The Journal of Technology, Learning, and Assessment

Volume 5, Number 3 · December 2006

Individualizing Learning
Using Intelligent
Technology and Universally
Designed Curriculum

Michael Abell

www.jtla.org
A publication of the Technology and Assessment Study Collaborative
Caroline A. & Peter S. Lynch School of Education, Boston College
 Volume 5, Number 3

Individualizing Learning Using Intelligent Technology and

Universally Designed Curriculum

Michael Abell

Editor: Michael Russell

[email protected]
Technology and Assessment Study Collaborative
Lynch School of Education, Boston College
Chestnut Hill, MA 02467

Copy Editor: Kevon R. Tucker-Seeley

Design: Thomas Hoffmann
Layout: Aimee Levy

JTLA is a free on-line journal, published by the Technology and Assessment Study
Collaborative, Caroline A. & Peter S. Lynch School of Education, Boston College.
Copyright ©2006 by the Journal of Technology, Learning, and Assessment
(ISSN 1540-2525).
Permission is hereby granted to copy any article provided that the Journal of Technology,
Learning, and Assessment is credited and copies are not sold.

Preferred citation:
Abell, M. (2006). Individualizing learning using intelligent technology
and universally designed curriculum. Journal of Technology, Learning,
and Assessment, 5(3). Retrieved [date] from http://www.jtla.org.
Abstract:
The American education system and its rigorous accountability and performance stan-
dards continually force educators to explore new ways to increase student achievement.
The improvement in computer technology and intelligent computing systems may offer
new tools for student learning and higher academic achievement. These systems have the
potential to meet individual student learning needs using universally designed curricula
and assessments. The purpose of this paper is to present a conceptual framework that
harnesses the potential of intelligent learning systems, machine learning models, and
universal design for learning principles to help formulate next generation instructional
materials. By using intelligent and interactive curricula, educators could begin to move
away from information disseminator into a facilitator of the learning experience.
Individualizing Learning Using
Intelligent Technology and
Universally Designed Curriculum

Michael Abell
University of Louisville

Introduction
On a national level, increased accountability standards continue to
challenge educators to explore new methods to increase achievement.
Cuban (1993) noted the reform efforts of the 1980s and 1990s sought
to incorporate advanced technologies into schools to improve self-
directed learning and active engagement. A significant goal for students
is to become self-directed, thoughtful, and independent learners while
educators make teaching and learning more productive. Efficiency and
productivity are often subtly intertwined with reform efforts that look to
maximize learning according to Chester (2002). However, efficiency and
productivity though should not come at the cost of quality and academic
rigor.
Meyer and Rose (2000) espouse the concept of Universal Design for
Learning. Universal Design for Learning is a theoretical framework that
guides the development of curricula that meets the needs of all students
(Rose & Meyer, 2002). One component of this new paradigm focuses on the
development of accessible interactive curriculum materials. These mate-
rials would engage the learner in new and empowering ways that align to
their unique approach to learning. These learning materials can be altered
and scaffolded based on learner needs and cognitive style. McKenzie (2000)
defines instructional scaffolding as curriculum and instruction that con-
tains six characteristics: 1) Provides clear direction and reduces students’
confusion; 2) Clarifies purpose by helping students understand why they
are doing the work and why it is important; 3) Keeps students on task
by providing structure and clear pathways to learning. Students can make
decisions about which path to choose or what things to explore along the
path but they cannot wander off of the path, which is the designated task;
4) Clarifies expectations and incorporates assessment and feedback using
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.


models of exemplary work, rubrics, and superior student work samples;

