A SSESSING E YE -T RACKING T ECHNOLOGY FOR L EARNING -S TYLE DETECTION IN A DAPTIVE G AME -B ASED L EARNING Tracey J. Mehigan Ian Pitt IDEAS Research Group,

1

ASSESSING EYE-TRACKING TECHNOLOGY FOR LEARNING-STYLE DETECTION IN ADAPTIVE GAME-BASED LEARNING

Tracey J. Mehigan

Ian Pitt

IDEAS Research Group,

Dept. of Computer Science, UCC

2

INTRODUCTION

Background Adaptive eLearning systems

Learning-styles Felder-Silverman LSM

FSILS Questionnaire

Eye-tracking & eye-tracking technologies Detecting Global / Sequential learners Detecting Visual / Verbal learners

Potential for Mobile GBL Eye-tracking for GBL in mobile environments

3

BACKGROUND

Measurement of learning-styles facilitates the provision of adaptive content to specific learner needs

Data traditionally gathered using questionnaires

Recently, user interaction with learning systems has proven a valuable method of data gathering for learning-styles

Eye-tracking technology could provide a better method of user data collection for learning-style analysis

4

ADAPTIVE LEARNING SYSTEMS

User interaction with learning systems has proven a valuable method of data gathering for learning-styles

A number of studies in recent years have looked at the inference of learning-styles using the Felder-Silverman model

Bayesian Networks (García et al) Behaviour Patterns (Graf and Kinshuk) Feed Forward Neural Networks (Villaverde et al) Mouse movement patterns (Spada et al) Accelerometer Interaction (Mehigan et al)

5

LEARNING-STYLES & PERSONALITY MODELS

One of the most analysed cognitive features in the development of adaptive systems in eLearning environments

Learning-style models classify students according to where they fit on a number of scales depending upon how they process and receive information

Learning Style & Personality ModelsMyer-Briggs Model The Big Five Model

Felder-Silverman LSM

6

FELDER-SILVERMAN LSM

The Felder-Silverman Learning-Style Model (Felder & Silverman 1988) is a widely employed model for inferring individual learning-styles

Based on four learning dimensions:

Active /ReflectiveSensitive /IntuitiveGlobal / SequentialVisual / Verbal

7

LEARNING STYLES QUESTIONNAIRE

HTTP://WWW.ENGR.NCSU.EDU/LEARNINGSTYLES/ILSWEB.HTML

8

EYE –TRACKING

“Eye-tracking works by reflecting invisible infrared light onto an eye, recording the reflection pattern with a sensor system, and then calculating the exact point of gaze using a geometrical model. Once the point of gaze is determined, it can be visualized and shown on a computer monitor”

(Tobii.com)

9

EYE-TRACKING RESEARCH

Eye tracking technology is widely used in many disciplines

Commerce, learning difficulties, etc.

Studies have been conducted based on HCI, visual cognition and web accessibility

Research has explored saccade velocity, blink rate and the degree of eyelid openness for determination of user’s tiredness level

This can complement other information gained by the system through user behavioural patterns

Few examples of eye-tracking-based work in eLearning and subsequently, game based learning

However, researchers at a UK university are developing computer games which can be controlled by eye movements

10

TOBII EYE-TRACKING TECHNOLOGY

Calibrate for the user’s vision

System records the user’s eye movements while s/he observes the test scene(s)

Analyse the data on individual and / or group level

Gaze plotsHeat maps Statistics

Can be based on Areas of Interest (AOI’s)

11

DETECTING GLOBAL / SEQUENTIAL LEARNERS

Sequential learners would show a slower vertical speed of eye movement between fixation points and a longer focus time than their Global counterparts

10 Participants selected from Dept of Computer Science UCC – reflecting a balanced sample of Global and Sequential learners

2 Screens Presented to the User A Learning Screen A Task Screen

To ensure that the user gaze is following the screen content and not the cursor arrow the mouse is only used to move between screens.

12

GLOBAL & SEQUENTIAL LEARNERSGAZE PATTERNS & HEAT MAPS

Global Sequential

13

DETECTING VISUAL & VERBAL LEARNERS

Visual learners (as defined in terms of the Felder-Silverman LSM) will exhibit longer total time (fixation) duration on visual learning content (images/graphics)

Verbal learners will exhibit longer total time (fixation) duration on textual learning content

10 Participants selected from Dept of Computer Science UCC – reflecting a balanced sample of Visual and Verbal learners

2 Screens Presented to the User A learning screen A task screen

To ensure that the user gaze is following the screen content and not the cursor arrow the mouse is only used to move between screens.

14

VISUAL & VERBAL LEARNERSGAZE PATTERNS & HEAT MAPS

Visual Verbal

15

Mobile learning decreases limitation of traditional learning systems through the mobility of portable devices

The incorporation of mobile devices provides an opportunity for ubiquity and collaboration in education

The importance of the mobile phone to teenage identity and the development of social friendship networks facilitates incorporation of mLearning

Mobile devices could be used to encourage young people to learn in a beneficial way

MOBILE LEARNING

MOBILE GBL Mobile learning decreases limitation of learning location with the mobility of general portable devices

Mobile GBL is usually Situated and Ubiquitous

“We see a potential move away from immersive, game based learning, represented in traditional eLearning simulation systems, toward the advancement of situated mLearning environments” (Dede)

Game based learning can also improve learning in specialist areas where students become engaged in a situated learning environment which occurs within the game (Williamson et al )

Mobile GBL can also be incorporated in blended and other learning environments

16

16

EYE-TRACKING FOR MOBILE GBL

17

Potential to track user learning-styles through avatar movement via eye-based interaction

The provision of adaptive systems based on the Felder-Silverman model could potentially offer students increased motivation to learn through personalised content

Felder-Silverman’s model provides matching teaching-styles to each learning-styles and therefore content can be specifically tailored to the needs of each individual learner

17

18

EYE-TRACKING MOBILE GBL ENVIRONMENTS

Eye-trackers are becoming smaller and less cumbersome, potentially offering a new method of mobile device interaction

Recently ‘Gaze Gesturing’ has emerged as a means of controlling device interaction (Drewes et al)

Tobii ‘Glasses’ offer the next generation of mobile eye-tracker

Recent work conducted by Miluzzo et al facilitates the use of forward facing mobile device cameras for eye-tracing purposes

19

CONCLUSION

Background Adaptive eLearning systems

Learning-styles Felder-Silverman LSM

FSILS Questionnaire

Eye-tracking & eye-tracking technologies Detecting Global / Sequential learners Detecting Visual / Verbal learners

Potential for Mobile GBL Eye-tracking for GBL in mobile environments

QUESTIONS??

COMMENTS!!

20