Technology Developments in Touch-Based Accessible Graphics: A Systematic Review of Research 2010-2020
Matthew Butler, Leona Holloway, Samuel Reinders, Cagatay Goncu, Kim Marriott
TTechnology Developments in Touch-Based Accessible Graphics:A Systematic Review of Research 2010-2020
Matthew Butler
Monash UniversityMelbourne, [email protected]
Leona Holloway
Monash UniversityMelbourne, [email protected]
Samuel Reinders
Monash UniversityMelbourne, [email protected]
Cagatay Goncu
Monash UniversityMelbourne, [email protected]
Kim Marriott
Monash UniversityMelbourne, [email protected]
ABSTRACT
This paper presents a systematic literature review of 292 publica-tions from 97 unique venues on touch-based graphics for peoplewho are blind or have low vision, from 2010 to mid-2020. It is thefirst review of its kind on touch-based accessible graphics. It istimely because it allows us to assess the impact of new technolo-gies such as commodity 3D printing and low-cost electronics onthe production and presentation of accessible graphics. As expectedour review shows an increase in publications from 2014 that we canattribute to these developments. It also reveals the need to: broadenapplication areas, especially to the workplace; broaden end-userparticipation throughout the full design process; and conduct morein situ evaluation. This work is linked to an online living resourceto be shared with the wider community.
CCS CONCEPTS • Human-centered computing → Accessibility . KEYWORDS
Systematic Literature Review, Assistive Technology, Tactile Graph-ics, Blind, Low Vision
ACM Reference Format:
Matthew Butler, Leona Holloway, Samuel Reinders, Cagatay Goncu, and KimMarriott. 2021. Technology Developments in Touch-Based Accessible Graph-ics: A Systematic Review of Research 2010-2020. In
CHI Conference on HumanFactors in Computing Systems (CHI ’21), May 8–13, 2021, Yokohama, Japan.
ACM, New York, NY, USA, 15 pages. https://doi.org/10.1145/3411764.3445207
Since the first school for the blind was founded in Paris by ValentineHaüy in 1784, raised line drawings, called tactile graphics, and 3Dmodels have been used to convey graphical content to people who
Permission to make digital or hard copies of all or part of this work for personal orclassroom use is granted without fee provided that copies are not made or distributedfor profit or commercial advantage and that copies bear this notice and the full citationon the first page. Copyrights for components of this work owned by others than ACMmust be honored. Abstracting with credit is permitted. To copy otherwise, or republish,to post on servers or to redistribute to lists, requires prior specific permission and/or afee. Request permissions from [email protected].
CHI ’21, May 8–13, 2021, Yokohama, Japan © 2021 Association for Computing Machinery.ACM ISBN 978-1-4503-8096-6/21/05...$15.00https://doi.org/10.1145/3411764.3445207 are blind or have low vision (BLV) [13]. Tactile graphics, along withimage descriptions, are now the most common method for prov-ing accessible graphics to BLV people. They find particular use inschools to provide BLV students with access to educational graphicssuch as STEM diagrams, charts and maps, and are frequently usedin Orientation & Mobility (O&M) training. Tactile graphics, how-ever, suffer several drawbacks. They require specialist equipmentto produce and they are usually designed by an experienced tran-scriber, meaning that production is slow and costly. Thus tactilesare not well suited to, for instance, on-the-fly creation of graphics inthe classroom, and because of their static, two-dimensional naturethey are restricted in the types of graphical information they candirectly convey. The use of tactile graphics also requires explicittraining and braille literacy, limiting the number of BLV peopleable to effectively use them.Consequently, a major focus of assistive technology developmenthas been to develop alternatives to tactile graphics that overcomesome of their limitations. For instance, interactive audio labels doaway with the need to understand braille and refreshable tactiledisplays remove the need for printing the tactile graphic. And overthe past decade researchers have leveraged emerging technologiesincluding ‘maker’ technologies such as 3D printing. Another focusof research has been to automate or at least semi-automate the pro-duction of tactile graphics or other accessible formats. But despitethis activity there has been no broad critical review of the researchinto these new ‘touch-based’ technologies for presenting accessiblegraphics to BLV people.This paper presents a systematic review and critique of researchinvolving new technologies for touch-based graphics for BLV peo-ple over the period of 2010 to mid-2020. Systematic reviews are vitalas they allow a research field to reflect and improve on best practice[6]. Our review focuses on research investigating new productionor presentation methods related to touch-based graphics. By thiswe mean presentation methods that, like tactile graphics, rely onthe reader using the position of fingers or hands to understand thespatial layout of the graphic. While other senses such as hearingcan also be used, spatial understanding needs to be driven primarilyby touch.Primary research venues for assistive and inclusive technologieswere reviewed, as well as keyword searches for literature outsideof these venues. In total, 352 papers were identified based on titlesand abstracts. After an initial review, 59 were excluded because a r X i v : . [ c s . H C ] F e b HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al. they did not meet the selection criteria, leaving 292 papers in thereview corpus. These were analysed on the following criteria: pri-mary research contributions; graphic content; application domain;presentation technologies; methodology, including stakeholder in-volvement; and availability of outcomes.
Contributions:
This systematic review contributes to the field ofaccessible graphics in the following ways: • It is the first systematic review of research relating to touch-based accessible graphics. It critically reviews the work overthe past decade, highlighting how the field has evolved. • It summarises the results of studies comparing differenttouch-based presentation techniques. • It identifies key omissions and concerns, revealing a numberof recommendations for the assistive technologies researchcommunity. These include the need to: broaden applicationareas, especially to the workplace; strengthen comparativestudies; conduct more authentic evaluation by increasing insitu and longitudinal studies; and broaden end-user partici-pation in the initial design process. • It provides an accessible living resource named AGRep (Ac-cessible Graphics Repository), capturing what has been donein the field of accessible graphics technology, to be sharedwith the wider community and be continually updated. Thisrepository contains all papers from this systematic review.
