Evidence-Based Principles for
Selecting Eye Gaze AAC
Technology
Katya Hill, P.h.D., CCC-SLP
School of Health and Rehabilitation Sciences
Department of Communication Science and Disorders
Department of Rehabilitation Sciences and Technology
AAC Institute
Overview on session: Evidence-Based Principles
for Selecting Eye Gaze AAC Technology
• Focus on principles and evidence
• Identify candidates for eye gaze
• Identify features of eye gaze systems
• Discuss collecting clinical & personal evidence
2
Evidence-Based Principles for Selecting Eye Gaze AAC Technology
Focus on principles and evidence to
guide clinical decisions for selecting
eye gaze technology
3
Remote Eye Gaze Technology
• Vision-controlled direct selection technique
• More than bright and dark pupil tracking
4
The important details
• pupil-center/corneal-reflection method
• LED reflects a small bit of light off the surface of the
eye’s cornea
• The light reflects off of the retina and causes the pupil
to appear white
• Pupil effects are methods used to determine the
gazepoint
5
Viewing the Eye
6
Glint-Pupil Vector
7
How our eyes work
• Eye movement while reading
• Saccades and fixations
• Visual perception and language processing
• Errors: losing place, skipping words or not adequately
processing visual information
• Poor readers: language skills vs. eye movement control
8
http://www.visionforlearning.co.nz
9
10
11
12
Evidence-Based Principles for Selecting Eye Gaze AAC Technology
IDENTIFYING CANDIDATES
APPROPRIATE FOR EYE GAZE AS A
SELECTION TECHNIQUE.
13
Overview of clinical populations
• Amyotrophic Lateral Sclerosis (ALS)
• Traumatic Brain Injury (TBI)
• Cerebral Palsy
• Multiple Sclerosis (MS)
• Muscular Dystrophy (MD), spinal muscular
atrophy, Werdnig-Hoffman Syndrome
• Spinal cord injuries
• Strokes
14
User considerations: Vision
• Good control of at
least one eye.
• Common eye
movement problems
– Nystagmus
– Alternating strabismus
15
User considerations: Vision
• Adequate vision
• Corrected with glasses
• Inadequate visual
acuity
• Soft contact lenses
• Diplopia
• Blurred vision
• Cataracts
• Homonymous
hemianopsia
16
User consideration: Vision
• Absence of side effects
from medication
• Baclofen
• Anticonvulsants
• Antidepressants
17
Physiology Affecting Eye Tracking
Top image: ptosis
(drooping) eyelid
Bottom image: early
cataracts visible over
both pupils
18
User considerations: Physical
• Positioning
– Ability to maintain a
position in front of the
eye gaze screen/camera
• Continuous,
uncontrolled head
movement
19
User considerations
• Cognition
• Memory
• Language
20
Evidence-Based Principles for Selecting Eye Gaze AAC Technology
THE DESIGN AND FEATURES OF EYE
GAZE SYSTEMS
21
Primary components
Language
representation methods
Vocabulary
Method of Utterance
Generation
Secondary components
User Interface
Selection method
Outputs
Tertiary components
Hill & Scherer, 2008; Hill, 2010 in press
22
Examination of eye gaze technology as a selection
technique
• Infrared exposure
• Calibration & defaults
• Gaze point predictions
• Dark versus light pupils
• 1 versus 2 cameras
• Narrow versus wide angle camera lens
• Sitting or reclining
23
Camera Moveability & Position
24
Flexible cameras allow unique client positions
Another unique position
On-screen eye image display
• Eye images are
displayed for the user
• Decreases frustration:
user knows what the
camera sees. This is
not a feature on many
systems.