5) Points students to worthy sources that reduce confusion, frustration,
and time and offers them choices; and 6) Reduces uncertainty, surprise,
and disappointment by offering multiple routes to success.
The new scaffolded materials leverage the power of technology blended
with research based curriculum aligned to state and national education
standards. Other examples of these interactive materials include curric-
ulum units with embedded digital text, hyperlinked glossaries, concept
maps, graphics, audio, video, and virtual reality. These materials would
help meet broader student ability levels while engaging students based on
their own approach to learning. The traditional printed textbook does not
offer teachers and students the flexibility needed to meet diverse learning
needs when compared to universally designed digital curriculum (Rose
& Meyer, 2002). Behrmann (2001) advocates the use of instructional
curriculum and materials in electronic format. The advantage to this
approach is not the curriculum itself but rather the way the curriculum can
be digitally displayed and altered through scaffolding and customization to
meet the individual student’s preferred approach to learning and unique
ability level. When using digital versions of curriculum, students can
display and engage the content in ways not possible with printed text.
This would be beneficial in a larger intelligent learning system. Students
would engage the learning process utilizing scaffolded audio, video,
graphics, and text using personalized settings built into the curriculum
which offers millions of potential learning variations. These digital versions
could improve productivity and efficiency in schools, though more signifi-
cant uses may be found in reducing barriers to learning and increasing
personalization. Digital versions also reduce the cost and minimize the
time consuming adaptations previously done by teachers. By placing
the digital curriculum and accompanying tools such as text-reader soft-
ware in the hands of students, educators give the controls to those who
understand their approach to learning best: the students themselves. So
how can researchers and curriculum developers begin to construct more
intelligent and interactive curriculum materials?

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.


A Model Guided by Learning Styles and

Emerging Digital Media
Systems of any nature require grounding in appropriate theory and
logic. Intelligent learning systems and accompanying digital media also
require such grounding in the area of learning principles. Combined with
the growth of digital media (e.g. streaming video, websites, audio, Blogs,
etc.) students and teachers are now offered new types of information and
learning resources that can be integrated directly into daily instructional
activities leveraging computer and internet based technology.
Intelligent learning systems of the 21st century could provide students
of all ability levels new ways to personalize and engage the learning process.
This would avail students of the growing array of digital content now avail-
able. Learning resources have grown to include easy access to on demand
streaming video in classrooms (United Streaming, 2005). Other resources
include the growing availability of digital text from major publishers
(Association of American Publishers, 2002), online assessments aligned to
state accountability standards (Kentucky Department of Education, 2004),
and informal teacher created assessments allowing immediate classroom
feedback (Qwizdom®, 2004) which could be incorporated into intelligent
learning systems. Assistance would be needed to guide students through a
process that would create personalized learning profiles.
Currently, a major stumbling block is the lack of intelligent guides, also
referred to as agents (Seaward, 1998), that can manage and coordinate
the flow of learning materials based on unique student learning prefer-
ences. Bork (1997) noted the low quality of individualization in student
learning and that computer-based learning systems do the same. He con-
tinues by encouraging more research on advanced algorithms leading to
interactive computer environments that understand student needs. This
understanding could result in highly interactive learning and improved
engagement. A learning environment such as this should emulate the
unique learning style of the individual student. Seaward (1998) explored
the use of computer based digital assistants (e.g., icon or animation repre-
senting a participant) to facilitate learning. Specific analysis focused on the
use of digital assistants built into Microsoft Word and Office. These assis-
tants facilitated the improved utilization of the software by new users.
Results revealed that fewer technical support calls were made once the
assistants were introduced resulting in lower support costs. The assistants
acted as a guide by asking questions and directing new users to the neces-
sary features they were unable to locate. This same premise could be used
by students as a way to interact with the curriculum. Students would pre-
select learning style preferences. Then the computer and accompanying

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.


databases would present interactive curriculum or learning material in the

most opportune format for that student. These materials would all be built
from the unique learning style of the student.

Importance of Unique Learning Styles

Baldwin and Sabry (2003) espouse the importance of learning style
integration into interactive learning systems. Interactivity is important to
adaptive learning systems. The interactivity allows students to engage their
unique learning style based on intrinsic learning preferences. These pref-
erences would help guide the intelligent learning system to individualize
the curriculum and assessments to the individual student level. Baldwin
and Sabry also expand on the importance of digital presentation methods
found in interactive learning systems. These methods allow engagement
across a wide variety of learning styles using multi-media (e.g., video, text,
graphics, audio) technologies by allowing students to engage their stron-
gest learning modality. Learning styles encompass a broad range of abilities
and would be a vital component to the intelligent interactive curriculum.
Felder (1993) presents a model of learning style dimensions that theoreti-
cally could be presented by a digital assistant (Seaward, 1998). The digital
assistant would present various questions to determine the student’s
predominant learning style. This learning style assessment would repre-
sent the first component of an intelligent learning environment based on
individual learning needs. This approach is not without inherent weak-
nesses though.
The challenge lies in what method the intelligent learning agent would
use to initially process computational variables that are attributed to the
way a person learns best. A simplistic method would use levels of shear
cognitive or creative abilities determined by diagnostic tests. This would
not allow for significant variability across individual’s learning differences.
It might also limit the dimension by which the human brain learns best
and ways to categorize this. As a first step in the process, a learning style
inventory could provide a more practical yet quantifiable measure (Table
1, next page). This information would then be transferred into intelli-
gent learning systems and could serve as the foundational measure for
complex algorithms that in the future would predict the most effective
types of content presentation methods pertinent to the student’s learning
style. Nevertheless, learning style inventories have not held up flawlessly
under the rigors of research (Atkinson, 1991, Duff, 2004) even though
they may engage students and promote multidimensional learning experi-
ences. Learning style inventories encompass a broad range of preferences
and modalities. Some of these learning modalities are captured by Felder
(1993) and aligned to student learning preferences and processing styles
as seen in the following table.
J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.


