenter• Verbal prompting by experimenter• Clicking tile before the end of the clip • Thinking which tile to click before the

end of the clip

Observational Measures

• Laughing• Clapping• Talking about the characters• Talking about clips• Making sounds from clip• Recall task• Commenting on the television series• Commenting on the game being played

Study 1: Results

Interactive Narrative Systems

Rewrite Complete

M SD M SD

smiley 4.60 0.68 4.40 0.68

sticky 5.60 1.67 5.35 1.35

Study 1: Results

• No significant difference between the two conditions in terms of– Smiley Face Scale and Sticky Ladder – Observational measures.– Recall measure

• Children in the Rewrite condition received more verbal help

• They also received significantly more prompting

Relationships between the Measures

- Verbal expressions of enjoyment (the relationship between laughing, talking about the characters, talking about the clips, making sounds, the recall task, commenting on the series and commenting on the game)

- Non-verbal relationships (clapping was found to have a strong relationship with smiling)

- Errors and help - strong relationship between prompting and verbal assistance.

Study 1: Summary

• No significant differences were found between the two conditions regarding enjoyment

• No significant differences were found between the two conditions regarding understanding

• However, regarding the usability – the help and errors – significant differences were found regarding verbal assistance and prompting.

Study 2: Aims

• Study 1 looked at the different games in terms of how much the children enjoy playing the games and understanding.

• The measure of understanding was too crude we therefore decided to explore that further with a more sensitive measure

• Study 2 investigated children’s undertsanding and enjoyment of an interactive narrative system

Study 2: Method

• Forty Eight children (24 girls and 24 boys), 16 children in each condition.

• Aged 7-8 years in Year 3• Three conditions

– Complete Explicit with Narrative– Complete Explicit without Narrative– Domino game

• Enjoyment of the game was measured using the ‘sticky ladder’ measure.

Structured Recall Task

• What can you remember• Memory probes

– What did these characters do– What happened next– Identity Parade. Which of these

characters were in the video.

Study 2: Results

• We have only just completed this study so the results are still being analysed.– There was no difference between the

conditions in terms of the number of characters CORRECTLY identified in the identity parade

– There was a difference between the conditions in terms of the number of characters INCORRECTLY identified in the identity parade.

Discussion

• The aims of the studies where– To investigate how much the children

enjoy the interactive narrative.– To investigate children understanding of

the interactivity narrative.– To investigate whether children can use

the domino interface to direct a narrative.

Project

• The project was supported by ESPRC and DTI

• Open University– Annika Wolff– Paul Mulholland– Zdenek Zdrahal

• Peppers Ghost Production Company

• Edinburgh Virtual Environment Centre

• University of Bath– Anna Reeves– Nicole Pieper

• Staff and children at Old field Park Junior School