Buehler, Erin Et Al

Investigating the Implications of 3D Printing in Special Education
ERIN BUEHLER and NIARA COMRIE, University of Maryland, Baltimore County

MEGAN HOFMANN, Colorado State University
SAMANTHA McDONALD and AMY HURST, University of Maryland, Baltimore County
Consumer-grade digital fabrication such as 3D printing is on the rise, and we believe it can be leveraged to
great benefit in special education. Although 3D printing is infiltrating mainstream education, little research
has explored 3D printing in the context of students with special support needs. We describe our studies on
this topic and the resulting contributions. We initially conducted a formative study exploring the use of 3D
printing at three locations serving populations with varying ability, including individuals with cognitive,
motor, and visual impairments. We found that 3D design and printing perform three functions in special
education: (1) STEM engagement, (2) creation of educational aids for accessible curriculum content, and
(3) making custom adaptive devices. As part of our formative work, we also discussed a case study in the
codesign of an assistive hand grip created with occupational therapists at one of our investigation sites. This
work inspired further studies on the creation of adaptive devices using 3D printers. We identified the needs 11
and constraints of these therapists and found implications for a specialized 3D modeling tool to support their
use of 3D printers. We developed GripFab, 3D modeling software based on feedback from therapists, and
used it to explore the feasibility of in-house 3D object designs in support of accessibility. Our contributions
include case studies at three special education sites and discussion of obstacles to efficient 3D printing in this
context. We have extended these contributions with a more in-depth look at the stakeholders and findings
from GripFab studies. We have expanded our discussion to include suggestions for researchers in this space,
in addition to refined suggestions from our earlier work for technologists creating 3D modeling and printing
tools, therapists seeking to leverage 3D printers, and educators and administrators looking to implement
these design tools in special education environments.
CCS Concepts: r Social and professional topics → People with disabilities
Additional Key Words and Phrases: 3D printing, assistive technology, children, cognitive impairment, digital
fabrication, developmental disability, rapid prototyping, special education, visual impairment
ACM Reference Format:
Erin Buehler, Niara Comrie, Megan Hofmann, Samantha McDonald, and Amy Hurst. 2016. Investigating
the implications of 3D printing in special education. ACM Trans. Access. Comput. 8, 3, Article 11 (March
2016), 28 pages.
DOI: http://dx.doi.org/10.1145/2870640
This work is supported by the National Science Foundation under grant number EEEC-0540865.
Authors’ addresses: E. Buehler, N. Comrie, S. McDonald, and A. Hurst, University of Maryland, Baltimore
County, Information Systems Department, 1000 Hilltop Circle, Baltimore, MD 21250; emails: {eri4, cniara1,
sam30, amyhurst}@umbc.edu; M. Hofmann, Colorado State University, Fort Collins, Department of Com-
puter Science, 1873 Campus Delivery, Fort Collins, CO 80523; email: hofmann.megan@gmail.com.
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted
without fee provided that copies are not made or distributed for profit or commercial advantage and that
copies show this notice on the first page or initial screen of a display along with the full citation. Copyrights for
components of this work owned by others than ACM must be honored. Abstracting with credit is permitted.
To copy otherwise, to republish, to post on servers, to redistribute to lists, or to use any component of this
work in other works requires prior specific permission and/or a fee. Permissions may be requested from
Publications Dept., ACM, Inc., 2 Penn Plaza, Suite 701, New York, NY 10121-0701 USA, fax +1 (212)
869-0481, or permissions@acm.org.
c 2016 ACM 1936-7228/2016/03-ART11 $15.00
DOI: http://dx.doi.org/10.1145/2870640
ACM Transactions on Accessible Computing, Vol. 8, No. 3, Article 11, Publication date: March 2016.
11:2 E. Buehler et al.
1. INTRODUCTION
Three-dimensional (3D) printers are a promising technology gaining acceptance in
mainstream education as a means to engage students with hands-on interactions [Tides
Center 2014]. However, little has been published on the role of this technology in
special education. We feel this technology can contribute to special education through
(1) supporting STEM (Science, Technology, Engineering, and Math) engagement in a
historically underserved population, (2) creating curricular materials, and (3) creating
assistive technology (AT).
3D printing technology can offer students a powerful tool for creativity and explo-
ration and an engaging introduction to STEM topics. For example, students can print
simple machines to help visualize elementary concepts in physics and engineering or
design their own small-scale buildings as part of an architecture lesson. Finding ways
to increase participation in science and engineering is a universal challenge, but it is
particularly important for students with disabilities, as they are severely underrepre-
sented in these disciplines [National Science Foundation 2013].
Customized learning aids in the classroom and individualized assistive technology
are both possibilities for 3D printing in special education. In the classroom, for exam-
ple, a history teacher could download an open-source model of a pharaoh’s tomb and
have a durable plastic diorama, which would be content specific and could be handled,
helping students with visual impairments as well as tactile learners. In the case of
assistive devices, a therapist could use a 3D printer to create a custom tablet mount for
a wheelchair, tailoring the design to both the tablet case and the individual student’s
chair, circumventing purchasing processes and exorbitant costs. In order to achieve
these goals, special education institutions must first gain access to 3D printing technol-
ogy and then make the time to learn how to use, operate, and maintain this equipment.
To understand how special education institutions are currently using 3D printing
technologies, we studied three organizations providing special education and technol-
ogy access services. We observed a classroom and conducted interviews with teachers,
therapists, administrators, and technical experts. We worked with two schools that
support middle- and high-school-aged children, one focused on students with cogni-
tive impairments and the other on students with vision impairments. Our third site
was the technology division of a national organization offering training and support to
blind individuals. Our analysis revealed insights into the current uses of 3D printing
in special education and how we can better support its use.
Based on observations and interviews at our main investigation site, we identified
a potential project leveraging 3D printing for the creation of assistive devices. We
engaged in an iterative codesign process with occupational therapists from that site
to explore this application of the technology. The codesign process resulted in a case
study exploring 3D printed objects as assistive devices and was the motivation for
GripFab, a prototype design software intended for the simple creation of 3D printed
objects to support persons with limited dexterity and gripping ability. The software
requirements were derived from therapist feedback and the prototype software was
brought back to therapists twice for evaluation and to evoke further discussion of the
merits and potential problems related to in-house or self-designed assistive objects.
In this article, we first summarize existing research on 3D printing in education,
DIY accessibility, and learning. We describe our investigation into the current use and
practices of 3D printing at three sites and present each investigation site as a case
study. We extended our original work [Buehler et al. 2014] by describing the expecta-
tions and perceptions of 3D printing from the perspective of multiple stakeholders in
special education. We also introduce GripFab, a specialized piece of modeling software
to create custom hand grips that was designed for clinicians at one of our investigation
Investigating the Implications of 3D Printing in Special Education 11:3
sites. We present applications of this prototype software and we discuss the opportu-
nities and challenges involved in efficiently implementing 3D modeling and printing
for special education use. Finally, we offer revised suggestions for future designers and
technologists in this space, as well as our lessons learned related to 3D printing in
special education environments for therapists, educators, and researchers.
2. RELATED WORK
Our research builds on existing explorations of making in children’s education, Do-
It-Yourself (DIY) assistive technology, and Universal Design for Learning. We briefly
describe relevant work from these topics to help frame our research goals next.
2.1. Education and Making

With the rise in personal fabrication, there has been an increased interest in how
children can interact with fabrication technologies. Encouraging engagement in STEM
fields is a prevalent topic in education and one that researchers in our field are ad-
dressing using digital fabrication. Leduc-Mills and Eisenberg [2011] have developed
a series of technologies and conducted workshops on involving children in 3D design.
They developed the UCube [Leduc-mills and Eisenberg 2011], which is a tangible in-
terface for designing objects in three-dimensional space. Researchers have conducted
workshops to explore the implications of making in the classroom as a reinvention of
constructionist learning and discussed the accessibility of tools for children in main-
stream contexts [Blikstein and Krannich 2013; Krannich et al. 2012; Leduc-Mills et al.
2013]. Posch and Fitzpatrick [2011] also conducted an extensive series of workshop
studies exploring children’s expectations and outcomes when using 3D design tools.
These workshops used open-source tools and personalized take-home projects, finding
that “An especially crucial factor for the children’s sense of satisfaction appeared to be
the technical challenge and mastery of the task and the personal engagement associ-
ated with the result.” Additionally, we are starting to see books published on the topic
of making and DIY skill building for children [Kemp 2013; Libow et al. 2013]. Again,
these focus on constructionist learning and STEM engagement.
Our study supports and extends this work by furthering the understanding of ed-
ucator perspectives on making in the classroom, as well as examining the needs of
students with disabilities both in terms of value gained from making and their access
to this new wave of digital fabrication. We extend our prior work [Buehler et al. 2014]
in this area by providing more in-depth discussion of the educator and administrator
perspectives on 3D printing in special education.
2.2. Making and Assistive Technology