Accessibility guidelines recommend the use of tactile graphics forpresenting graphics, such as maps or charts, in which spatial rela-tionships are important [8, 37]. Tactile graphics, however, requirespecialist equipment to produce and, as specialist transcribers aretypically required, are expensive and time consuming to produce.Furthermore, the reader needs sufficient tactile reading skills, in-cluding the ability to read braille, to best understand the graphiccontent.The widespread use of touch screen technologies, the arrivalof low-cost maker technologies, and developments in computervision and machine learning that can facilitate automated tran-scription, have opened up opportunities to take the provision oftouch-based graphics into a new era. As such, new techniques forthe creation of accessible graphics is a significant area of research.Key areas of research include new approaches to 3D-printed tactilemodels [7, 18, 25, 46, 51], haptic feedback [11, 14, 38, 39, 50], roboticsystems [22], touch-triggered auditory feedback [10, 42, 43], andautomated tactile generation [9, 33, 48, 54], such as utilising onlinemapping systems [17, 49, 55].Given the rapid rate at which new research is being conductedin the use of these new technologies for touch-based graphics, it istimely to review the field in order to better understand the nature ofthis research, and to identify best practice to better inform researchin this context.
Systematic Literature Reviews (SLRs) are conducted across manydifferent research fields. They are undertaken by researchers inorder to examine a research field more holistically by understandingits recent areas of focus, the gaps, and looking to the future of thefield. When conducted in a rigorous manner they provide a vehicleby which a research field can reflect upon and improve its bestpractice.While most research is underpinned by a review of related work,it is often very targeted and typically self-serving to the researchbeing presented. SLRs seek to be broader in scope, synthesising“existing work in a manner that is fair and seen to be fair” [29], aswell as remove bias [36]. Not only can they provide a more completepicture of the type of work being undertaken, they can also examinea breadth of settings, method and contexts [3, 29]. Thus they allowus to understand the current state of the field in a more holisticmanner.Further to this, SLRs capture a point in time and reveal what hasled the research field to this point. This is important in technology-related fields, such as assistive technologies, where advances intechnology and techniques have a large influence on the directionthat a research field takes.Finally, SLRs allow us to identify the key challenges moving for-ward, regarding both topic and method. They not only identify thecurrent state of the art, but also the gaps that are yet to be addressedand provide data for rational decision making [36]. They can beinfluential in guiding the research agenda for the years to follow.As such, “The aim of an SLR is not just to aggregate all existingevidence on a research question; it is also intended to support thedevelopment of evidence-based guidelines for practitioners.” [28].For example, the work of Barkhuus and Rode [2] presents a sys-tematically derived set of accepted evaluation methods in the CHIcommunity and demonstrates how they are important to specificinfluential research communities.There is evidence of the benefit of systematic reviews withinthe domain of assistive technologies. Kelly and Smith [27], in theirmulti-decade review found that, while of benefit to the professionalcommunity, most work lacked methodological rigour. Wabiński andMoscicka [53] analysed works relating to automatic tactile map pro-duction and found that very few of the working solutions reportedever made it to widespread adoption. More recently, Brulé et. al.[6] sought to better understand the nature of empirical evaluationof assistive technologies for BLV users, and found that meaningfulparticipation by BLV end users in evaluation was lacking, as wellas a disproportionately low involvement of low vision participantsrelative to their proportion in the BLV community. Other reviewsfocus on a subset of work, such as the development of interactivemaps [12, 55], and highlight research questions specific to that con-text. These works demonstrate that systematic reviews can shine alight on key issues and provoke the assistive technologies researchcommunity to reflect and improve on their practice. echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan
This SLR has adopted a process consistent with that recommendedby Kitchenham [29]. The process is also consistent with that re-ported in [6], a systematic review in a similar domain relating toBLV technology.
This work focused on a number of questions regarding researchpublications in touch-based graphics over the past decade. These areconsistent with other SLRs, whereby it is the intention to capturea snapshot of current practice, identify gaps, and foreshadow thefuture of research in the area. Key questions are: • What are the main types of graphics and application domainsstudied in research exploring touch-based graphics? • What are the primary presentation technologies and howhave they changed over the past decade? • What methodological approaches are used and how inclusiveare they of the end users? • What are key gaps or challenges that researchers shouldconsider moving forward?This SLR aims to cover work relating to technological develop-ments in touch-based graphics for people who are blind or havelow vision (BLV). On this basis, the research team documenteda number of key inclusion/exclusion examples that informed theselection of works. Notable inclusion criteria required that: • The work must relate primarily to novel touch-based graph-ics representations. In this context novelty refers to focus ona new production or presentation approach. A broad inter-pretation of ‘new’ was adopted when considering work forinclusion; • Auditory work could be considered if a touch interactionstrategy was the basis; • Traditional touch-based graphics, such as tactile graphics,are only included if part of a comparative study with newtechnologies or if exploring new production techniques. Pa-pers solely describing the use of tactile graphics were notregarded as ‘novel’.Key exclusion examples were: • Blind or low vision users are not an identified focus of thework; • Primarily focusing on braille text; • Exclusive focus on user interfaces; • Audio interfaces with no touch interaction.A ten-year period was chosen for the review, rather than a longerperiod. This was for two primary reasons. Firstly, a large volumeof relevant papers was identified in this period (over 350). It wasalso felt the last decade covered the emergence of topical newtechnologies such as 3D printing.In order to identify relevant work, multiple search approacheswere used to reduce publication bias [29]. A systematic reviewof ‘gold standard’ venues was combined with a keyword search.Gold standard venues were those that fell into one of the followingcategories: • Research publication venues with known interest in assis-tive technology, i.e.: ACM SIGACCESS Conference on Com-puters and Accessibility (ASSETS); ACM CHI Conferenceon Human Factors in Computing Systems (CHI); Interna-tional Conference on Computers Helping People with Spe-cial Needs (ICCHP); PErvasive Technologies Related to As-sistive Environments (PETRA); ACM Transactions on Acces-sible Computing (TACCESS); International Conference onTangible, Embedded and Embodied Interaction (TEI); ACMTransactions on Computer-Human Interaction (TOCHI); andACM Symposium on User Interface Software and Technology(UIST). These include those recognised as top-tier venues by[6]. • Industry journals: Journal of Visual Impairment and Blind-ness (JVIB); Journal of Blindness Innovation and Research(JBIR); and British Journal of Visual Impairment (BJVI).All works in these venues were reviewed, using titles and ab-stracts as the primary guide. As per [29], a liberal and inclusiveapproach was taken, knowing that works could be removed at alater stage of the process. All works in the venues were considered,including full papers, short papers, late-breaking works, etc.In order to identify work published in other venues, a keywordsearch was also undertaken. This could capture work presentedoutside of venues specifically associated with assistive technolo-gies. A Google Scholar advanced search was conducted to identifyarticles between 2010 and 2020, using the following keywords: • At least one of: disability; accessibility; blind; low vision; • Plus at least one of: tactile; touch; haptic; • Plus at least one of: graphic; diagram; map.Rather than search specific databases e.g. ACM/IEEE, GoogleScholar was used as it includes these databases. We did not use exactphrases in the search, thus the search results included synonyms.We reviewed the first 100 search results, whereby no new relevantpapers were found on the next results page. After conducting thesearch using these two methods, 352 papers were included in theinitial corpus.