27
Environmental issues affecting
eye tracking
• Figure on left: light from window reflecting on
eyeglass lens
• Figure on right: reflection eliminated by changing
angle of earpieces
28
User interface considerations
• Color and detail
• Alphabet displays
• Graphic symbol displays
• Navigation
• automaticity
29
Examples of user interfaces
30
Examples of user interfaces
31
Examples of user interfaces
32
Comparing Eye Tracking Systems
33
Evidence-Based Principles for Selecting Eye Gaze AAC Technology
PROCEDURAL PROTOCOLS FOR EVALUATING
EYE GAZE SYSTEMS – PERFORMANCE AND
OUTCOMES DATA
34
Computer-based evaluations and
LAM datalogging
35
COMPASS – Computer Access
Assessment Software
• Assessment tool for computer access & AAC
• Eight skill tests
• Skills assessed include:
– keyboard and mouse use
– navigation through menus
– switch use.
36
Copy Spell & Core Word Sentences
• Examples:
– I want to think about that.
– Why do you think that?
– He helped her with that.
– I can do this myself.
– Do you feel cold?
– She likes to be happy
– Give it to me.
37
Self-Created Utterances/
Language Sample Context
• Examples:
– Picture Description
– Story Retell
– Narratives
– Conversation
38
Performance Measures
• Communication Rates
– Ave. communication rate (wpm)
– Peak communication rate (wpm)
• Rate by language representation method
• Selection rate (bps)
• Rate index (wpb)
• Accuracy
39
Video clip
40
Know your numbers
41
Know your numbers
Analysis of LAM data from ECO; Unity 144;
Subject: K. Hill 26 January 2009 Task: Copy Core Sentences
Keyboard:
ECOPoint
right index
Acceptance time (sec)
Comm. Rate (average)
Comm. Rate (peak)
Fastest Utterance
Comm. Rate (SEM)
Comm. Rate (SPE)
Selection Rate (bits/s)
Errors (%)
32.6
60.0
0.7
13.8
30.0
Where would you
like to go
I want to go
there
36.5
20.9
10.8
0
11.3
6.7
13
5
42
User Satisfaction - Outcomes
43
Quest & PIADs
QUEST
PIADs
• Self-administered 12item questionaire
• 26-item, self-rating
questionnaire
• Evaluates user
satisfaction
• Yields three scores:
– Device,
– Services,
– total QUEST
• Describes user experiences
along three dimensions:
– Competence: Measures feelings
of competence and usefulness.
– Adaptability: Signifies a
willingness to try new things.
– Self-esteem: Indicates feelings of
emotional wellbeing and
happiness
44
Surveying client preferences and
satisfactions – outcomes data
Rating Scale:
SA = strongly agree; A=agree; N=no opinion, D=disagree; SD-strongly disagree
SD
D
N
S SA
I use multiple methods to communicate with family
and care givers
My most effective method of communication is my eye
gaze system
I have no problems with calibration of my system
I use my eye gaze system to write and save
messages/letters
I am very satisfied with my accuracy using my system
45
Authors using eye gaze systems
46
External evidence sources
•
Text entry by gaze: Utilizing eye tracking. In I. S.
Mackenzie & K. Tanaka-Ishii (Eds.)
•
COGAIN: Communication by Gaze Interaction
at http://www.cogain.org/
•
Additional References
47
Conclusions and Wrap-Up
• Safety and protection of clients
• Identify evidence-based marketing
• Need for data-driven decisions
• Suggest an ASHA technical paper
48
Questions and Discussion
49
Thank you!
Katya Hill, Ph.D., CCC-SLP
Associate Professor
AAC Performance and Testing Lab
6017 Forbes Tower
University of Pittsburgh
Pittsburgh, PA 15260 US
www.aacinstitute.org
412-383-6659
[email protected]
50
References
Anson, D., George, S., Galup, R., Shea, B., & Vetter, R. (2001). Efficacy of the Chubon vs. the QWERTY keyboard. Assist
Technol, 13(1), 40-45.
Anson, D., Moist, P., Przywara, M., Wells, H., Saylor, H., & Maxime, H. (2006). The effects of word completion and word
prediction on typing rates using on-screen keyboards. Assist Technol, 18, 146-154.
Bates, R. E. A. (2006). Enhancing the performance of eye and head mice: A validated assessment method and an
investigation into the performance of eye and head based assistive technology pointing devices. Unpublished PhD
thesis, DeMontfort University, Leicester, UK.