DIY fabrication addresses two important issues in assistive technology—cost and cus-
tomization. Financial costs of the device and assessment can be astronomical for the
average user. Proper fit and personalization are key issues in AT abandonment. With
3D printing, end-users can create their own custom designs at reduced cost [Hurst and
Tobias 2011]. Brown and Hurst’s [2012] VizTouch software creates 3D printable tactile
mathematical graphics for use in accessible education and business. Finally, Hurst and
Kane [2013] offer example tools designed to make DIY more accessible to a variety of
populations, including encouraging children in STEM topics [Kane and Bigham 2014].
Creating assistive technology for children raises additional challenges that must be
addressed before children will adopt these technologies. Several of these challenges are
highlighted in the works of Dawe [2006] and Copley and Ziviani [2004], in which barri-
ers to adoption are identified with respect not only to AT devices but also to the people
and environments surrounding young users. The exploration of Hayes et al. [2010] into
visual support for children with autism uncovered other key factors related to children
and AT. Children change and outgrow devices very quickly, which can lead to problems
when being fitted for these items. These problems include the manufacturing time for
custom-ordered devices and prolonged waiting periods for paperwork processing be-
tween medical/insurance providers and manufacturers when ordering generic items.
Another aspect of AT adoption versus abandonment is the perception of others. Chil-
dren are susceptible to popular opinion and the sense of belonging, making the stigma
around AT a serious issue.
In our earlier work, we investigated how 3D printing can address these concerns in
the context of special education, providing inexpensive and customized solutions for
young students [Buehler et al. 2014]. This article extends our prior work, including
additional findings regarding the stakeholders in this setting and the presentation of
a prototype study examining a novel software design tool, GripFab, targeted to occu-
pational therapists. We also expand on our discussion and provide further suggestions
and insights regarding these technologies.
2.3. Universal Design for Learning
Universal Design for Learning (UDL) research has explored the use of tangible learn-
ing aids for students with varying abilities. UDL is an extension of universal design
for accessibility applied to education. UDL follows three guiding principles: (1) flexible
methods of presentation, (2) flexible methods of expression and apprenticeship, and
(3) flexible options for engagement [Rose et al. 2002]. By offering educators a way to
create their own 3D printed objects, they can in turn offer multimodal interactions for
their students. The benefits lie in presentation of information and demonstration of
knowledge gained. A student with a visual impairment may gain access to graphical in-
formation tactilely, or a student with a cognitive impairment might demonstrate math
skills by manipulating objects rather than writing out equations. When discussing
multimodal interactions, the concept of tangibles is a natural connection. Manches and
O’Malley [2011] provide an extensive review on the topic of tangibles and manipulatives
in education and how to support these interactions.
Our research examines the potential for 3D printers to assist with the creation of
educational aids, adaptive devices in support of learning, and manipulatives—physical
objects designed to promote comprehension through interaction. We also present expec-
tations and perceptions of faculty and staff in special education settings. In extension
of our prior work, we document applications of 3D printing as suggested by special
education instructors.
2.4. State of 3D Modeling and Printing Technology
In order to create an object to be manufactured on a 3D printer, one needs a 3D model.
Depending on the situation, an individual may find a preexisting 3D model that some-
one else designed through popular online repositories such as Thingiverse1 or create
their own using 3D modeling software. There is a wide variety of 3D modeling software
created for novices (individuals with little to no formal training) and experts (individ-
uals who commonly have engineering or animation training). Two examples of novice
design tools include Tinkercad2 and Sketchup.3 Both tools are direct manipulation
WYSIWYG (What You See Is What You Get) interfaces that output a standard format
(.stl) and are free to use. They offer uncluttered interfaces and objects are built by
combining primitive shapes (such as cubes, triangles, and spheres).
1 http://www.thingiverse.com/.
2 http://www.tinkercad.com/.
3 http://www.sketchup.com/.
Table I. Participants Interviewed at Each Site, Listed by Role

Instructors Administrators OTs AT Specialists
Site A 3 3 2 2
Site B 1 1 0 0
Site C 0 0 0 1
Users who have difficulty using on-screen tools may prefer to design models tangibly
instead of using on-screen tools. Within the research community, there has been an
influx of tangible design tools that enable users to design 3D models by manipulating
physical objects. A 3D model is created by 3D scanning these physical objects [Izadi
et al. 2011; Follmer et al. 2010], or objects may have embedded electronics that can
specify their configuration to the computer [Huang and Eisenberg 2012]. While users
can quickly create low-resolution primitives with these systems, their complexity is
commonly limited.
Consumer-grade 3D printers are available in a variety of types; one of the most
common types used at two of our investigation sites is fused deposition modeling (FDM).
These printers heat plastic filament and extrude fine layers of plastic that ultimately
create a 3D object. A second type used at our other investigation site is a powder bed
printer. This printer combines powder with layers of a binding agent that can be tinted
to generate a 3D object in full color.
In our study, we describe challenges and opportunities related to both novice design
tools and consumer-grade 3D printers given the state of this technology at the time of
our fieldwork.
3. DATA GATHERING METHODS

Our investigation set out to gain an understanding of how 3D printing can support spe-
cial education and therapy. We conducted interviews and observations at three diverse
sites: Site A, a special education school serving students with cognitive impairments
and multiple disabilities; Site B, a school serving students with visual impairments;
and Site C, the technology services department of a national organization working
with blind individuals. We conducted long-term classroom observations at Site A and
completed a series of interviews with staff, instructors, and therapists at all three sites.
3.1. Interviews with Faculty and Staff

We defined a set of stakeholders (Table I) we perceived would be relevant to the intro-
duction of 3D printing in special education. These stakeholders included faculty, staff,
and administrators. The stakeholders at Sites A and B give support to students and
represent important perspectives in the delivery of curriculum and assistive technol-
ogy to students in multiple educational settings. At Site C, we spoke with a technology
coordinator who provides information to organizations and end-users on assistive tech-
nology related to vision impairments.
We recruited participants through email, explaining our research interests and invit-
ing faculty and/or staff to participate. We conducted 13 interviews across three sites
(Table I). We conducted individual semistructured interviews lasting 30 to 60 minutes,
and all but one was conducted in person and on site. These interviews occurred after
we began our observations at Site A and were guided by that data. The interviews
were structured around the following themes: software and hardware challenges, stu-
dent engagement with 3D printing, printer maintenance, safety, intellectual property
rights, and time committed to 3D printer use.
We demonstrated 3D-printed objects in the interviews to illustrate the capabilities
of current printers (Figure 1). These objects included 3D-printed tactile graphics, small
Fig. 1. Examples of 3D-printed objects to support special education. Pictured: tactile graphics for students
with visual impairments, a small sculpture and moving nautilus gear for hands-on engagement, and two
assistive devices including an easy-grip bottle opener and a pillbox.
models, and working gears. Interview questions and prompts were designed to elicit
discussion and further insight on observed themes, but also served to reveal new issues
not previously identified. The questions were tailored for the role of the interview
participant. For example, we asked technology instructors about the strategies they
used to manage student expectations in the classroom; we asked technical experts about
failures and successes using the technology; and we asked nontechnical instructors and
therapists about their perceptions of and uses for 3D printers in their work setting.
3.2. Classroom Observations at Site A

We conducted observations at Site A for one school year while attending six periods of
an information technology class each week. This course’s curriculum taught computer
skills, ranging from office productivity to computer hardware and networking. The
instructors and administrators of this school were also interested in incorporating 3D
printing into the curriculum. In order to remove obstacles for them to get the printer up
and running, our research group provided the classroom with a 3D printer, supplies,
and technical support. Our goal was to gather real-world stories about 3D printing
in the classroom and remove obstacles that might prevent them from utilizing this
technology.
Class sections contained between three and 13 students, two instructors, and one to
four student aides, and lasted 45 minutes. Students were grouped by their experience
and their support needs from instructors. Thus, students with more experience and
lower support needs were placed together, and those students with less experience and
greater support needs were grouped separately. We took detailed notes on classroom
activities and artifacts such as instruction worksheets, student 3D design files, and 3D
printed objects.
3.3. Codesign and Prototyping with Occupational Therapists

A development stemming from our observations and other interactions at Site A led
to a long-term codesign partnership with some of the occupational therapists (OTs).
Together we worked on an assistive device project that went through several iterations
of prototyping and evaluation (Section 5). The end product was a hand grip printed to
fit a stylus for a specific student with limited hand dexterity. This codesign task was
the motivation for a custom design software prototype that we developed and named
GripFab.
We conducted a pilot test of GripFab with four OTs at Site A, including the two
OTs who participated in the hand-grip design project. The participants were given a
crash course in 3D printing technology and given a demonstration of GripFab before
being asked a series of questions to gauge their interest in GripFab and its potential
applications as design software at their site.
In a follow-up to our pilot study, we completed a focus group with three OTs and
an art instructor at Site A, and we demonstrated a revised version of GripFab and
brought several printed examples of new hand-grip designs. During the focus group,
we asked for feedback on both the software and the new models, and we also used these
items to elicit feedback on the important features and software requirements for these
participants to engage in 3D modeling and printing techniques.
3.4. Data Analysis Techniques

We transcribed and hand-coded observations from field notes. This process entailed
two coding passes noting high-level themes associated with the expectations, goals,
uses, and challenges related to 3D printing. From this coding, certain themes and
stakeholders emerged. These topics were then used to inform a collection of semistruc-
tured interview questions to further explore the issues associated with 3D printing in
education. Interviews were also transcribed, hand-coded, and finally examined using
affinity diagramming. The richness of data gathered from Site A enabled us to split
this location into findings that applied in the classroom and findings that applied to
therapeutic services on the campus. For the codesign and prototyping sessions with
OTs, focus groups were again transcribed and hand-coded. We describe our findings by
site (Sections 4, 5, 6, and 7) and stakeholder type (Section 8), describe outcomes from
our prototype design software (Section 9), and address common themes and challenges
(Section 10).
4. SITE A (CLASSROOM): SPECIAL EDUCATION SCHOOL