Based on the identified aims of the SLR, the research team createda set of analysis criteria by which each paper would be reviewed.Each criteria was discussed by the entire team to ensure agreementof its meaning was reached. Criteria were presented in a GoogleForm, with one form submission made for each paper.A small cross-section of papers were reviewed by the entireteam to assess the appropriateness of the analysis criteria. Theresearch team then met to discuss the outcomes of the prelimi-nary review, during which a revised and expanded set of analysiscriteria was developed. Using a revised Google Form to capturedata, all papers were reviewed against the new criteria. Resultswere captured for analysis in an accompanying Google Sheet. Thefollowing criteria were analysed: contributions; graphics and theirapplication domain; presentation technology; production methods;research methodology; evaluation methods and participation; anddistribution of technology outcomes.All 352 papers were fully reviewed. During this process eachpaper was also re-assessed for relevance [29] and could be excluded
HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al. from the corpus if it was deemed to be outside the scope of the liter-ature review, as based on the initial inclusion criteria. If a reviewerbelieved a paper should be excluded, another reviewer would alsoanalyse the paper. If both believed the paper should be excludedthen it was formally removed from the data set. If there was dis-agreement, a third reviewer would make the final decision. Thisensured reliability of the inclusions and exclusions [29]. Examplesof papers that were excluded at this stage include those where onlywidely adopted production and presentation methods with tactilediagrams were used, work where the context of application was notidentified as being for BLV end users, and papers where the workonly related to accessibility of user interfaces and not graphicalinformation.
To support the authors as well as the wider touch-based graph-ics research community, the corpus of identified documents hasbeen catalogued in a publicly accessible web format, based on theinteractive sentiment visualisation tool SentimentViz [31]. LikeSentimentViz, our tool - AGRep (Accessible Graphics Repository)- is designed to help users to “discover interesting patterns andfacilitate data exploration” [31].Using AGRep (Figure 1) it is possible to explore and perform cus-tom searches on the corpus of resources based on our categorisationmarkup, access persistent identifiers or URLs for each resource, andview tabulated summary data. Additionally, as AGRep is envisionedas a living resource capturing what has been done in the field ofaccessible graphics technology, users can update the corpus withnewly published resources in the future.
AGRep has been customised toimprove its level of accessibility. During early stages of developmentthe researchers held an informal session with a BLV participantto understand how the accessibility of the tool could be improved.This involved a walk-through of a test deployment of SentimentVizon the participant’s personal technology.Based on this informal session, the following modifications weremade to the source code of SentimentViz to increase its accessibilityfor the purpose of AGRep: • A help dialogue page added, outlining how to best use andnavigate the tool; • Alternate text added to all images, icons and glyph symbols; • The tabbing order of page elements optimised to improvescreen reader performance; • A change from grid-based layout of resources to row-basedlayout; • Elements that update per user input marked up so that screenreaders can interpret dynamically changing content.The AGRep tool contains all papers from the corpus of this SLR,as well as more recent 2020 works that meet the inclusion criteria.The process for external parties to add new works for inclusionis given on the repository website. The AGRep tool is located at https://agrepository.github.io
On completion of the full review, the final corpus contained 292 pa-pers from 97 unique venues. Following are the basic characteristicsof the corpus. Note that the largest body of work came from ICCHP.As this conference runs every second year, data is aggregated intwo-year blocks, so as not to misrepresent alternate years whenICCHP was not held.Table 1 shows the distribution of papers over the decade. Thenumber of papers can be seen to have significantly risen in 2014. Apossible reason for this will be discussed when we examine papercontent later in this review.The corpus was drawn from 97 unique venues. As can be seenin Table 2, the vast majority of venues from which papers werechosen (70.41%) were represented by only one paper in the corpus.Only 4 venues had on average one or more paper per year. Thesewere: ICCHP (61 papers); ASSETS (44 papers); CHI (35 papers); andTACCESS (10 papers).The majority of papers collected were classified as ‘Full Papers’by their venue (Table 3). There were also a number that were consid-ered as ‘Short Papers’ or were an accompanying paper to a posteror demonstration. Short papers typically do not have the same ex-pectation of rigour regarding evaluation, and include such thingsas preliminary studies or late-breaking work. The majority of short,poster and demo papers were in the top three venues (ICCHP, AS-SETS and CHI) with a little under half being considered ‘Full Papers’in those specific venues.Across the 292 papers, there were 641 unique authors. All authorswere counted in the metrics regardless of the author order andweightings were not applied to authorship. The large number ofauthors is consistent with the finding that there are a very highnumber of venues with only one