Biscaldi M, Fischer B, Hartnegg K. (2000). Voluntary saccadic control in dyslexia. Perception, 29, 509-521.
Calvo, A., Chio, A., Castellina, E., Corno, F., Farinetti, L., Ghiglione, P., et al. (2008). Eye tracking impact on quality of life of
ALS patients. Paper presented at the Computers Helping People with Special Needs, Berlin.
Cooper, R., Hill, K., Koester, H., & Spaeth, D. (2004) Advances in data logging across assistive technology platforms. Paper
presented at the 2004 RESNA Annual Conference. Atlanta, GA. June, 2004.
Day, H., & Jutai, J. (1996). Measuring the psychosocial impact of assistive devices: The PIADS. Canadian Journal of
Rehabilitation, 9(2), 159-168.
Demers, L., Weiss-Lambrou, R., & Ska, B. (1996). Development of the Quebec user evaluation of satisfaction with assistive
technology (QUEST). Assist Technol, 8(1), 1-13.
Feng, J., Karat, C.-M., & Sears, A. (2005). How productivity improves in hands-free continuous dictation tasks: lessons learned
from a longitudinal study. Interacting With Computers, 17(3), 265-289.
Fischer B,Hartnegg K. Auditory, Visual, and Optomotor Develomental Deficits, and Training in Dyslexia. In: Trends in Dyslexia
Research, 2005 Chapter 1, pp 1-33.
51
References, cont.
Hansen, J. P., Torning, K., Johansen, A. S., Itoh, K., & Aoki, H. (2004). Gaze typing compared with input by head and hand.
Paper presented at the ACM Symposium on Eye Tracking Research and Applications, San Antonio, TX.
Hill, K. (2004). AAC evidence-based practice and language activity monitoring. Topics in Language Disorders: Language and
Augmented Communication, 24, 18-30.
Hill, K., & Romich, B. (2001). A language activity monitor for supporting AAC evidence-based clinical practice. Assistive
Technology, 13, 12-22.
Hill, K., Romich, B., & Spaeth, D. (2000). AAC selection rate measurement: Tools and methods for clinical use. In
Proceedings of the RESNA '00 Annual Conference. Orlando, FL. July. pp 61-63
Itoh, K., Aoki, H., & Hansen, J. P. (2006). A comparative study of two Japanese gaze typing systems. Paper presented at the
ACM Symposium on Eye Tracking Research and Applications, ETRA 2006, San Diego, CA.
Koester, H.H. and McMillan, W.W. (1998). Usability Testing of Software for Assessing Computer Usage Skills, Proceedings of
the RESNA '98 Annual Conference, Minneapolis, Arlington, VA: RESNA Press.
Lin, A., Weaver, J., Aedo, L. (2009). Eye tracking technology: Overview of current systems. Presentation at the annual
CSUN conference. Los Angeles, CA: March 16-21, 2009.
MacKenzie, I. S., & Soukoreff, R. W. (2002). Text entry for mobile computing: Models and methods, theory and practice.
Human-Computer Interaction, 17, 147-198.
52
References, cont.
Majaranta, P. & Räihä, K. (2007). Text entry by gaze: Utilizing eye tracking. In I. S. Mackenzie & K. Tanaka-Ishii (Eds.),
Text entry systems: Mobility, accessibility, universality (pp. 175-187). New York: Morgan Kaufmann Publishers.
Pannasch, S., Helmert, J. R., Malischke, S., Storch, A., & Velichkovsky, B. M. (2008). Eye typing in application: A
comparison of two systems in ALS patients. J. Eye Movement Research, 2(4), 1-8.
Stein JF. (2000). Monocular occlusion can improve binocular control and reading in dyslexics. Brain, 123, 164-170.
Wobbrock, J. O., Rubinstein, J., Sawyer, M. W., & Duchowski, A. T. (2008). Longitudinal evaluation of discrete
consecutive gaze gestures for text entry. Paper presented at the ACM Symposium on Eye Tracking Research and
Applications, ETRA 2008, Savannah, GA.
53