Table 1: Learning-style Dimensions (modified from Felder, 1993)

Dimensions Categorization Preferences

Active–reflective Information processing: Likes trying things, discussion, application
through active and interactive of things learned. Difficulty attending to
engagement in physical activity lectures.
or discussion.
Active–reflective Information processing: Prefers to think about what they learn
through introspection quietly first. Prefer working alone.
Difficulty sitting in lectures without
chance to reflect on what
has been learned.
Sensing–Intuitive Perception of information: Tends to like learning facts, solving prob-
sights, sounds, physical sensation. lems using familiar and well-established
methods. Dislikes complications, surprises,
to be tested on material that has not been
fully covered in class. Tends to be patient
with details and good at memorizing facts
and doing hands-on (laboratory) work.
Tends to be more practical and careful.
Sensing–Intuitive Perception of information: Likes innovation and prefers discovery-
memories, ideas, insights. based approaches, finding relationships,
dislikes repetition and impatient with
details. Good at grasping new concepts
and usually comfortable with abstractions
and mathematical formulations. Tends to
work faster. Gets bored with courses that
involve a lot of memorization and rote
learning
Visual–verbal Perception of sensory information: Tends to remember best what they see:
pictures, diagrams, graphs, static pictures (e.g. diagrams) or dynamic
demonstration. pictures (e.g. videos, DVDs).
Visual–verbal Perception of sensory information: Tends to get more out of words (written
sounds, written, spoken words, and spoken explanations).
formulas.
Sequential–global Progress towards understanding: Tends to gain understanding/find
in logical and small incremental solutions in linear manner, with steps
steps following each other logically. Sequential
learners may not fully understand the
material or establish a link with other
parts, but able to know about specific
aspects of a subject.
Sequential–global Progress toward understanding: Tends to learn in large jumps, absorb
in non-linear way, large jumps, material almost randomly, and may be
holistically able to solve complex problems quickly.
Strongly global learners may be fuzzy
about details or have serious difficulties
understanding until they have the big
picture.

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.


The Digital Learning Environment

The interface, or look and feel an intelligent learning system has, is
important to how a student or teacher will engage with it. Students and
teachers may benefit from an interface where controls are easily under-
stood, customizable based on student needs, easily accessible, and even
available on-line. So what might such a system look like?
The actual interactive curriculum materials presented in a system
to students is only one part of a broader digital learning environment
though. The composition of the environment itself is very important. One
can look at examples of rudimentary intelligent learning environments
presented in various web-based formats. Current models center on the
customization of user profiles exemplified by such websites as Amazon.
com or My.yahoo.com. These interactive environments provide extensive
customization features based on individual profiles and defined param-
eters set forth by the user. The differentiating factor between these sites
and higher order intelligent learning systems is cognitive diagnosis and
adaptive remediation (Shute & Psoka, 1994 & Psoka, 1994). Wasson
(1997) outlined a number of important factors impacting advanced
educational technologies that could influence the diagnosis and remedia-
tion of student learning needs. These factors include artificial intelligence
(AI) routines involving planning, qualitative reasoning, plan recognition,
expert systems, and knowledge representation and management. These
components give the universally designed digital curriculum the ability to
dynamically configure specific student centered instruction and remedia-
tion. In addition, a variety of student data (e.g. standardized and informal)
could be ported into the system to provide various levels (low to high
customization) of student centered content and feedback.
The environmental interface could be structured similar to an online
portal with select components being downloadable. The interface would
have extensive customization and intelligent learning features based on
student learning preferences and instructional needs. A model of the dis-
criminate levels of the system (Table 2, next page) adds further depth
and customization to the intelligent learning system environment. This
is accomplished by allowing district, building, and teacher level content
to be added to the student’s personal profile. This would then allow stu-
dents to access state mandated content or assessment materials in a way
that corresponds to the student’s unique learning style while also meeting
instructional goals. Researchers may consider the following framework
for intelligent student learning environments that offer specific levels and
interface components based on the educational organization’s need.