paper (Table 2). Table 4 shows that488 authors (76.13%) contributed to only one paper in the corpus,i.e. have only one work relating to touch-based graphics in thelast 10 years. This suggests that a significant amount of researchis a ‘one-off’ by authors: they explore a single idea regarding aninnovation with touch-based graphics and do not further developor evaluate it. In accord with this, very few authors have multiplepublications. Only 32 authors (5%) have 5 or more papers publishedrelating to touch-based graphics over the past decade.The long tail in both venues and authors shows little evidenceof ‘deep dives’ into a particular research direction by researchers inthe field. While this is somewhat speculative, it raises the concernthat most authors do not have a deep knowledge of what othershave done in the field and are possibly ‘reinventing the wheel’.It also raises the question of whether one-off research will leadto meaningful change in BLV community practice. On the otherhand, it may be that to effect change in this area, many differentapproaches need to be tried to ‘see what sticks’. Indeed, this wasillustrated clearly in the Transforming Braille Project, initiated in2011 to develop a low-cost refreshable braille display and bring itto market, whereby over 60 projects were examined before leadingto a commercialised product [45]. This does not, however, stop usconsidering the question of how work in this domain can be bettershared and built upon to effect meaningful change more quickly. echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan
Figure 1: AGRep website
Year 2010-11 2012-13 2014-15 2016-17 2018-19 2020Number of Papers 46 33 73 59 73 9
Table 1: Number of Papers by Year
Paper Count 1 2 3 4 5 6 7 10 35 44 61Number of Venues 69 16 2 1 1 2 2 1 1 1 1Percentage 70.41% 16.33% 2.04% 1.02% 1.02% 2.04% 2.04% 1.02% 1.02% 1.02% 1.02%
Table 2: Number of Papers by Venue
Paper Type Full Short Poster/Demo Thesis Experience Chapter Work In Progress WorkshopCount 171 59 37 7 6 4 4 4
Table 3: Number of Papers by Type
Paper Count 16 13 10 9 8 7 6 5 4 3 2 1Num Authors 1 1 1 2 2 5 6 14 17 25 79 488Percentage 0.16% 0.16% 0.16% 0.31% 0.31% 0.78% 0.94% 2.18% 2.65% 3.90% 12.32% 76.13%
Table 4: Number of Papers by Author
HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al.
Graphics and their application domain provide an insight into theprimary use cases of the research undertaken within the corpus.Graphic Type Paper CountSTEM 115Math 35Science 34Graphs 28Technology 11Geometry 5Biology 2Maps and Plans 110General or Unspecified 104Art and Culture 53Paintings 22Sculpture and Artefacts 15Book Images 8Drawings 3Photographs 3Architecture 2Other 8
Table 5: Diagrams by Type
Three categories dominated the types of graphics represented inthe corpus: STEM related (115); Maps and Plans (110); and Generalor Unspecified (104). STEM related graphics encapsulated a numberof other sub-types, including those relating to math, science andtechnology. “General or Unspecified” was a catch-all for graphicsthat were not explicitly defined or could be applied across domains.These included those represented by work exploring general pre-sentation systems. Graphics captured by “Other” include: Puzzlesand Games (4); Screen Layout (2); and Patterns (1) and Handwriting(1).The focus on STEM related graphics and maps and plans is tobe expected as it reflects that the main application areas for tactilegraphics are in education and Orientation and Mobility (O&M)training. Maps and plans also lend themselves to technology driveninnovations. For example, they are a strong use case for 3D printing,audio labelling, and in particular automated creation, where opendata is easily available.“Arts and Culture” is the other main category of graphic foundin the corpus. This is somewhat surprising as they are not a typicaluse case for the provision of tactile graphics. It is exciting to see thatresearchers are exploring innovation outside of the more ‘obvious’application areas of tactile graphics.We also examined if the type of graphic had varied over the pastdecade. We found there was no significant change in focus. Theonly exception being a small increase in graphics relating to Artsand Culture in the latter years.The intended application domain of the research publications(Figure 6) is consistent with the types of graphics they consider.Education and O&M dominate, which is reflective of the the primaryapplication areas for tactile diagrams. The category of “Other” onlycaptured one other use case, being for research. Intended Application Paper CountEducation 106O&M 83Daily living 46Not specified 38Museum or gallery 32N/A 7Workplace 5Other 1
Table 6: Diagrams by Use
Analysis revealeda strong focus on Education and O&M. This reflects a focus onmore functional or utilitarian use, as clearly graphics in these fieldsprovide critical information that BLV people need in their day today lives.Graphics that are not related to daily living, but rather quality oflife across the whole life span, are much less of a focus. An exceptionare those relating to arts and culture. We believe an important areaof future research is to consider other graphics, such as for instance,sports graphics, which are common in popular culture but currentlyunavailable to the BLV community.An even more glaring omission is that graphics for the work-place were only considered in 5 of the papers in the corpus. This isvery low and suggests significant inequities that may exist in theprovision of accessible graphics in the workplace. This should berecognised as not just an opportunity for future research but onethat must be addressed to support BLV people in the workplace,not just in schools or mobility training.