Site A is a special education school that caters to students with moderate to severe
cognitive disabilities and, in some cases, multiple disabilities. Students are referred
to this school from public schools if they demonstrate a need for additional support
services. We focused on a single classroom teaching technology literacy skills.
4.1. 3D Printing and Design Resources

This site used a MakerBot Replicator 2x4 (Table II), on loan from our research lab for
the duration of the study. These printers produce objects by additive manufacturing,
heating plastic (ABS or PLA) filament. The filament is fed by a stepper motor into a
nozzle that extrudes melted plastic onto a heated build plate in progressive layers. The
primary design software being used at Site A was Tinkercad, a web-based 3D design
tool that is free to use and is designed to support novice users. We suggested this
software for its volume of tutorial materials and for its novice-friendly interface.
4 http://www.makerbot.com/.
Table II. Specifications for the Printers Used at our Sites
4.2. Primary Users

The instructors introduced the students to the concept of 3D printing by showing
videos on 3D printing and giving live printing demonstrations in the classroom. The
instructors encouraged students to explore existing 3D designs available for viewing,
customizing, and downloading from Thingiverse, a 3D printing community that en-
courages sharing of printable designs. Students selected designs they wanted to see
printed before moving on to experimenting with their own designs. Instructors were the
most frequent users of the printer and performed all maintenance and troubleshooting
tasks, such as starting and stopping prints, filament changes, and printer calibration.
We helped instructors when problems arose that they could not resolve, as they had no
prior experience with 3D printers before our study.
4.3. Use of 3D Printing

Students were given access to Tinkercad and, after completing the website’s tutorials
(three class periods for some students, more for others), students were encouraged
to create their own designs for printing. However, limited curriculum time and diffi-
culty mastering tools prevented most students from finishing custom designs. Instead,
students commonly looked for designs from open-source sites to print or modify.
During our observations, the printer was frequently used to produce items for a
recurring school fundraiser. The students at this site were encouraged to engage in
entrepreneurial activities like creating goods or providing services, and the faculty
felt the 3D printer was ideally suited to this exercise. Students selected designs from
Thingiverse and the technology instructors printed these designs to sell at a student-
run sales event (Table III). These objects were chosen because they were perceived
as having universal appeal and would sell at the event. The instructor chose to sell
existing designs rather than student designs because they felt the students had not
yet mastered the design software.
Table III. Top, Two Items Selected, Downloaded, and Printed by Students for a Fundraiser: (1) a Bracelet and
(2) a Cube toy. Bottom, Two Examples of Student-Designed Dog Figurines (3, 4)
4.4. Objectives and Obstacles

The technology instructor’s ultimate hope was for students to independently design
and print objects. The technology instructor wanted to see students use the printer
for fundraiser sales or as a type of in-house manufacturing for curriculum aids for
teachers outside of the technology classroom. Ideally, other instructors would request
that a model or other tactile manipulative be designed and printed by the students,
similar to a classroom-based engineering firm. The technology instructors felt this
type of experience would provide both STEM engagement and empowerment for their
students; however, the progress for these goals has been slow.
Even though Tinkercad is designed for novices, we observed that the students expe-
rience several challenges using this software to design new models. Examples include
difficulty manipulating camera angle; moving objects instead of adjusting view; confu-
sion about which handles changed scale, rotation, and position of objects; and difficulty
selecting multiple objects. These challenges were particularly characteristic of stu-
dents needing high support. Students could become frustrated and distracted by the
interface, and their interest in completing a design appeared to diminish.
In addition to having difficulty using the software, some students had difficulty
understanding the physical capabilities of the printer and its materials with respect
to their own design concepts. If an object is not positioned on the building plane, this
can cause complications during the printing stages either with software being able to
prepare the file for the printer or resulting in a failed print. The ability to perceive
designs on three planes (x, y, and z) challenged students with high support needs.
Fig. 2. An example of an unprintable design. This house has a floating roof and is situated below the virtual
print bed. Created by a student at Site A.
This resulted in students creating designs that were not seated correctly on the
build platform or designs that included object overhangs or details beyond the current
capabilities of the printer (Figure 2).
5. SITE A (THERAPISTS): SPECIAL EDUCATION SCHOOL

We found that other support staff at Site A were also interested in the 3D printer, and
this section describes how OTs wanted to use this technology.
5.1. Printing Resources

Once we introduced the OTs to the 3D printer and explained its capabilities, they saw
many opportunities for the technology in their work, including modifying or replacing
basic AT items like accessible clasps, custom technology cases, or novel designs. We
assisted therapists in the design and printing of a 3D object as a prototype for an AT
design to encourage the therapists’ exploration of this technology.
The therapists had access to the same MakerBot Replicator 2x in the instructional
classroom, but none of them chose to use the device directly due to time constraints and
training needs. Instead, we worked closely with them on the design and manufacturing
of custom objects. This group was interested in scanning physical models made from
clay, so we provided a MakerBot Digitizer5 (a 3D scanner that combines images of
a physical object and converts this to a closed 3D object for printing). We refined
and modified these models using a combination of open-source software, including
Tinkercad, MeshMixer,6 and Autodesk 123D.7
5.2. Primary Users

The OTs at Site A provided therapy to a diverse student population. Sessions lasted
for the duration of a class period (45 minutes) and often incorporated skills practice
or other specialized therapeutic routines for students. The therapists did not feel they
had the time or skills to commit to working with the printer directly, but described
their existing practices for modifying AT devices or creating new ones mostly through
the use of quick-fix supplies like Velcro, tape, string, paper, cardboard, and so on. We
felt we could translate these practices to low-fi prototyping and assisted two OTs with
the technology to create a final product.
5 http://store.makerbot.com/digitizer/.
6 http://www.meshmixer.com/.
7 http://www.123dapp.com/design/.
Table IV. The Adaptive Stylus Grip Went Through Several Design Iterations. Pictured: (1) Initial Air-Dry
Clay Models and an Early 3D-Printed Prototype of the Student’s Grip, (2) the Final Low-fi Clay Design;
(3) Scanning the Clay Model with the Digitizer, (4) Designing an Extension to Accommodate a New
Stylus, and (5) Final 3D-Printed Grip with Stylus Inserted

At Site A, we collaborated on a 3D-printed assistive technology project with two of
the school’s OTs. We saw this case study as an opportunity to find out what these
therapists could be doing with the printer, if given the right resources. The goal was to
create a unique stylus grip for a student with a limited grasp. The OTs had previously
tested over $150 worth of assistive and mainstream off-the-shelf stylus and stylus-like
products but found no satisfactory solution.
Prior to our design activity, the student was using the eraser tip of a pencil combined
with a soft gel grip, but this did not meet the exact needs of the student and sometimes
caused discomfort or fatigue. In order to create a custom device, we agreed it would be
best to create a new design shaped to fit the student’s hand and that the device could
be 3D printed.
This grip was designed through three iterative rapid prototyping sessions (Table IV).
The OTs created initial prototypes using lightweight, air-dried clay to form a base
model of the student’s grip, which was then 3D scanned by the researchers using the
Digitizer. Adjustments were made to the design to accommodate a basic stylus and the
design was printed using a Replicator 2x located in our lab. The plastic prototype was
delivered to the OTs, who conducted test sessions with the student. During testing,
concerns were raised about the hardness of the plastic. Our lab coated the grip in a
rubberizing aerosol spray to soften the texture and improve grasping. Both the new
texture and the bright color (picked by the student) were well received. The student
later purchased a different stylus and the task of modifying the grip prompted us to
explore other design options that we discuss in our software prototyping, in Section 9.

Therapists from Site A were excited at the prospects of 3D printing and were consid-
ering other individualized devices to create in the future. However, they currently see
the task of 3D design and printing to be someone else’s work, and see themselves as
consumers of that work. For these participants, 3D printing is largely meeting their
expectations because the design and test print processes are being obscured from them.
The OTs were interested in learning how to design and create 3D prints, but were
concerned about the required time commitments and learning curve to familiarize
themselves with the software. “I wasn’t sure how much was involved in the [design and
printing] process; it seemed like a lot of work” (OT, Site A). Each therapist works with
multiple students and has limited time with each student. Their interest was piqued
by the prospects of the 3D scanner used in the grip project, but because the grip needed
to be altered to correctly hold the stylus, the OTs again perceived the time to design
and implement these modifications as outside of their capacity.
6. SITE B: PRIVATE SCHOOL FOR BLIND AND VISION-IMPAIRED STUDENTS

Site B is a private organization that provides programs and resources for students who
are blind or vision impaired between the ages of 3 and 21. Some students have multiple
disabilities. The 3D printer at this site is located in a technology classroom.