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
10

Table 2: Intelligent Student Learning Environment Interface Levels

Access Levels Description Components

District Content available from district • System Administration Controls
level sources & programs. • E-learning components
System managed as a district • Digital curriculum & assessment tools
wide intelligent learning • Attendance records
portal. • File storage
School Content is school specific • Licensed digital curricula
targeting learning goals and • Online assessment system
outcome objectives. • School learning goals
• School wide assessment results
• Software tools & databases
Classroom Classroom level instructional • Lesson plans
material identified, controlled, • Online classroom assessments
or created by teacher. • Instructional content
• Online resources
• Intelligent learning & assessment tools
• Student skills/interests inventories
• Class grades
Student Student created content • Productivity tools (word processors,
including assignments, assess- spreadsheets, textreaders)
ment results offering intelli- • Databases
gent monitoring of curriculum • Student grades
progress, and test results. • Assignment and project files
• Individualized digital curriculum
• Email
• Digital content controls (background color,
text size, audio/visual features,
content placement)

Intelligent student learning environments are linked inherently to the

needs of the student and teacher and must not be forgotten. The interac-
tion between the teacher and student plays a critical role in the learning
process according to Wasson (1997). By developing intelligent learning
tools (Emiliani & Stephanidis, 2005), the teacher could begin to move away
from information disseminator to an information organizer and facili-
tator of the learning experience. More time would be dedicated to guiding,
modeling, tutoring, and organizing material using intelligent computing
as a tool for teachers and students. Intelligent computing could allow
more dynamic learning in class and outside of the regular school day. By
offering individualized interactive curriculum that is rigorous and aligned
to state standards, more non-traditional learning sites could be offered in
the community or home setting. Students who use an intelligent learning
environment could utilize less powerful but portable handheld computers
(Abell, Bauder, Simmons & Sharon, 2003) to access learning tools and the
general curriculum away from school or when a computer isn’t available.
This would be accomplished through a consistent yet dynamically changing
user environment.
J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
11

Constructing the Environment

How would such an environment be constructed or customized? The
environment would need to allow students and teachers access to the type
of information (e.g. curriculum, learning tools and exemplars, etc) that
meet their needs and cognitive level of students while challenging and
guiding them through the learning process. Complex relational databases
containing students’ unique learning profiles (e.g. preferences) aligned to
specific content level material would then begin to construct the learning
environment and accompany activities.
The digital learning environment and accompanying interactive
curriculum materials could be offered to students using complex algo-
rithms to cull appropriate student specific content. This might be offered
to students through an online portal with accompanying wireless connec-
tivity or through synchronization of curricula presented using portable
handheld devices. The interface of the online portal would offer extensive
customization and intelligent learning features. These features would be
based on student and teacher needs. Students would interact with the
content for class assignments or discovery learning which in turn would
trigger algorithms processing through agents to relational databases
resulting in individualized student specific content. The content would
be presented in the students’ preferred learning style while the learning
system also intelligently adjusts content levels and supports. Wu and Lee
(1998) advocate a more concise approach to the development of intelligent
learning systems. They outline a model that includes: a) a systems approach
to intelligent learning systems; b) a paradigm hierarchy; and c) the emer-
gence of an agent model capable of initiating action. The computational
hierarchical layers Wu and Lee (1998) elaborate on include three levels
(Figure 1, next page). The first is an instruction-interaction layer which is
at the top and addresses problems by treating them in a logical manner. All
problems are defined by the instructional tasks needing action. The second
level is the architectural layer. This middle layer acts as a conduit between
the instructional-environment layer and the lower level processing layer.
Architectural layer components include the capabilities and performance
characteristics of the aggregated functional components of entities in the
lower processing level interconnected with the instructional-interaction
layer. The third level is the processing layer. This layer is the lowest level
where detailed interoperations and processing components are performed.
This processing results in intelligent system output based on unique
student parameters (e.g., learning profile characteristics).