Contribution Paper CountCase study or application area 120Presentation method 88Production method 74Interaction method 69Comparative study (of presentation methods) 34Other 18Systematic Review 7
Table 7: Paper Contributions
Next we considered the area in which a paper made its primarycontribution (Table 7). A paper could make more than one contri-bution.The majority of works collected were those with a new applica-tion of existing technologies or techniques. That is not unexpected,and reflects that many works came from publication venues outsidethose related specifically to assistive technologies. Those works of-ten explored the provision of touch-based graphics in a new context,for example the arts, cartography, etc. echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan
Behind this, three contributions shared the majority of focusof other papers. These were new presentation methods, new pro-duction methods, and new interaction methods. It is significantand positive to see that researchers are not simply focusing onnew presentation methods for touch-based graphics but are alsodevoting equal attention to their creation and the ways in whichthe reader can interact with them.Within new production techniques, the dominant topic is theautomation of accessible graphic creation, especially within thecontext of maps and plans. This is very important, as the timelyand cost effective production of accessible graphics is a significantbarrier to their more widespread use, especially in contexts outsideof education and O&M. As such it is heartening to see the stronginterest by researchers but it remains an area that would benefitfrom increased attention.Other contributions that were captured by papers in the corpusincluded: reviews of technology (6); development or evaluationof guidelines or standards (5); exploration of design spaces andapproaches (5); feasibility or usability studies (3); and understandingof tactile perception (2).We also evaluated whether a paper was an improvement of anexisting production technique or presentation technology, i.e. if itwas heavily based on a preceding piece of research. While this is amore subjective measure than our other measures we believe it isstill suggestive. Within the corpus, 38 studies were considered tobe focused on improving existing work. This relatively low numberis consistent with the ‘long tails’ of both authors and venues aspresented in section 4.1, and again should be an area of reflectionfor researchers.
We discovered thatdifferent venues appear to implicitly have a slightly different focusregarding contribution. Table 8 shows the contributions of the pa-pers in each of the 4 primary publications venues. ICCHP was moreheavily skewed toward new production methods and case studyareas, while ASSETS and CHI explored presentation and interactionmethods more strongly. This may be consistent with the nature ofICCHP having (in general) greater involvement by practitioners.Notably, ASSETS and ICCHP had a larger number of papers thatwere improvement of existing techniques and technologies thanthe other venues.
Given our focus on new presentationtechnologies, we were interested to summarise the findings fromuser studies comparing different presentation technologies. Weextracted these papers from the corpus and a summary of the studiesand results can be seen in Table 14 in the Appendix. Most studiesfocused on maps but a few used simple charts or STEM materialsas stimuli. The tasks were quite varied. Almost all were laboratorystudies, with only 3 in situ evaluations [24, 30, 40].Overall, comparisons with more than a single study support thefollowing conclusions: •
3D printed model with/without audio labels vs tactile graphic [4,15, 19–21, 25, 30, 41] : 3D model is preferred, better short-termrecall using 3D model (but not long-term recall). Some indi-cation that the representation of third dimension is easier to understand on a 3D model. Symbols are easier to discrim-inate and iconic symbols more understandable with a 3Dmodel. • Touchscreen with overlay and audio labels vs tactile graphics [5,35] : Touchscreen with overlay and audio labels preferred andfaster. • Touchscreen with audio labels with/without vibration vs tactilegraphic [16, 23, 52, 55] : Tactile graphic is preferred and isexplored more quickly. Some indication that it is difficult tounderstand detailed geometry with touchscreen. • Force feedback with audio labels with/without vibration vstactile graphic [44, 56] : Faster with tactile graphic. Conflictingresults on accuracy.However, this examination revealed that more controlled studiesare required to better understand the comparative advantages ofthese different presentation methods. These studies should considera wider range of graphics and more systematically investigate howlow-level tasks are supported as well as the use of these methodsin situ and longitudinally.
Table 9 identifies the presentation technologies used in the corpus.These technologies can be combined. The ‘base’ presentation tech-nologies which provide information about the spatial layout of thegraphic are, in descending order of occurrence: tactile graphics; 3Dmodels; touch screens; refreshable tactile displays; and force feed-back devices. These base technologies were frequently combinedwith the other presentation technologies. In the case of 3D models,audio labels were common while touch screens also provided audiolabels and sometimes tactile overlays and/or vibratory feedback.It may be surprising that tactile graphics, a well established tech-nology, features so prominently. This reflects that they are currentlythe ‘gold standard’ for accessible graphics provision and so are afrequent benchmark for comparing new presentation technologiesor are the focus of new production methods.
Figure 2 shows the changes in pre-sentation technologies over time. The most striking change is thesharp increase in the use of 3D printed models from 2014-2015. Thisis not unexpected, given that commodity printers arrived at thistime, and over the past 5 years they have both dropped in priceand been more widely adopted by researchers and practitionersin assistive graphics provision. There has also been an associatedincrease in the use of audio labels, as the two technologies arefrequently combined.We believe that the primary reason for the overall increase inpublications about touch-based graphics noted earlier is the ar-rival of low cost ‘maker’ technologies such as 3D printing in 2014.This led to lower research and development costs for prototypingnew presentation technologies as well as interest in how thesetechnologies could be directly used to create accessible graphics.
We also see that there has beendiminishing attention to force feedback devices and sonification.The early part of the decade had a focus on the use of force feedbackdevices, such as the ‘Phantom Omni,’ but their use dropped off
HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al.
Case study /application Presentationmethod Productionmethod Interactionmethod Comparativestudy SystematicReview ImprovementASSETS 15 17 12 17 3 0 10CHI 20 13 7 12 2 1 4ICCHP 23 15 23 10 3 0 9TACCESS 8 3 0 3 1 0 2
Table 8: Paper Contributions by Key Venue
Total ASSETS CHI ICCHP TACCESSTactile Graphic 105 14 11 25 53D Model - 3D printed 102 19 11 12 3Audio Labels 98 13 13 18 4Touch Screen 56 12 7 7 2Other 42 6 11 4 3Refreshable Tactile Display 37 5 5 10 1Vibration 32 6 1 5 2Tactile Overlay 28 6 6 4 13D Model - Other 24 3 1 5 2Force Feedback Device 21 2 1 6 3Sonification 18 4 3 2 3Low Vision Visuals 12 5 4 0 1
Table 9: Technologies Overall and by Key VenueFigure 2: Technology Research Over Time echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan almost completely by decade’s end. We conjecture this is becauseof their expense and limited uptake by the BLV community.The reduction in sonification research is curious. However, itmay be that our requirement that sonification is employed in touch-based interfaces has resulted in a downward trend that is not repre-sentative of the broader research agenda. Even though this may bethe case, with the availability of new ways to trigger audio labelling(such as with a moble phone, or with capacative touch sensors on3D models), this seems a presentation technology worth pursuing.