The 3D printer at this site is a ZPrinter 4508 resin composite machine. This printer and
most of its materials were donated to the school at no cost. Resin composite printers
have a slightly different printing method from the printers discussed so far. In this
printing process, the layers are created by sealing together resin powder with a binding
agent, enabling for higher-resolution, full-color prints (Table II).
For designing, Site B used Rhinoceros 3D,9 which includes both a mouse-driven
interface with more advanced tools than those found in Tinkercad and a command-
line-driven design mode, where students can supply programming commands to draw
shapes. Rhinoceros is a tool for experts with a very different learning curve from open-
source tools like Tinkercad and has a high cost ($1,695). This software was chosen for
its command-line accessibility and supporting tutorials.
6.2. Primary Users

At the time of our interview, the school’s technology instructor had been the primary
user of this printer, owing to the need to gain familiarity with the software and in-
corporate the printer into future curriculum. A small group of students had tried the
8 http://www.3dsystems.com/.
9 http://www.rhino3d.com/.
Fig. 3. Example objects printed at Site B. On the left is a math manipulative project aimed at creating a
tactile Cartesian grid. On the right is a decorative snowman figurine designed by students.
printer and software as part of an extracurricular activity. The instructor has plans to
create and implement lessons for the students in his class in the future.

This site had the printer for 1 year and printed few objects (10 to 20). Example designs
include a snowman figurine and a geometry manipulative (Figure 3), both created
with Rhinoceros 3D. The technology instructor at Site B received a request to create a
geometry manipulative that would allow visually impaired students to tactilely interact
with their math curriculum. Three students from the technology class created the
snowman figurine as part of a software tutorial.

Similar to Site A, this site’s goal is to teach students enough skills for them to become
independent 3D designers, and the instructor had concerns about the students’ ability
to master the modeling software. Specifically, they were concerned that the planned
3D modeling curriculum might prove a challenge for the vision-impaired and blind
students. The instructor anticipated struggles to fully grasping command-line-based
design and issues relating to the build plate and design symmetry. The printer at this
site required very delicate calibration including tasks that bordered on the impossible
for students, such as using a mirror to make adjustments to components underneath
and/or deep inside the machine.
Another goal at this site was to provide tactile graphics and other manipulatives to
students and teachers. Unfortunately, the durability of prints made by this type of 3D
printer was an issue, as objects that come from this printer may not be strong enough
to endure extensive handling. The resin composite prints are very fragile when they
first come out of the print bed. Thick or large objects are often heavy and can easily
break during the final stages of vacuuming off excess powder before applying a final
coat of resin for strength. Additionally, after constant handling, prints can still crumble
or crack, and gluing them back together is not always successful. This is a serious issue
for a population that relies heavily on touch and needs durable prints.
7. SITE C: TECHNOLOGY CENTER

Site C is the technology center within a national organization that provides advocacy
and training for individuals with vision impairments. At this location, we interviewed
an assistive technology expert about the exploration and promotion of 3D printing as
a tool for tactile graphics and supporting alternative access to educational materials,
such as hand-held models.

Site C currently employs two plastic extrusion printers, a MakerBot Replicator and a
Cubex Duo10 (Table II). While other sites focused primarily on the cost or convenience of
a 3D printer when deciding on which model to acquire, this site also focused on customer
support. The technology expert from this site pointed out that many consumer-grade
printer manufacturers are small companies with limited technical support, static FAQs,
or forums that aren’t always accessible. Having a live person to call up and ask for
troubleshooting assistance makes a difference for end-users with visual impairments.
This site uses open-source software wherever possible, including Tinkercad and 123D
Design. Other resources, such as Thingiverse.com and GrabCAD11 (an open engineer-
ing group offering free tools and model-sharing communities), are used to find existing
models.
7.2. Primary User

At Site C, we spoke with a technology expert who provides information about and
training on assistive technology tools related to visual impairment. This expert is the
primary user for this site’s two 3D printers and is self-taught in CAD design and 3D
printing. As part of a small department serving a large population, this participant
leverages 3D printing only when other options are not practical and is the sole user at
this site.

The printers at this site are used for demonstrations, training, and support material for
events put on by the organization. When individuals or groups contact Site C for infor-
mation on 3D printing, the technical expert can provide the seeker with information on
types of printers, hardware costs, software suggestions, and ideas on how 3D printers
can be used to support visual impairment. This primarily includes tactile graphics and
models to support access to information. Due to time and ability constraints, printed
designs are roughly an 80–20 split between open-source designs and creating novel
designs. Figure 4 illustrates some of the models that have been printed at this site.

This site’s current goals for the 3D printer are to create tactile graphics and hand-held
models and provide information for other individuals or organizations that would like
to know more about using 3D printing in support of visual impairments. However, our
participant mentioned the tedious and time-consuming task of mastering CAD. They
also described serious issues regarding support for users.
“For something that’s supposed to be consumer-grade, that’s just not workable. Their
support is good, but for some things I have to wait three days for a response!” (Tech.
Expert, Site C). Other possibilities, such as 3D scanners, are perceived as promising
ideas, but likely not as practical as advertised. “I’m very wary because I feel like so
much of it is marketed as, ‘Oh it will scan and you’ll get a print-ready model’ and I’m
very skeptical of that” (Tech. Expert, Site C).
8. 3D PRINTING STAKEHOLDERS IN SPECIAL EDUCATION

Stakeholders were identified based on in-class observations and on the structure of
the different sites utilizing 3D printers. We recognize that this is by no means a com-
plete list of stakeholders. In particular, we acknowledge that parents, other caregivers
10 http://www.cubify.com/.
11 http://www.grabcad.com/.
Fig. 4. Models printed at Site C to support information access for visually impaired users. Pictured: a planet,
the Sphinx, the Chrysler building, a DNA helix, & a landmass.
outside of the school setting, and the children themselves should be involved in this
research. For this study, we had to exclude these populations for reasons of access and
other constraints that are discussed in Section 10.5. Instead, we are focusing on the
faculty and staff providing education and access to tools such as 3D printers, in addi-
tion to leveraging these tools themselves as a means of contributing to the quality of
life and education for their students. To follow, we provide a brief description of each
stakeholder group and their high-level reflections on 3D printing technology in special
education settings.
8.1. Administrators’ Perceptions and Expectations

Administrators include those involved in management of personnel, curriculum design,
and/or site planning. This group had little to no hands-on experience with 3D printing
but was often involved in deciding how or why a 3D printer was added to their location.
They shared high-level thoughts about safety, reliability, and educational outcomes for
3D printing. Overall, they perceived printers as safe to use with supervision. Their
only concerns were related to safety regulations and standards for equipment, but
they felt that printers were well within the guidelines they already had in place. One
administrator pointed out that the school had a cooking class and felt an oven was
equally or more dangerous than a 3D printer. This perception could indicate a lack of
understanding, as most printers have parts that heat up to temperatures much higher
than a conventional oven. The administrators’ lack of experience with the printers may
be coloring their expectations and the needs of the employees they supervise.
We interviewed four administrators (two assistant principals, one vice principal, and
one curriculum director). These stakeholders had the least interaction with students
and also the highest expectations of what students could do with a 3D printer. Ad-
ministrators described optimistic scenarios in which students would have the ability
to design and produce custom objects with ease, providing design and print services
to other parts of the school campus and even to the local community. These admin-
istrators were interested in students making both educational aids and assistive ob-
jects. “[3D printed AT] is the thing that’s going to revolutionize care for people. Truly
individualized solutions” (Admin., Site A). In addition to creating low-cost AT solutions

and classroom manipulatives, they perceived this technology as affording students with
employability skills and as a means of practicing math and engineering. An adminis-
trator at Site A suggested that 3D modeling and printing “. . .presents engineering in
a manageable way” and went on to explain that the student body “. . .deals best with
hands-on approaches to learning.”
While some administrators understood the time constraints of 3D printing, few had
concerns about the amount of time necessary for students or staff to learn 3D design
and printing tools. They described a need for training or workshops for faculty in order
to spread awareness of what 3D printers could do and to discuss the best classroom to
host printers. Administrators were interested in a catalog where staff members could
choose from an existing set of objects for their classrooms or place special requests that
the students learning 3D printing would design and fulfill.
8.2. Occupational Therapists’ Perceptions and Expectations
Our investigation sites offered different services to students (depending on student
need and resources) including therapy. We initially interviewed two occupational ther-
apists at Site A, and later we returned to this site and got feedback from three additional
OTs as part of our GripFab study. General feedback we received about 3D printers from
all OTs is included in this section. These participants had limited exposure to 3D print-
ing, and their knowledge of printers came from recent media coverage on printers or
from local sources such as other staff members, friends, or family who told them about
the technology.
This stakeholder population offered potential applications for 3D printers creating
adaptive or assistive devices both for therapy sessions and everyday interactions. Ob-
jects they considered for 3D printing included the grips we explored in our design
case study, modifications to devices and AT such as cases and screen guards, and spe-
cialized fidgets (small objects with textures and/or articulated components that can
be “fidgeted” with) for students with sensory and attention needs. They were particu-
larly keen on the idea of customizing objects for their students: “. . .because it can be
customized it opens the door to these things [individualized solutions].” The OTs felt
there were likely more applications, but that they would need time and exposure to
the printer before they were able to realize more possibilities. Therapists also gave us
more information on the costs of customized AT and potential safety concerns for 3D
printed objects in this context.
Therapists at this site indicated that their biggest concern with 3D printing was the
time investment to learn the design skills and software tools to create novel 3D-printed
objects. At the onset of our interviews, the OTs were unfamiliar with the time needed to
create medium and large prints. “[I] thought I could go to someone and say, ‘Hey I need
this,’ and they could print it in 15 minutes. . ..” After we described the time constraints
on these prints, the OTs agreed that a 3D printer was still a useful tool. However,
they wanted to reduce the trial and error of their designs as much as possible and felt
that this level of design efficiency would only come with mastery of 3D modeling tools.
This brought the conversation back to their own time constraints working with a large
student population.
These participants had very few safety concerns about the use of the printer or of
3D-printed objects. They cited existing protocols for testing and vetting any new piece
of assistive or adaptive technology and the close observation of students already in
place at a special education school. The OTs agreed with administrators about the
need for workshops or training for faculty to help promote understanding and brain-
storming about the use of 3D printers in their schools. They indicated that 3D-printed
solutions to accessibility challenges would be considerably less expensive. Therapists
were interested in alternative materials for 3D-printed objects, including soft textiles
and rubber- or foam-based filaments to create easy-to-grip or flexible objects.
8.3. Technology Instructors’ Perceptions and Expectations