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
12

Figure 1: Intelligent Learning System Layers

User Level Custom Content

Leveraging
Learning Styles
Instructional
Environment
Layer

Computer Machine Learning –

Computation Level Algorithm Processing
Architectural Layer

Database Level Data

Solution Processing Layer Input/Output
Level

Intelligent learning systems depend heavily on well planned and inte-

grated systems that adjust and adapt to the learner’s needs which are
guided by learning goals. Consideration should be given to the standard-
ization of instructional content at the meta level. This standardization
would utilize tagging or markup language that would allow databases to
quickly analyze and process information requests from students while
recognizing their unique learning styles and preferences. This processing
would depend on access to appropriately tagged digital content which
agents could then process, parse, and display relevant instructional
content. The National Instructional Materials Accessibility Standard
(NIMAS) (CAST, 2005) presents the first nationally recognized stan-
dard for digital content creation and markup. NIMAS will help guide the
digitization of accessible instructional materials by publishers in a system-
atic and hierarchical manner organized around the individual document
meta-information. This structure works from the highest level container
of a book’s major divisions down to the furthest subdivisions that nest
within chapters and includes information supplementary to the main text
and narrative flow (CAST, 2005). The NIMAS format is a significant step
forward on the road to intelligent learning systems but is only one of many
necessary components.
J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
13

Another model is advocated by Amigoni, Gatti, Pinciroli, and Roveri

(2005). They advocate the utilization of a distributed hierarchical task net-
work that combines both centralized and distributed features from which
information can be accessed. This would allow centralized processes to
access closed networks. The processes or agents would distribute licensed
copyrighted content, state specific curriculum, or assessment materials
while working within predefined, approved user parameters such as
learning style preferences, grade levels, and even content specific informa-
tion. Distributed features could allow specific content to be accessed from
wider less stringent or controlled sources such as public websites, govern-
ment databases, and select media services.

Assessment Systems
The same parameters for intelligent learning systems and content
structure are critical and warrant integration with accessible assessment
systems. Agents would process and respond to student input resulting in
assessment results parsed from structured or informal assessment content
using a NIMAS like schema to catalogue and store student test content.
Agents would also direct the storage of assessment results and preferences.
Research in the area of accessible assessment (Abell, Bauder & Simmons,
2004; Abell & Lewis 2005; Dolan et, al., 2005) advocates the same impor-
tance be placed on universally designed online assessment methods which
benefit and guide the instruction of all students regardless of ability level.
The researchers explore the flexibility within the assessment environment
resulting from the use of digital content (e.g. text, video, audio, or combi-
nations of). Flexible content representation offers more opportunities to
connect with individual student learning styles, thereby increasing knowl-
edge transfer. Quellmalz and Kozma (2003) support the use of informa-
tion and communication technologies (ICT) to help students solve complex
problems. The ICTs such as simulation, visualization, modeling tools, web
connectivity, and online communication allow learners to engage with the
assessment process in new and naturalistic ways. These tools along with
universally designed assessment materials could be incorporated into
the instructional environment layer leveraging machine learning models.
Models such as Latent Semantic Analysis (LSA) utilize human scored essays
as a vector triggering continuous computational analysis of the similarity
between each to-be-scored essay and each of the previously scored essays
(Landauer, Laham, & Foltz, 2003). LSA does not depend solely on direct
prediction which is based only on intuitive judgments from a predefined
set of index variables. As the field of intelligent computing develops and
matures, various forms of machine learning models emphasizing categori-
zation such as Naive Bayes (Tzeras & Hartmann 1993), Rocchio (Rocchio
1971), k-nearest neighbors (Yang 1994), and decision trees (Quinlan