One of the most uncommonpresentation technologies is low-vision visuals, though it has in-creased over time. This suggests research in this area may be skewedtoward people who are blind as opposed to having low-vision de-spite the fact that there are many more people with low vision thanblindness. This was consistent with the findings of [6]. It may alsoreflect that the majority of people with low-vision do not wish touse touch-based graphics, however this would need to be furtherinvestigated.
The technologiesrepresented in the corpus can also be considered in light of theresearch contributions of each work. Figure 3 shows the four mainresearch contribution types, with the associated number of papersfor each technology. Note that there can be more than one tech-nology associated to the contribution (e.g. a case study may utiliseboth tactile graphics and 3D printed models). Similarly, a paper canbe categorised with more than one contribution.Examining contribution against technology reveals a number ofinteresting differences. When looking at case studies, 3D printingis the most common. This seems remarkable given that commod-ity 3D printing only emerged midway through the review period.However, it is likely that the rapid high levels of interest in 3Dprinting by both researchers and practitioners in accessible graph-ics and early positive evaluations fuelled interest in a wide varietyof applications.New production methods were dominated by tactile graphics.Within the corpus, new production methods focused heavily onautomatically converting data or existing graphics into a tradi-tional tactile graphic. As such, this is unsurprising. However, thishighlights a significant research opportunity: How can similar au-tomation processes be used in the production of 3D printed models?This is especially important given the amount of work exploringthe use of 3D printing for presenting touch-based graphics, and thatthe design of 3D models is often a barrier to their wider adoption.We were surprised that new presentation methods were domi-nated by audio labels. However, this is likely due to the fact theycan be used in conjunction with both touchscreens and 3D printedmodels. It is encouraging to see a commodity technology such astouch screens being used in the exploration of new presentationmethods, as well as the more recently emerging refreshable tactiledisplays. 3D printed models are somewhat underrepresented in thiscontribution category. This is likely due to their more recent emer-gence; they are first being evaluated for their efficacy in differentapplication areas before being fully explored as the basis for newpresentation methods.
We were particularly interested to understand the level of involve-ment of end users in the research. While BLV people are the primaryend users, end users also include sector professionals such as tran-scribers or O&M trainers.Table 10 shows the involvement of participants in each researchstudy, across the design, prototype and evaluation stages of eachstudy. The participants are divided into three categories: BLV, Pro-fessional and Sighted. Note that participants might take part acrossmultiple stages within a single study. Similarly, members from mul-tiple participant groups may be used within a single stage within astudy (e.g. both blind and low vision participants within an evalua-tion).Importantly, BLV participants are by far the most representedgroup, with the strongest representation by blind participants inthe evaluation stage. One thing that is noticeable is that BLV partic-ipants, however, are much less likely to be involved in the designor prototyping stages than the final evaluation. This is potentiallya serious problem as it means end user feedback is only consid-ered late in the development. The following section explores BLVrepresentation further.
Participation by BLV peo-ple across the corpus was in 62.80% of the studies, which may beconsidered low for work that is intended to be for the benefit of aparticular community. This increases marginally if sector profes-sionals are included, to 66.89% (Figure 11).However, this must be considered in light of the paper type(Figure 12). Short papers, posters and demonstrations had lowerrepresentation. This is primarily a constraint of either the spaceafforded the work, or that it is often the initial stages of new work.Full papers present a better insight into end user involvement, andas such are the focus of discussion. When only considering fullpapers, the involvement of BLV people in the research increasesto 73.26%. While this is strong, there is still considerable scope forimprovement.It could be expected that the primary venues associated withassistive technologies would have a higher involvement by theBLV community. When considering full papers only, this is mostlythe case, with ASSETS, CHI and TACCESS all showing levels ofengagement by BLV pople in over 85% of papers. Indeed, whensector professionals community are considered as well, all threevenues have 100% involvement. ICCHP, however, is an exception.BLV involvement is still only in the order of 69% and there is noinvolvement with professionals within the corpus. Involvementby end users in the other venues, including venues not normallyassociated with assistive technologies or accessibility, is lower thanthe top 3 venues. Of note is that only 1 publication out of 143 papersin other venues had the involvement of a sector professional.One concern that emerges is that studies are skewed toward blindparticipants as opposed to participants with low vision. This mayreflect the lack of focus on low-vision visuals that we previouslyidentified. It is, however, consistent with the findings of [6, 47] andmay reflect that researchers believe touch-based graphics are ofgreater importance to those who are blind. It may also be that newkinds of touch-based graphics such as touchscreens with audiolabels or 3D printed models will be of benefit to people with low
HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al.
Figure 3: Technologies by Research Contribution
Participants Stage Number of Studies % of Total Papers Total ParticipantsBLV Blind Design 36 12.29% 344Prototype 48 16.38% 234Eval 150 51.19% 1516Low Vision Design 13 4.44% 85Prototype 15 5.12% 73Eval 60 20.48% 296Professionals Professionals Design 20 6.83% 416Prototype 10 3.41% 107Eval 16 5.46% 133Sighted Sighted Design 5 1.71% 135Prototype 11 3.75% 98Eval 17 5.80% 288Blind Folded Design N/A N/APrototype 8 2.73% 61Eval 26 8.87% 432
Table 10: Participants
Participant Type Number of Papers % of Total Papers Total ParticipantsBLV 184 62.80% 3204BLV + Professional 196 66.89% 3227
Table 11: Participation by End Users (BLV and Sector Professionals) vision, however a focus on blind participants more stringentlyevaluates the technology.
