Top Banner
Effects of Narrative Levels on Comprehension : Theoretical Framework and Methodology Baptiste Campion Université catholique de Louvain (UCL) Groupe de Recherche en Médiation des savoirs (GReMS) Louvain-la-Neuve (Belgium) [email protected] http://www.uclouvain.be/comu/ AISB07 - AI and Narrative Games for Education Newcastle 2nd-5th april 2007
15
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: AISB'07

Effects of Narrative Levelson Comprehension :

Theoretical Framework and Methodology

Baptiste Campion

Université catholique de Louvain (UCL)Groupe de Recherche en Médiation des savoirs (GReMS)

Louvain-la-Neuve (Belgium)

[email protected]://www.uclouvain.be/comu/

AISB07 - AI and Narrative Games for EducationNewcastle 2nd-5th april 2007

Page 2: AISB'07

• Current PhD research led under supervision ofPr Philippe Marion (UCL) and Daniel Peraya(Univ. of Genève)

• Attempt to combine cognitive and narratologicalapproaches for studying interactive narrative

• Object: Science popularization interactivenarrative

• Current presentation: work in progress

General Overview

Page 3: AISB'07

• There exists different uses of narration ineducative interactive documents

• Is the educative ʻeffectʼ similar in each case ?

Whatʼs the matter ?

Interest of dinstinctions between situations

Interest of modelling presumed effects of narrationand testing real effects

Interest both for classic and interactive narratives

Interest for researchers and designers

Page 4: AISB'07

Comprehension is defined followingModel of Comprehension of Van Dijk andKintsch (1983)

Comprehension

Comprehension process: double process(top down and bottom up) of construction ofa coherent representation

I define ʻeducative effectsʼ in terms ofcomprehension.

Page 5: AISB'07

The reader must build a representation ofnarrative referent

• Story Schemata (for ex. Mandler 1984)• Mental Models Theory (Johnson-Laird 1983)• Consistent and coherent with narratology

Understanding narrative

Page 6: AISB'07

Following Herman (2002) narrative comprehensiongoes through the construction (by the reader) of astoryworld

Narrative comprehension

This storyworld is:

•A mental model of ʻwhatʼs going on?ʼ•Set up by bottum-up (microdesign) and top-down(macrodesign) process

Page 7: AISB'07

Use of narrative forcomprehension

Herman (2003):

“My hypothesis is that stories provide, to a degree thatneeds to be determined by future research, domain-general tools for thinking”

→ Storyworld is the base on whichcomprehension of educative matter transmittedthrough a narrative is allowed

Page 8: AISB'07

The ʻlevelsʼ of narrationWe postulate educational use of narrative on at leasttwo levels:

•Surface level (storyworld related to the story)•Deep level (storyworld related to knowledge domain)

→ Different comprehension effects ?

Page 9: AISB'07

• Deep level narrative should lead subjects to builda relatively unified representation

• Surface level narrative should oblige subjects towork with two levels of representation : one forthe story itself and one another for the educativecontent

Research Assumptions

Page 10: AISB'07

Research Assumptions (II)

STORY

A&B mixedStoryworld useful for:•The story•The educative content

A. Storyworld useful forthe story

+B. Representation of

educative content

Page 11: AISB'07

Methodology• Quasi-experimentation• Comparison of representation of a scientific

phenomenon whether explained with a surface levelnarrative or a deep level narrative

• Coherence of representations is observed withquestionnaires : definition question, problem-solvingquestion and drawing the phenomena

• Indicators : relations between elements, specificvocabulary, conjunctions and disjunctions, ability ofabstracting and re-use gathered info

Page 12: AISB'07

Current experimentations• Subjects : 100 children of 5th year elementary school

(+/-11 year old)• Experimental material : 3 versions of the explanation

(short website) of a scientific phenomenon (decayformation)

• Deep level, surface level and control group (non-narrative unfollowing Adam criteria [1996])

• Individual passation

Page 13: AISB'07

Current experimentations (II)Summary of (quasi) experimental design:

Deep levelconditon

Surface levelcondition

Controlgroup

Deep levelnarrative

Surface levelnarrative

Non-narrativeexplanation

Same questionnaire

Identification questionnaire

Page 14: AISB'07

• Forthcoming results• Elements to take in consideration:

Nuances in results Extraction operations Never forget whatʼs the goal of an

educative document: thepresented distinction make nosense without this

Conclusions

Page 15: AISB'07

Thank you for your attention.

Baptiste Campion

[email protected]

Université catholique de LouvainRuelle de la Lanterne Magique 14BE 1348 Louvain-la-Neuve (Belgium)