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
14

1986) warrant consideration and further research. Baker (2003) encour-

ages the field to adopt models for technology design and assessment that
would include a set of minimum features: 1) linkage to cognitive demands
or requirements (e.g. problem solving, verbal comprehension); 2) methods
for attributing validity for various assessment purposes; 3) procedures
for generating multiple instances of a task, item, or simulation; 4) ana-
lytic approaches for providing reports targeted to users; 5) quality control
routines to assure content quality, appropriateness to the learner; and
6) fairness. More also needs to be done to help teachers see value in this
type of assessment model.
Stiggins (2004) postulates that educators have inherited an assess-
ment legacy that actually prevents teachers from tapping the full power
of assessment for school improvement. Maximum learning comes from
active engagement between the teacher and student. This engagement
allows students to decide if they are likely to succeed, if meeting the stan-
dards is worth the effort, and finally the results of possible failure. Students
look at these factors on a personal level. Stiggins (2004) further alludes
that instructional decisions, in addition to being personal, also occurs in
a fluid day to day instructional environment, not once a year resulting in
standardized test scores. Using online assessment tools allows teachers
and students to quickly and efficiently assess individual learning and class
wide instructional progress. Future scenarios could see paper and pencil
quizzes replaced with online or personal digital assistant (PDA) based
quizzes that students could take during class or on their own time at home.
Online access, PDA, or digital ink devices (Economist, 2000) would offer
a variety of options for students to take and review assessment results.
Other interactive assessment methods using interactive class feedback
(Qwizdom, 2004) could be used which depend less on online connectivity
while still tracking results at the student and classroom level for teachers
and students to analyze.

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
15

Technology Infrastructure and

Interactive Media Challenges
To have a viable intelligent learning system, students and teachers
require easy access to technology and digital content that is aligned to
the curriculum. The intelligent learning system is intimately connected to
universally-designed instructional content offered in digital format that
includes the technology needed by students to display and interact with it.
The technology infrastructure currently available in today’s society
including schools continues to grow and will be a benefit to intelligent
curricular applications. This growth is in part due to the growth of the
internet and free wireless network access. The necessary components of
intelligent learning curricula are currently available or becoming available.
These include robust computer networks and storage capabilities, wire-
less network access, affordable computers and handheld display devices
such as cell phones or Palm Pilots to display digital content, inexpensive
broadband Internet access, national file format standardizations such as
NIMAS, and textbook publishers offering materials in digital formats.
More work is required to shape these components into model intelligent
learning system applications.
A number of components necessary for intelligent learning curricula
need research and development efforts to move forward. These areas
include the textbook publishing industry continuing to move toward
digital content production and sale using market driven pay for fee models
while adopting file format standardizations such as NIMAS. The national
file format standardization found in NIMAS allows researchers to delve
deeper into ways that digital content can be personalized using different
tag structures. Personalization of the learning environment also comes
with risks and criticisms though.
Clark (1983) noted that technology or “media” is not the driving force
behind learning; it is the content that is presented through the media itself.
Research in the use of technology in education is ambiguous and falls on
both sides of the debate. This is supported by the research of MacArthur,
Ferretti, Okolo, and Cavalier (2001) who conducted an extensive review
of educational technology research. Their review of evolving technology
applications used for students with literacy problems found that select
students benefited while others did not. This hastens the point that tech-
nology may be more beneficial when used in the context of a student’s
unique needs rather than universally.
Individual student needs are often varied and significant. This raises
the point of how much control and flexibility of the curricula should be
given to students in an effort to help them achieve at high levels. Too
J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
16

much flexibility and control of content in the hands of students could

have an unintended effect of lowering expectation and student challenge.
Close monitoring by teachers, instructors, or learning coaches may need
to be instituted to allow students the appropriate challenges needed in the
learning process itself.

Future Research and Next Steps

Advanced research is needed to construct models that probe, analyze,
and predict student learning needs. This research should take into account
current student data such as ability level, interests, standardized and non-
standardized test results, as well as personal learning goals and profiles.
One way to assemble intelligent learning curricula may be to align indi-
vidual student learning characteristics with tagged instructional content
using pedagogical tags rather than purely structural tags which would not
take into account the students learning characteristics.
To integrate more intelligent technology with learning processes,
linkages are also needed between learning and cognitive scientists along
with computer and computational researchers. Great strides can be made
through better understanding of cognitive learning styles and how to
align these to intelligent computing to create custom learning environ-
ments and materials for all students. The blending of intelligent computing
with learning style preferences and the subsequent schema could provide
the conceptual models needed to build more advanced and personalized
learning systems.
Research might also examine the coordination and execution of algo-
rithms that create learning style specific environments for individual
learners perhaps using pedagogically tagged content. This would allow
computer databases to analyze and process student requests and create
instructionally relevant curriculum material aligned to state standards
and student learning styles. This research should focus on a description
of the environment, goals, and student capabilities in the form of possible
learning profiles. Lastly, flexible computer models that allow for a variety
of learning styles to be met through intelligent learning systems should be
explored.
Though numerous challenges lay ahead in developing intelligent inter-
active learning systems and accompanying curricula, the potential bene-
fits to society warrant the physical, intellectual, and financial investment.
This investment in the development of intelligent learning systems has
the potential to impact every individual regardless of ability by opening
new doors to learning and potentially a better life.