This group represented the primary point of contact with the 3D printers at each lo-
cation. These stakeholders are most versed in the design, printing, and maintenance
processes. Instructors provided their war stories about printing in an educational set-
ting. This group spoke to challenges for youth in the design and printing process, as
well as the shortcomings of the software and hardware involved, and shared their
ideas for improvements and new uses for the printers. This helped us juxtapose the
desires of the other stakeholders versus the realities of the technology, particularly in
the case of stakeholders who were far removed from the printing process, such as the
administrators.
We interviewed three technology instructors, two from Site A and one from Site B.
These participants shared several challenges related to teaching 3D printing in a spe-
cial education classroom. These stakeholders identified time management, software
complexity, and hardware reliability as serious obstacles to successful 3D printing
curriculum. Technology instructors indicated that they did not have enough time as
educators to master 3D design and print technologies and that they also struggled to
create enough time for 3D printing in their existing, packed technology curriculums.
They described steep learning curves for their students when teaching 3D modeling
software, student difficulties using mouse-driven design interfaces, and problems man-
aging expectations and maintaining student motivation during the design process.
The instructors at Site A struggled with persistent hardware failures that impacted
instructor and student morale and engagement with the technology. The students
at Site A are most familiar with procedural education where scheduling and task
completion are an important part of the learning environment. The instructors stated
that printer failures or struggles with the design software were very disruptive: “We do
A, B happens, and C is what you get. With everything they do. If that doesn’t happen,
you lose them” (Tech. Instructor). While we were able to supply a more reliable printer,
the instructors continued to manage student expectations.
8.4. Nontechnology Instructors’ Perceptions and Expectations

Two of our investigation sites offered academic courses, including classes on history, art,
mathematics, music, and so forth. We interviewed one nontechnology instructor at Site
A. Our goal was to identify perceptions of 3D printing and to discover alternative uses
for the printers not currently being explored at the investigation sites. This instructor
taught art but had a very antitechnology stance for his classroom.
The art instructor described admiration for 3D printing and similar technologies,
but stated that he preferred hand-crafted arts and his only interest in 3D printing
for his classroom was an alternative means of sculpting or painting during his three-
dimensional art unit. “They always use computers at home. In here [art class room], I
want them to get an experience they don’t get at home. Hammers, chisels, other tools. I
don’t want them to use the computers” (Nontech. Instructor, Site A). Despite interfaces
that use scanners to transform literal hand-made 3D objects into digital designs or
modeling software that provides users with the metaphor of a clay block to be directly
manipulated, this instructor insisted that 3D printed objects were not part of his
traditional definition of three-dimensional art. A single instructor is not representative
of all nontechnical teachers; however, the example highlights that not all educators are
interested in this technology.
8.5. Technology Specialists’ Perceptions and Expectations

Technology specialists at investigation Site A provided support to their colleagues and
to students. The specialists at Site A had a cursory knowledge of 3D printing, but they
were not the primary users at their site. At Site C, the technology specialist was the
primary user and assistive technology expert. By supporting special education needs
via technology, these stakeholders regularly evaluated and informed the applied use of
a variety of assistive technology solutions. They gave insights into costs, alternatives,
and appropriate deployment for 3D printer technology.
We interviewed three assistive and information technology specialists. Their per-
ception and interest in 3D printing aligned with the therapists at Site A. Like the
therapists, they had few worries about the safety of 3D printers and instead were con-
cerned with the time and personnel needed to operate it. Unlike administrators, the
specialists at Site A did not necessarily see students as the lead designers of AT devices.
They saw, instead, a specialized position somewhere between AT and OT in which an
individual or a team was dedicated to creating assistive solutions with the printer. The
specialist at Site C echoed similar thoughts, describing a high time commitment and a
need to have at least one person at any institute trained explicitly and extensively on
how to use the printers and 3D modeling software.
This group indicated that 3D printed assistive solutions would be a tremendous
cost savings over existing products. “There’s a lot of assistive technology that’s so
expensive—it’s so expensive! And it’s because it’s such a niche—a super niche market.
Even something as simple as an adaptive penholder is going to be ridiculously expen-
sive. [I see] a lot of potential for the 3D printer to come in and create these things that
people could use for much cheaper” (Tech. Specialist, Site A). They also wanted more
variety in printable materials to support a range of AT products, including mixed mate-
rials, textiles, rubber, and metal. One specialist who works mostly in AT described the
struggles of making individualized AT devices. “How [to] customize something when
the standard isn’t appropriate for a particular person? . . . We do try to make them
[custom AT] and making them takes time” (AT Specialist, Site A).
In addition to using the printer to create AT objects, these participants identified
the printer as a way to support UDL by creating difficult-to-find, nonexistent, or costly
educational aids and manipulatives. They also described a need for more open-source
designs of such objects. Similar to other stakeholders, these specialists favored a collec-
tive set of designs made available freely to educators and care providers to more easily
create and print AT and UDL objects in 3D.
9. DESIGNING ASSISTIVE TECHNOLOGY 3D MODELING SOFTWARE FOR THERAPISTS

After working closely with the OTs at Site A through interviews and the codesign of
a novel stylus grip for their students, we developed a prototype 3D modeling software
to enable them to easily create 3D-printed accessible hand grips. In this section, we
describe the motivation, initial requirements gathering, and findings from demonstra-
tions and focus groups of this software with additional OTs at Site A.
9.1. Need for Customized 3D Modeling Software

As described in Section 5, we worked in cooperation with two OTs to create a 3D-
printed customized stylus grip for a student. Our grip design fit the student’s hand
and stylus well and was considered a success. However, the student found that the
stylus itself was not as responsive to her touchscreen device and purchased a new
and larger stylus that did not fit our grip. The OTs asked us if it would be possible to
modify the grip design to accommodate this new stylus. This request highlighted the
common problem that many AT devices are outgrown or require changes or updates to
keep pace with changing electronics. Other examples of changing AT needs provided
by OTs included students breaking devices or cases and student families purchasing
new devices requiring new cases, mounting, and so forth.
The OTs made it clear to us that more requests like this one were likely, and we
needed to provide a design tool that would empower them to build other custom devices.
Given the OTs’ reticence to spend time learning existing software, we needed a solution
to quickly and easily create variations on designs to accommodate change.
9.2. Requirements for 3D Modeling Software for Special Education Settings
From our interviews and design work with the OTs, we came up with a basic set of
requirements for a minimal design tool. Even though the hand grip was explicitly cre-
ated for one student, the therapists indicated that multiple students could benefit from
a custom grip modification for an array of objects. We designed GripFab, a simplified
design tool for generating customizable hand grips. In our initial conversations with
OTs, we recognized that we needed to create an interface that would (1) minimize
time commitment, (2) offer simple customization without mouse-driven design to im-
prove model accuracy, and (3) obscure as much complexity as possible to mitigate the
intimidation factor associated with CAD modeling tools.
9.2.1. GripFab Features. To fulfill the aforementioned requirements, we abstracted 3D
model generation down to parametric fields and option-selection settings. GripFab uses
a tabbed interface to break up the design process emulating a step-by-step wizard. This
forces the user to consider sections of their grip separately and limits the overwhelming
number of options available to the user. Each tab offers the user a small set of options
related to each feature of the hand grip: (1) selecting the base shape, (2) adjusting the
dimensions of the hole in the grip accommodating the object to be held, and (3) optional
support called a “barrel” that can be used for holding longer objects or to provide an
extension for a user with limited range of motion. GripFab combines Java and an open-
source, console-based design tool called OpenSCA12 to apply the designer’s settings
and render a model as an .stl file. All necessary files for rendering are saved locally
and then removed once the .stl is fully rendered. This interface masks the rendering
process from the user, leaving a clean .stl file for printing.
9.2.2. GripFab Version 1 - Pilot Study Feedback. We pilot tested the first version of GripFab
in an hour-long focus group at Site A with four OTs, including the two involved in the
original hand-grip design project. The session included a brief demonstration that
was followed by a set of semistructured questions regarding the OTs’ experiences
with hand motor impairments and their current practices for devising solutions for
accessibility challenges. Participants were shown an example grip designed to hold a
spoon (Figure 5) and asked to make a grip design for a large highlighter using GripFab.
One participant controlled the software while the other three provided feedback on the
usability of the application.
The OTs gave favorable reviews of the software and grips and provided us with a
list of improvements. To minimize the time commitment, the OTs requested a profile
system that would load default values about a student (e.g., left- or right-handed). They
requested dynamic images to help visualize the models, but they did not want to use
these images for direct manipulations of models. For the software, they requested para-
metric field and tool labels with more OT-friendly language to help alleviate confusion
over CAD vocabulary. The OTs also described a need for more diverse base grip shapes
and provided us with a set of art supplies spanning several shapes and dimensions,
including pencils, markers, paintbrushes, glue sticks, and other supplies.
12 http://www.openscad.org/.
Fig. 5. A 3D printed assistive hand grip for a spoon.
Fig. 6. Screenshot of GripFab showing variables related to accessory width.
9.2.3. GripFab Version 2 - Focus Group 1 Feedback. We implemented a few changes to