The numbers of participants in eachstudy is also variable. In studies presented in full papers, there wasa median of 10 BLV participants. While this may appear to be a relatively low number of participants, it is understandable giventhe low incidence of blindness and the difficulty that can arise withrecruiting participants. With small sample sizes it is difficult to findstatistical significance, so this low number may account for thegreater number of qualitative studies (161) that were conducted in echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan
NumPapers BLVInvolvement Percentage Num FullPapers Full PapersBLV Percentage Full BLV+ Prof. PercentageASSETS 44 27 61.36% 16 14 87.50% 16 100.00%CHI 35 28 80.00% 18 17 94.44% 18 100.00%ICCHP 61 40 65.57% 39 27 69.23% 27 69.23%TACCESS 10 9 90.00% 10 9 90.00% 10 100.00%Other 143 80 55.94% 89 59 73.75% 60 75.00%
Table 12: Participation by Venue the corpus in comparison to quantitative studies (111). However,this is not necessarily a problem, as qualitative studies can providerich data and important insights. Indeed [6] warns against rejectingpapers based on participant numbers in this field.
One noticeable limi-tation of many of the studies undertaken was that they were under-taken in controlled environments, such as a laboratory (148 studies)rather than in the field or in situ (47 studies). While research thatinvolves prototype bespoke technologies may be constrained intheir ability to be conducted outside of a controlled environment, itdoes suggest that little of the work being evaluated is at the stagewhere it can used or distributed in its end user environment.Moreover, in almost all studies the user learned the system andwas tested on its use in the same session. While short studies canprovide immediate feedback for further development, they do notaccount for the time it can take to develop new tactile literacies,become comfortable with a new presentation method, or transcendlearning curves that a new technology may have. As such, morelongitudinal studies should be considered to truly evaluate newdevelopments in touch-based graphics and to obtain greater buy inand adoption.
The evaluation under-taken in most studies was also strongly focused on function, espe-cially when exploring new presentation methods. Rarely did thestudies encapsulate other important factors such as engagement,pleasure or stimulation. Given that an important part of the develop-ment of visual graphics is in their aesthetic pleasure, it is suggestedthat research in this area starts to consider more holistic evaluationcriteria.
As previously stated, commercial work was not considered as partof the literature review. However, in order to gauge the potentialongoing impact of the research in the corpus, the availability of itsoutcomes was also evaluated during analysis. This was based uponstatements in the paper and so probably understates the actualavailability of the research.Table 13 shows the availability of the outcomes of each researchpaper in the corpus. Of the work where the nature of availabilitywas clear, almost half the outcomes of the research were consid-ered to be unavailable to the broader research community or enduser, due to the research being undertaken with a custom proto-type. Close to 40% of the outcomes are available in some form,whether it be through the use of commodity technologies or with Availability Paper CountNot Available (Custom Prototype) 136Available 107Commodity technology 81Open source 15Commercialised by the researchers 11Not applicable / Not reported 45
Table 13: Outcome Availability the outcomes being explicitly made available through open sourceor commercialisation.It is encouraging that 81 papers did use commodity technologies.However, it is also acknowledged that there is significant prototyp-ing being undertaken utilising low cost technologies (such as 3Dprinting, simple electronics, or simple vision systems). This maysignal a shift toward research that will not require significant costto become more widely adopted. As such, continued work shouldstrive to balance innovation with widely adopted commodity tech-nologies as well as with low cost development technologies in orderto lead to impact in the BLV community.
This paper presents a systematic review of literature relating totouch-based graphics over the period of 2010 to mid-2020. Analysisof the work has provided insights into the changing nature of thisresearch, as well as highlighting ‘calls to action’ for researchers inthe field. The following are key areas for researchers to consider aswell as opportunities for future work: • Broadening Application Areas:
The areas of application fortouch-based graphics should be broadened to include graph-ics outside the education and O&M context. While thosecontexts are fundamental to daily living for the BLV commu-nity, it is important for researchers to consider other domainareas. An application area in considerable need of attentionis graphics for the workplace. This is not just an opportunityfor future research but is also vital for increased support forBLV people in the workplace. • Strengthening Comparative Studies:
More controlled studiesare required in order to better understand the comparativeadvantages of new presentation methods. Comparative stud-ies can then provide important evidence that can informdecision making regarding adoption of new technologies
HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al. by the wider BLV community and its stakeholders. Thesestudies should also consider a wider range of graphics. • Authentic Evaluation:
Laboratory studies dominate the re-search surveyed. While this is an important first step inresearch innovation, it is also critical that new developmentsin touch-based graphics are evaluated in the context in whichthey will be primarily used and evaluated over longer peri-ods of time. This should be a priority for researchers, as itwill also support research becoming more widely adopted inthe BLV community. • BLV Involvement:
While BLV involvement in comparativestudies and evaluations is relatively high, there is much lessinvolvement in initial design or prototype evaluation. Wesuggest that the design of new technologies for touch-basedgraphics, like other areas of assistive technology design,would greatly benefit from using a more participatory designprocesses, in particular during the creation of accessiblegraphics. It is also important that the low vision communityis equitably represented in research studies.It is hoped that this systematic review of work can support theassistive technologies research community by highlighting gaps incurrent practice as well as opportunities for work in the future. Itforms the basis for a web-based living resource that we hope willbe a valuable tool for accessibility researchers for many years tocome, and lead to more meaningful impact of this research for theBLV community.
Cagatay Goncu is supported by the Australian Research Council(ARC) grant DE180100057.