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
17

References
Abell, M., Bauder, D. Simmons T., Sharon, D. (2003). Using personal
digital assistants (PDA) to connect students with special needs to the
general curriculum. Closing the Gap, 22(1), 20
Abell, M., Bauder, D., & Simmons, T. (2004). Universally designed online
assessment: Implications for the future. Information Technology and
Disabilities Journal, 10(1). Retrieved July 26, 2005 from http://www.
rit.edu/%7Eeasi/itd/itdv10n1/abell.htm
Abell, M. & Lewis, P. (2005). Universal design for learning: A statewide
improvement model for academic success. Information Technology and
Disabilities Journal, 11(1). Retrieved August 20, 2005 from http://
www.rit.edu/~easi/itd/itdv11n1/abell.htm
Association of American Publishers (2002). AAP Facilitates Ground-
Breaking Discussion of Accessible Curriculum Materials. Retrieved
December 16, 2004 from the Association of American Publishers
website: http://www.publishers.org/press/releases.cfm?PressRelease
ArticleID=108
Amigoni, F., Gatti, N., Pinciroli, C., & Roveri, M. (2005). What planner
for ambient intelligence applications: IEEE transactions on systems.
Man & Cybernetics, 35(1), 7–22.
Atkinson, G. (1991). Kolb’s learning style inventory: A practitioner’s
perspective. Measurement and Evaluation in Counseling and
Development, 23(4), 149-161.
Baldwin, L. & Sabry, K. (2003). Learning styles for interactive learning
systems. Innovations in Education & Teaching International, 40(4),
325–341.
Baker, E. (2003). Reflections on technology-enhanced assessment.
Assessment in Education, 10(3), 421–425.
Behrmann, J. (2001). Electronic materials can be important for students
with disabilities. Teaching Exceptional Children, 34(2), 87–88.
Bork, A. (1997, June). The future of computers and learning. Technology
Horizons in K–12 Education (T.H.E.) Journal, 24(11), 69–78.
Center for Applied Special Technology (CAST). (2005). National
instructional materials accessibility standard report (Version 1.0).
Wakefield, MA. Retrieved July 20, 2005 from the National
Instructional Materials Accessibility Standard (NIMAS) website:
http://nimas.cast.org/about/report/index.html#3techpanelb

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
18

Chester, E. (2002). Making school reform work. Public Interest, 148,

85–96.
Clark, R. E. (1983). Reconsidering research on learning from media.
Review of Educational Research, 43(4), 445–459.
Cuban, L. (1993). Computers meet classroom: Classroom wins. Teachers
College Record, 95(2), 185–211.
Digital ink meets electronic paper. (2000, December 9). Economist
Technology Quarterly, 19–21.
Dolan, R. P., Hall, T. E., Banerjee, M., Chun, E., & Strangman, N. (2005).
Applying principles of universal design to test delivery: The effect
of computer-based read-aloud on test performance of high school
students with learning disabilities. Journal of Technology, Learning,
and Assessment, 3(7). Available from http://www.jtla.org
Duff, A. (2004). A note on the problem solving style questionnaire: An
alternative to Kolb’s learning style inventory? Educational Psychology,
24(5), 699–709.
Elimiani, P. L. & Stephanidis, C. (2005). Universal access to ambient
intelligence environments: Opportunities and challenges for people
with disabilities. IBM Systems Journal, 44(3), 605–619.
Felder, R (1993) Reaching the second tier: Learning and teaching styles
in college science education. Journal of College Science Teaching, 23(5),
286–90.
Kentucky Department of Education (2004). Kentucky state improvement
grant: Assistive technology component. Lexington, KY: University of
Kentucky Interdisciplinary Human Development Institute.
Landauer, T., Laham, D., & Foltz, P. (2003). Automatic essay assessment.
Assessment in Education, 10(3), 295–309.
MacArthur, C., Ferretti, R., Okolo, C., & Cavalier, A. (2001). Technology
applications for students with literacy problems: A critical review.
The Elementary School Journal, 101, 273–378.
McKenzie, J. (2000). Scaffolding for success: Beyond technology, questioning,
research and the information literate school community. [Electronic
version]. Retrieved October 29, 2006 from the From Now On (FNO)
website: http://fno.org/dec99/scaffold.html
Meyer, A. & Rose, D (2000). Universal design for individual differences.
Educational Leadership, 58(3), 39–43.