GripFab and also created several potential base grip designs to accommodate the art
supplies provided to us at the end of pilot testing at Site A. Changes to GripFab included
the incorporation of two new base models and the ability to angle the barrel support
(Figure 6). The new printed hand grips were a mixture of novel designs based on needs
described by the OTs in the pilot study and modifications of existing designs from AT
catalogs and Thingiverse. We conducted an official focus group with three OTs and one
art teacher, including two OTs from the pilot study.
Again, we saw emphasis on time savings, simplifying complex design techniques,
and obscuring complexity. Even though the OTs praised the step-by-step process of
Table V. Top Five Grip Designs as Selected by OT and Art Faculty at Site A: (1) Refined Pinch
Holding a Highlighter, (2) Articulated Container Grip, (3) Bottle Grip, (4) Rocker Grip Holding
a Large Crayon, and (5) Built-Up Grip Holding a Thin Paintbrush
creating grips, they suggested providing a tutorial to help guide them through the
design process. They also requested new features, such as the ability to change the
scale of the base model and increase/decrease the width, length, and/or overall size of
the base grip to provide more flexibility and customization of the grips. Previously, the
ability to scale or stretch a design has been controlled during the slicing step between
generating the 3D model and preparing it to print. By reducing the number of programs
they must interact with, the OTs are again looking to reduce complexity even though
this is adding in more features to the design interface. Concerned about the print time
necessary to create several variations of a grip, the OTs also wanted the option to print
alternative interlocking barrels on preexisting grips. Ideally, they would have a base
model that fit a particular student’s need and they would print hot-swap barrels to
accommodate all the different hand implements the student interacted with during
the school day.
9.3. Utility of 3D-Printed Assistive Grips

In addition to learning about the requirements of design software, we also identified
high-level feedback on creating assistive hand grips. In our second and final official
focus group, we brought 12 different grip base designs ranging from bases modeled after
household tools like screwdrivers, existing grasping tools sold by AT manufacturers,
and grip designs found on Thingiverse. We asked the OTs to evaluate the usefulness
of each grip model and whether or not it would make a beneficial addition to the base
model library in GripFab.
The five grips the OTs described as most useful were the rocker, refined pinch, built-
up, articulated container grip, and bottle grip (Table V). The attributes of these grips
gave the user flexibility, a steady comfortable grasp, and mobility support. The refined
pinch is a generalized version of the custom grip design we created for the student
in our codesign project. The articulated container grip was made explicitly for the
operation of a glue stick. The grip clasps around the glue stick, providing the user with
the ability to turn the glue stick upside down, sideways, and upward. The bottle grip
is an existing design available on Thingiverse [Minifacture 2013] that we selected for
thicker objects and containers found in the art classroom, specifically a small bottle
containing glitter. The rocker grip is a T-shaped grip based on existing AT designs for
persons who have some grasping ability but only lateral wrist movement. A built-up
grip is an exaggerated handle based on a technique described by the OTs wherein they
wrap material around nonassistive handles to make them easier to grasp for people
with limited dexterity.
The OTs preferred to have generalized grips over user-specific ones, a default average
adult hand size, and the ability to attach extra support materials to the grips to give
the user more stability when interacting with objects. For instance, the extra support
material would allow them to take the lid off the jar, hold it, and pour with stability. The
OTs stated that they would like to use GripFab and a 3D printer in their daily practice if
we can incorporate the remaining feedback on the interface and add these five designs
as base models. They described a 2-week waiting period for ordering and receiving AT
products from catalogs and manufacturers, suggesting that 3D-printed options would
save them wait time. The OTs also said that based on their understanding of 3D printer
material costs, using 3D-printed grips would also be more cost effective.
9.4. Implications for Self-Designed Assistive Technology and Disruptive Service Delivery
GripFab is one example in a movement toward self-designed assistive technologies.
To a certain extent, customization has always been a component of AT design. Ther-
apists and end-users regularly make low- and medium-fidelity modifications to AT by
incorporating tape, clay, cardboard, Velcro, and other artifacts to improve the fit and
utility of off-the-shelf solutions. Not all people with disabilities have the level of motor
ability and dexterity necessary to make use of these materials. At the same time, con-
temporary design tools also have barriers to access, relying heavily on mouse-driven
techniques and requiring training in CAD to generate novel designs in 3D. Tools like
GripFab or the Thingiverse Customizer are part of an emergent toolset that supports
designers who can access a computer. Our findings indicate that the use of parametric
settings may enable a wider user population to create bespoke AT objects, but we have
not yet completed extensive testing to discern which populations will benefit most from
this practice and if this modality is the best option over other modalities for specific
user abilities and level of design skills.
This accessible self-design concept does not need to be limited to 3D printing. In-
stead, we propose that the new model for manufacturing be made more inclusive by
way of accessible fabrication tools. With the creation of accessible fabrication, we can
support and promote end-user self-design of AT. When combined with the user’s con-
structionist learning from the use and design of these items, this creates a disruption
to existing service delivery and service design. We, as HCI researchers, can then foster
collaboration between end-users and clinicians in the creation of novel or custom AT.
By empowering the end-user as a designer of his or her own AT solutions, we can also
incite engagement, leading to better adoption and reduced abandonment rates.
10. DISCUSSION
These studies reveal that there is support and interest in using 3D printing in special
education and therapy. Through this work, we identified several recommendations for
technologists related to the development of 3D hardware and software, we identified
issues of interest to schools and therapists, and we provide our lessons learned studying
3D printing technology in special education settings.
10.1. 3D Printer Accessibility Recommendations
We believe it is important that the 3D printer be considered as an assistive technology
in the special education classroom and not simply another piece of technology. We make
the following recommendations specific to this cause.
10.1.1. Provide Accessible Feedback. One significant obstacle we observed for novices
learning to use this technology was developing an accurate mental model of the printer.
Common problems included not understanding why prints failed, the limits of a given
printer’s capabilities, and the relationships between troubleshooting techniques and
the problems they are intended to solve. As we move toward a future where students
with diverse abilities use these printers, it is important to provide accessible feedback
about when the printer is working and when there are problems.
10.1.2. Printer Safety. 3D printers can be dangerous when misused. While we did not
see specific concerns from any of our stakeholders about the safety of these machines
for students, it should be considered for future audiences. All of the 3D printers our
participants interacted with had enclosed all hot and/or moving parts, so they would be
difficult to touch while the printer was on. Additionally, students at these sites are mon-
itored very closely, and the administrators felt that it was the teacher’s responsibility
to keep students safe.
10.1.3. Offer Appropriate Customer Support. Given that 3D printing is not yet as robust
as other consumer technology, it is important to provide consumers with appropriate
support. While this is true for all end-users, it is particularly true for assistive technolo-
gies where end-users are accustomed to having access to experts, technical support,
and repairs. This was made clear to us in our interview at Site C, where purchasing
decisions were informed based on available support.
10.2. 3D Modeling Accessibility Recommendations
10.2.1. Make Accessible Software. As 3D printing becomes more common, 3D modeling
software must be accessible. Specifically, this software must support screen readers,
switch input, and other common computer access customizations. According to our
participant at Site C, there is currently no open-source 3D modeling software that is
accessible to a screen reader. The accessibility of these tools should also be considered
for end-users who have difficulty remembering complex task sequences or have limited
short-term memory.
We suggest GripFab as one example created toward increased accessibility. The
OTs found contemporary CAD tools to be inaccessible based on their skillsets and time
affordances. By identifying design needs and discussing interface options, we were able
to create a single-purpose program that enables the OTs to create and modify 3D models
without extensive training or CAD experience. This is not to say that technologists
should create a single design tool for every special case of 3D printing, but instead
provide guidelines for identifying needs and skills of special designer populations and
working with those individuals to arrive at more approachable software.
10.2.2. Consider the Learning Curve. It would also be prudent to explicitly design 3D
modeling software to support a range of expertise and to support the transition from
novice to expert performance. While many of our participants were excited to use 3D
printing, most found the current 3D modeling software intimidating. For these users,
novice tools don’t provide enough control, but expert tools come at too high of a cost in
terms of outright expense, learning, and time investment. Adding optional features to
support novices such as detecting unprintable designs and integrating tutorials into
the modeling software might assist users when they are struggling.
10.2.3. Encourage Sharing of Existing Models. Not all end-users want to create custom
designs, and it is often enough to make minor tweaks to existing object models. Faculty
who were not directly involved in teaching technical topics felt they would be more
comfortable using a catalog of existing items, similar to Thingiverse.com, rather than
learning CAD. This concept was echoed by administrators at Site A, who felt that a
central repository of tools, designs, and support would be the best fit for the teaching
faculty not currently involved with the 3D printer. Several participants wanted cur-
riculum support, such as miniature models or student-tailored objects that could be
selected from a list and printed on demand.
10.2.4. Support Editing Existing Models. A mix of existing 3D modeling tools or minor
modifications to existing tools may be enough to support some of the stakeholders
discussed in this article. While we are starting to see customizable 3D models (such as
Customizer on Thingiverse), these tools are limited. 3D scanning physical objects may
offer an easier way to create custom designs; however, many current 3D scanners have
low resolution or require complex cleanup to make a printable model.
10.3. Recommendations for Special Education and Accessibility Organizations

We saw a range of exposure, comprehension, and skills related to 3D printing that
impacts decisions about choosing appropriate tools for modeling and printing. To ac-
commodate this, we recommend carefully surveying the skills and application goals of
users before choosing a printer or set of modeling tools.
10.3.1. Budget Time for Training. Participants with 3D printing experience expressed
concerns about the time necessary to train themselves, other staff, or students to
create novel 3D designs. They feared mastering these tools would take days, weeks,
or even months. This is concerning given that our participants consider learning 3D
printing as “extracurricular” and not what they were explicitly hired to do.
10.3.2. Consider Printer Reliability and Maintenance. As is a common complaint with the
recent 3D printing boom, there are serious setbacks in the consistency and quality
of printer performance [Lutz 2013]. These shortcomings include maintenance, trou-
bleshooting, and unanticipated poor-quality print outcomes.
These issues are crucial for accessibility and education applications. For example,
when making custom-fitted assistive devices, precision is key to successfully replicating
details or measurements. When teaching 3D design and prototyping to students in
special education, consistency and reliability impact the learning experience. Students
with visual or cognitive impairments may already endure obstacles to accessing the
design and modeling end of 3D printing; adding the complexity of unreliable print
output is an additional discouraging factor.
10.3.3. Develop a Plan to Share Resources Equally. The placement of a 3D printer in a
school is important and impacts who will use it. We believe that since the 3D printer
at Site A was physically located in the technology classroom, the therapists were more
hesitant to use it than if it had been placed in their space. As pointed out by the
technology instructors at this site, the students are sensitive to disruptions in daily
classroom activities. Having the printer running on a nonprinter day could distract
students from their other learning goals. This, combined with the amount of time
needed to print medium to large objects, hindered the OTs’ ability to use the printer in
the classroom. As the market of consumer 3D printers under $500 continues to grow,
it may make sense for educators to invest in multiple inexpensive printers rather than
one high-end printer. This would enable higher production rates for printed objects
and reduce conflicts of use like those described between the OTs and the technology
classroom.
10.4. Recommendations for Therapists

10.4.1. 3D Printing Custom-Fit Objects Is Possible. Our design sessions with Site A’s OTs
demonstrated how 3D printing can provide cost-effective and highly individualized
assistive technology solutions, although extensive help is currently required. If we can
empower these therapists to design and prototype solutions on their own, they can
utilize in-house 3D printing to overcome several challenges in assistive technology
such as limited availability, high costs, and poor fit.
10.4.2. Budget Extra Time for Iteration and Fitting. While it is possible to create custom
accessibility solutions, the time between iterations from design to prototype can be
slow. Even though it only took 2 hours to 3D print the custom stylus grip discussed
earlier, the development took longer than expected. While design and print time were
within reason, this project unfolded over the course of several months. The ultimate
limiting factor on this project was access to the student, who was only available for
testing during short therapy sessions. We were restricted both by our access to the
student and by the student’s availability to test out the grip. We believe this process
could be faster if the end-user were able to be more involved in the design process.
10.5. Lessons Learned (for Researchers)

After nearly 2 years of studying 3D printing in special education settings, we have
noted pitfalls and workarounds that may be insightful for future researchers in this
area.
10.5.1. Access to Participants. Students in special education are a protected population,
as they are both minors and persons with disabilities. It is important to find out during
the early planning stages of a study what, if any, access the investigators will be granted
with the children in special education settings. Our investigations were often limited
to observation-only interactions with children.
Our strategy to gain insights about the experiences and the needs of the children as
users of 3D technology and 3D-printed assistive devices relied heavily on adult faculty
collaboration. We worked as partners with teachers and therapists in order to validate
our observations and to gain knowledge via their hands-on interactions with students.
This also afforded us more long-term wisdom to help sort out behaviors or obstacles
that are specific to particular students rather than more generalizable to an ability
level.
10.5.2. How Much Support to Offer. In order to get a realistic view of the trials associated
with 3D printing in special education, we had to be careful about the type of support we
provided to our participants at Site A. The other investigation sites we worked with had
access to 3D printing prior to our study and did not look to us for technical support.
By introducing a 3D printer to Site A, we became the primary source for technical
support. Especially in the early months of the study, when the technology instructors
were having repeated mechanical failures with their printer, we felt obligated to step
in. We recognized that we were quickly losing participant buy-in with the technology
and were able to replace their printer with a similar but more reliable model from the
same printer family.
Despite the increasing popularity and the decreasing price of consumer-grade 3D
printers, there are still issues of printer reliability. Among the problems that continue
to plague 3D printer users are issues of file formatting and slicing errors (wherein
problems with the geometry of a 3D model cannot be correctly translated to slices

for the printer), prints failing to adhere to the print bed, clogs or jams in the heated
extruder, and low-quality filaments entering the consumer market with manufacturing
defects that can negatively impact print success or print quality. In the real world, these
issues continue to impact educators of all student types. With this in mind, we tried to
encourage our participants at Site A to conduct their own troubleshooting and problem
documentation after we replaced their printer.
We found that while participants were quick to complain about inconsistencies with
their printer, they were also very invested in solving issues on their own. It became
evident the day-to-day maintenance of the printer was too much for the instructors at
times, but that their determination to be self-sufficient prevented them from reaching
out to us for help and slowed the progress of their class.
We suggest itemizing the support and maintenance needs of 3D printers with partic-
ipants prior to study deployment. Promotional material for consumer-grade 3D print-
ers makes the technology sound very plug-and-play and user friendly, but the reality
for our participants was closer to the trial and error sometimes encountered with
medium-fidelity prototypes. It is important to prepare the participants for the reliabil-
ity of this technology without influencing their perception and willingness to use the
printers. Talking through potential errors and hardware complications ahead of time
helps manage participant expectations and creates defined roles in the troubleshooting
process.
10.5.3. Financing Technology. On top of user training and the maintenance aspects of a
3D printer, we found that there are also several financial considerations when deploying
this technology in the wild. Spare parts and miscellaneous tools need to be available to
participants to care for the machine and polish off their prints. This includes extruder
nozzles and print bed coatings for the machine, wire cutters for filament, and items
like pliers, scrapers, and sanders for removing objects, snapping off support material,
and smoothing any rough surfaces from completed designs. The filament itself needs
to be kept in supply and participants will likely want multiple colors.
We found it useful to maintain an inventory of supplies provided and requested our
participants at Site A to maintain a log of printer activity and repairs. Similar to diary
or journal-style studies, we found that printer log information provided by participants
dropped off after a short period of time and rarely contained all of the data necessary
to keep accurate track of the materials being used. Still, monitoring the supplies and
checking digital history of print jobs stored on the printer allowed us to keep tabs on
the amount of plastic used based on total print time for the machine.
A further consideration for us was the fundraiser component of the Site A technol-
ogy classroom. Because the class had an industry model wherein the students were
intended to create and sell commodities based on the technical skills they were using,
they were then printing and selling objects using materials provided for the study. One
of our researchers made a habit of photographing and documenting items for sale to
help track the designs being printed and to account for the materials used. The school
was not asked to recoup costs or provide its own materials; instead, it was assumed
that use and sale of printed objects were benefits of being participants in our study.
11. CONCLUSIONS AND FUTURE WORK