J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
19

Psotka, S. (1994). Intelligent tutoring systems: Past, present and future.

In D. Jonassen (Ed.), Handbook of educational communications and
technology, (pp. 570–590). New York: Macmillan.
Quellmaiz, E. & Kozma, R. (2003). Designing assessment of learning
with technology. Assessment in Education: Principles, Policy, and
Practice, 10(3), 389–408.
Quinlan, R. (1986). Induction of decision trees. Machine Learning, 1,
81–106.
Qwizdom®. (2004). Interactive Response System. [Computer software].
Belfast, IR: Available at www. Qwizdom.co.uk.
Rocchio, J. (1971). Relevance feedback in information retrieval.
In Gerard Salton (Ed.), The SMART retrieval system: Experiments in
automatic document processing (pp. 313–323). Englewood Cliffs, NJ:
Prentice-Hall.
Rose, D. H.,& Meyer, A. (2002). Teaching every student in the digital
age: Universal design for learning. Alexandria, VA: Association for
Supervision and Curriculum Development.
Seaward, M. (1998). Interactive assistants provide ease of use for novices:
The development of prototypes and descendants. Computers in Human
Behavior, 14(2), 221–237.
Shute V. & Psotka, J. (1994). Intelligent tutoring systems: Past, present
and future. D. Jonassen (Ed.), Handbook of research on educational
communications and technology. New York: Scholastic Publications.
Stiggins, R. (2004). New assessment beliefs for a new school mission.
Phi Delta Kappan, 86(1), 22–28.
Tzeras, K. & Hartman, S. (1993). Automatic indexing based on Bayesian
inference networks. Proceedings of ACM SIGIR, (Pittsburg, PA),
22–34.
United Streaming. (2005). Discovery Education [Computer software]
Evanston, IL: Available at www.unitedstreaming.com
Wasson, B. (1997). Advanced educational technologies: The learning
environment. Computers in Human Behavior, 13(4), pp. 571–594.
Wu. A. & Lee. M. (1998). Intelligent tutoring systems as design.
Computers in Human Behavior, 14(2), 209–220.
Yang, Y. (1994). Expert network: Effective and efficient learning from
human decisions in text categorization and retrieval. Proceedings
of the 17th Annual International ACM SIGIR Conference on Research
and Development in Information Retrieval, Dublin, IR.
J·T·L·A
 Individualizing Learning Using Intelligent Technology and Universally Designed Curriculum Abell.
20

Author Biography
Michael M. Abell, Ph.D, is a researcher at the University of Louisville,
College of Education and Human Development. His interests include
issues surrounding access to the general curriculum, universal design,
psycho-educational assessment and technology integration benefiting
all students including those with disabilities. His research involves
grants and publications that facilitate the development of online
learning environments with interactive content as well as universally
designed instructional materials and assessments for classroom use.
Address: Michael M. Abell, ([email protected]), University of
Louisville, College of Education and Human Development, Louisville,
KY 40292.

J·T·L·A
The Journal of Technology, Learning, and Assessment

Editorial Board
Michael Russell, Editor Marshall S. Smith
Boston College Stanford University
Allan Collins Paul Holland
Northwestern University Educational Testing Service
Cathleen Norris Randy Elliot Bennett
University of North Texas Educational Testing Service
Edys S. Quellmalz Robert Dolan
SRI International Center for Applied
Special Technology
Elliot Soloway
University of Michigan Robert J. Mislevy
University of Maryland
George Madaus
Boston College Ronald H. Stevens
UCLA
Gerald A. Tindal
University of Oregon Seymour A. Papert
MIT
James Pellegrino
University of Illinois at Chicago Terry P. Vendlinski
UCLA
Katerine Bielaczyc
Museum of Science, Boston Walt Haney
Boston College
Larry Cuban
Stanford University Walter F. Heinecke
University of Virginia
Lawrence M. Rudner
Graduate Management
Admission Council

www.jtla.org
Technology and Assessment Study Collaborative
Caroline A. & Peter S. Lynch School of Education, Boston College