Our research has identified benefits of 3D printing in special education environments,
and it has uncovered several obstacles to adoption. As 3D printing technology pro-
gresses and becomes more reliable, we believe there are opportunities for STEM en-
gagement in children with varying abilities, providing tactile access to information
and educational content, and encouraging DIY and in-house assistive device design. In
extension of this work, we have also presented GripFab, a prototype modeling software
to support the design of individualized AT. Findings from the preliminary testing of
GripFab with therapists suggest that by offering these users appropriate tools and
support, customized assistive solutions are a viable application of 3D printers.
We have also offered new insights into the perceptions of various stakeholders in
special education with respect to 3D modeling and printing. To encourage the adoption
of this technology in special education, we identified barriers and points of considera-
tion for 3D printer manufacturers, 3D modeling software developers, special education
institutes and accessibility organizations, and therapists. These insights have been
updated from our earlier work to include explicit lessons learned on conducting field
research with 3D printers in special education environments and designers with dif-
ferent abilities.
In the future, we will continue to work with a wider range of instructors in non-
technical fields and work directly with students. We will also expand on our work with
therapists and other medical professionals using 3D printing in support of accessibility
and dynamic AT design. This will include further development of GripFab and simi-
lar tools coupled with longitudinal field deployments to evaluate the usability of the
software and the practicality of the objects made by 3D printers.
ACKNOWLEDGMENTS
We thank the schools and institutes involved, as well as all of our individual participants. We also thank
Shaun Kane and the members of the UMBC PAD Lab for their support with this research.
REFERENCES
Paulo Blikstein and Dennis Krannich. 2013. The makers’ movement and FabLabs in education: Experiences,
technologies, and research. In Proceedings of the 12th International Conference on Interaction Design
and Children (IDC’13). 613–616.
Craig Brown and Amy Hurst. 2012. VizTouch: Automatically generated tactile visualizations of coordinate
spaces. In Proceedings of the 6th International Conference on Tangible, Embedded and Embodied Inter-
action (TEI’12). ACM, New York, 131–138. DOI:http://dx.doi.org/10.1145/2148131.2148160
Erin Buehler, Shaun K. Kane, and Amy Hurst. 2014. ABC and 3D: Opportunities and obstacles to 3D
printing in special education environments. In Proceedings of the 16th International ACM SIGACCESS
Conference on Computers and Accessibility (ASSETS’14).
Jodie Copley and Jenny Ziviani. 2004. Barriers to the use of assistive technology for children with multiple
disabilities. Occupational Therapy International 11, 4 (January 2004), 229–243.
Melissa Dawe. 2006. Desperately seeking simplicity: How young adults with cognitive disabilities and their
families adopt assistive technologies. In Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems (CHI’06). 1143–1152.
Sean Follmer, David Carr, Emily Lovell, and Hiroshi Ishii. 2010. CopyCAD: Remixing physical objects with
copy and paste from the real world. In 23rd ACM Symposium on User Interface Software and Technology
(UIST’10). 381–382.
Gillian R. Hayes, Sen Hirano, Gabriela Marcu, Mohamad Monibi, David H. Nguyen, and Michael Yeganyan.
2010. Interactive visual supports for children with autism. Personal Ubiquitous Computing 14, 7 (April
2010), 663–680. DOI:http://dx.doi.org/10.1007/s00779-010-0294-8
Yingdan Huang and Michael Eisenberg. 2012. Easigami: Virtual creation by physical folding. In Pro-
ceedings of the Sixth International Conference on Tangible, Embedded and Embodied Interaction
(TEI’12), Stephen N. Spencer (Ed.). ACM, New York, NY, USA, 41–48. DOI:http://dx.doi.org/10.1145/
2148131.2148143
Amy Hurst and Shaun Kane. 2013. Making “making” accessible. In Proceedings of the 12th International
Conference on Interaction Design and Children (IDC’13). ACM, New York, 635–638. DOI:http://dx.
doi.org/10.1145/2485760.2485883
Amy Hurst and Jasmine Tobias. 2011. Empowering individuals with do-it-yourself assistive technology. In
Proceedings of the 13th International ACM SIGACCESS Conference on Computers and Accessibility
(ASSETS’11). ACM, New York, 11–18. DOI:http://dx.doi.org/10.1145/2049536.2049541
Shahram Izadi, Richard A. Newcombe, David Kim, Otmar Hilliges, David Molyneaux, Steve Hodges,
Pushmeet Kohli, Jamie Shotton, Andrew J. Davison, and Andrew Fitzgibbon. 2011. KinectFusion:
Real-time dynamic 3D surface reconstruction and interaction. In ACM SIGGRAPH 2011 Talks
(SIGGRAPH’11). ACM, New York, NY, USA, Article 23, 1 pages. DOI:http://dx.doi.org/10.1145/2037826.
2037857
Shaun K. Kane and Jeffrey P. Bigham. 2014. Tracking @stemxcomet: Teaching programming to blind stu-
dents via 3D printing, crisis management, and twitter. In Proceedings of the 45th ACM Technical
Symposium on Computer Science Education (SIGCSE’14). ACM, New York, NY, USA, 247–252.
DOI:http://dx.doi.org/10.1145/2538862.2538975
Adam Kemp. 2013. The Makerspace Workbench: Tools, Technologies, and Techniques for Making. Sebastopol,
CA: Maker Media.
Dennis Krannich, Bernard Robben, and Sabrina Wilske. 2012. Digital fabrication for educational contexts.
In Proceedings of the 11th International Conference on Interaction Design and Children (IDC’12). New
York, NY: ACM Press, 375–376. DOI:http://dx.doi.org/10.1145/2307096.2307174
Ben Leduc-Mills, Jaymes Dec, and John Schimmel. 2013. Evaluating accessibility in fabrication tools for
children. In Proceedings of the 12th International Conference on Interaction Design and Children –
(IDC’13). New York, NY: ACM Press, 617–620. DOI:http://dx.doi.org/10.1145/2485760.2485882
Ben Leduc-mills and Michael Eisenberg. 2011. The UCube: A child-friendly device for introductory three-
dimensional design. In Proceedings of the 10th International Conference on Interaction Design and
Children (IDC’11). 72–80.
Sylvia Libow Martinez and Gary Stager. 2013. Invent to Learn: Making, Tinkering, and Engineering in the
Classroom. Constructing Modern Knowledge Press, Torrence, CA.
Robert Lutz. 2013. Enhancing information technology education (ITE) with the use of 3D printer technology.
In Proceedings of the 14th Annual ACM SIGITE Conference on Information Technology Education
(SIGITE’13). ACM, New York, NY, USA, 157–158. DOI:http://dx.doi.org/10.1145/2512276.2512327
Andrew Manches and Claire O’Malley. 2011. Tangibles for learning: A representational analysis of
physical manipulation. Persaonl Ubiquitous Computing 16, 4 (July 2011), 405–419. DOI:http://dx.
doi.org/10.1007/s00779-011-0406-0
Minifacture. 2013. Adjustable Bottle Cap Opener / Cap Wrench. (2013). Retrieved January 20, 2015, from
http://www.thingiverse.com/thing:40131.
National Science Foundation. 2013. Women, Minorities, and Persons with Disabilities in Science and Engi-
neering: Special Report NSF 13-304.
Irene Posch and Geraldine Fitzpatrick. 2011. First steps in the FabLab: Experiences engaging children. In
Proceedings of the 24th Australian Computer-Human Interaction Conference (OzCHI’12). 497–500.
David Rose, Anne Meyer, and Nicole Strangman. 2002. Teaching Every Student in the Digital Age: Universal
Design for Learning. Association for Supervision & Curriculum Development.
Tides Center. 2014. Maker Education Initiative. Retrieved from www.makered.org.
Received March 2015; revised August 2015; accepted December 2015

Buehler, Erin Et Al

Uploaded by

Copyright:

Available Formats

Buehler, Erin Et Al

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Buehler, Erin Et Al

Uploaded by

Copyright:

Available Formats

Investigating the Implications of 3D Printing in Special Education

ERIN BUEHLER and NIARA COMRIE, University of Maryland, Baltimore County

2.1. Education and Making

2.2. Making and Assistive Technology

Table I. Participants Interviewed at Each Site, Listed by Role

3. DATA GATHERING METHODS

3.1. Interviews with Faculty and Staff

3.2. Classroom Observations at Site A

3.3. Codesign and Prototyping with Occupational Therapists

3.4. Data Analysis Techniques

4. SITE A (CLASSROOM): SPECIAL EDUCATION SCHOOL

4.1. 3D Printing and Design Resources

Table II. Specifications for the Printers Used at our Sites

4.2. Primary Users

4.3. Use of 3D Printing

4.4. Objectives and Obstacles

5. SITE A (THERAPISTS): SPECIAL EDUCATION SCHOOL

5.1. Printing Resources

5.2. Primary Users

5.3. Use of 3D Printing

5.4. Objectives and Obstacles

6. SITE B: PRIVATE SCHOOL FOR BLIND AND VISION-IMPAIRED STUDENTS

6.1. Printing Resources

6.2. Primary Users

6.3. Use of 3D Printing

6.4. Objectives and Obstacles

7. SITE C: TECHNOLOGY CENTER

7.1. Printing Resources

7.2. Primary User

7.3. Use of 3D Printing

7.4. Objectives and Obstacles

8. 3D PRINTING STAKEHOLDERS IN SPECIAL EDUCATION

8.1. Administrators’ Perceptions and Expectations

individualized solutions” (Admin., Site A). In addition to creating low-cost AT solutions

8.3. Technology Instructors’ Perceptions and Expectations

8.4. Nontechnology Instructors’ Perceptions and Expectations

8.5. Technology Specialists’ Perceptions and Expectations

9. DESIGNING ASSISTIVE TECHNOLOGY 3D MODELING SOFTWARE FOR THERAPISTS

9.1. Need for Customized 3D Modeling Software

Fig. 5. A 3D printed assistive hand grip for a spoon.

Fig. 6. Screenshot of GripFab showing variables related to accessory width.

9.2.3. GripFab Version 2 - Focus Group 1 Feedback. We implemented a few changes to

9.3. Utility of 3D-Printed Assistive Grips

10.3. Recommendations for Special Education and Accessibility Organizations

10.4. Recommendations for Therapists

10.5. Lessons Learned (for Researchers)

problems with the geometry of a 3D model cannot be correctly translated to slices

11. CONCLUSIONS AND FUTURE WORK

Received March 2015; revised August 2015; accepted December 2015

You might also like