REFERENCES [1] Sandra Bardot, Marcos Serrano, and Christophe Jouffrais. 2016. From tactileto virtual: using a smartwatch to improve spatial map exlporation for visuallyimpaired users. In
Mobile HCI ’16 . ACM, New York, NY, USA, 100–111. https://doi.org/10.1145/2935334.2935342[2] Louise Barkhuus and Jennifer A. Rode. 2007. From mice to men-24 years ofevaluation in CHI. In
Proceedings of the SIGCHI Conference on Human Factors inComputing Systems , Vol. 10.[3] Pearl Brereton, Barbara A. Kitchenham, David Budgen, Mark Turner, and Mo-hamed Khalil. 2007. Lessons from applying the systematic literature reviewprocess within the software engineering domain.
Journal of Systems and Software
80, 4 (2007), 571–583.[4] Megen E. Brittell, Amy K Lobben, and Megan Lawrence. 2018. Usability Evalua-tion of Tactile Map Symbols Across Three Production Technologies.
Journal ofVisual Impairment & Blindness
Human–Computer Interac-tion
30, 2 (2015), 156–194. https://doi.org/10.1080/07370024.2014.924412arXiv:https://doi.org/10.1080/07370024.2014.924412[6] Emeline Brulé, Brianna J. Tomlinson, Oussama Metatla, Christophe Jouffrais,and Marcos Serrano. 2020. Review of Quantitative Empirical Evaluations ofTechnology for People with Visual Impairments. In
Proceedings of the 2020 CHIConference on Human Factors in Computing Systems . 1–14.[7] Monica Carfagni, Rocco Furferi, Lapo Governi, and Yary Volpe. 2012.
TactileRepresentation of Paintings: An Early Assessment of Possible Computer BasedStrategies . Springer-Verlag, Berlin, 261–270. https://doi.org/10.1007/978-3-642-34234-9_26[8] DIAGRAM Center. [n.d.]. Decision Tree Image Sorting Tool. http://diagramcenter.org/decision-tree.html[9] Yong Chen, Zehira Haddad, and Jaime Lopez Krahe. 2014. Contribution to theAutomation of the Tactile Images Transcription Process. In
ICCHP: InternationalConference on Computers Helping People with Special Needs , Vol. I. Springer, 642–649. [10] James M. Coughlan, Huiying Shen, and Brandon Biggs. 2020. Towards AccessibleAudio Labeling of 3D Objects.[11] M. A. Darrah. 2013. Computer Haptics: A New Way of Increasing Access andUnderstanding of Math and Science for Students Who are Blind and VisuallyImpaired.
Journal of Blindness Innovation and Research
3, 2 (2013). https://doi.org/10.5241/3-47[12] Julie Ducasse, Anke M. Brock, and Christophe Jouffrais. 2018. Accessible inter-active maps for visually impaired users. In
Mobility of visually impaired people .Springer, 537–584.[13] Yvonne Eriksson. 1998. Tactile pictures. Pictoral representations for the blind,1784-1940. (1998).[14] Simon Gay, Marc-Aurèle Rivière, and Edwige Pissaloux. 2018. Towards Haptic Sur-face Devices with Force Feedback for Visually Impaired People. , 258-266 pages.[15] Stéphanie Giraud, Anke M. Brock, Marc Macé, and Christophe Jouffrais. 2017.Map Learning with a 3D Printed Interactive Small-Scale Model: Improvementof Space and Text Memorization in Visually Impaired Students.
Frontiers inPsychology
8, 930 (2017), 1–10. https://doi.org/10.3389/fpsyg.2017.00930[16] Nicholas A. Giudice, Hari Prasath Palani, Eric Brenner, and Kevin M. Kramer.2012. Learning non-visual graphic information using a touch-based vibro-audiointerface. In
Proceedings of the 14th international ACM SIGACCESS conferenceon Computers and accessibility . ACM, 103–110. https://doi.org/10.1145/2384916.2384935[17] Timo Götzelmann and Aleksander Pavkovic. 2014. Towards automatically gener-ated tactile detail maps by 3D printers for blind persons. In
International Confer-ence on Computers for Handicapped Persons . Springer, 1–7.[18] Noreen Grice, Carol Christian, Antonella Nota, and Perry Greenfield. 2015. 3DPrinting Technology: A Unique Way of Making Hubble Space Telescope ImagesAccessible to Non-Visual Learners.
Journal of Blindness Innovation and Research
5, 1 (2015).[19] Jaume Gual, Marina Puyuelo, and Joaquim Lloveras. 2014. Three-dimensionaltactile symbols produced by 3D Printing: Improving the process of memorizing atactile map key.
British Journal of Visual Impairment
32, 3 (2014).[20] Jaume Gual, Marina Puyuelo, and Joaquim Lloveras. 2015. The effect of volumetric(3D) tactile symbols within inclusive tactile maps.
Applied Ergonomics
48 (2015),1–10. https://doi.org/10.1016/j.apergo.2014.10.018[21] Jaume Gual-Ortí, Marina Puyuelo-Cazorla, and Joaquim Lloveras. 2015. Im-proving Tactile Map Usability through 3D Printing Techniques: An Experi-ment with New Tactile Symbols.
The Cartographic Journal
52, 1 (2015), 51–57.https://doi.org/10.1179/1743277413Y.0000000046[22] Darren Guinness, Annika Muehlbradt, Daniel Szafir, and Shaun K Kane. 2019.RoboGraphics: Dynamic Tactile Graphics Powered by Mobile Robots. In
The 21stInternational ACM SIGACCESS Conference on Computers and Accessibility . ACM,New York, NY, USA, 318–328. https://doi.org/10.1145/3308561.3353804[23] Michael E. Hahn, Corrine M. Mueller, and Jenna L. Gorlewicz. 2019. The Com-prehension of STEM Graphics via a Multisensory Tablet Electronic Device byStudents with Visual Impairments.
Journal of Visual Impairment & Blindness
Universal Learning Design International Conference ,Vol. 4. 49–56.[25] Leona Holloway, Kim Marriott, and Matthew Butler. 2018. Accessible maps forthe blind: Comparing 3D printed models with tactile graphics. In
Conferenceon Human Factors in Computing Systems - Proceedings , Vol. 2018-April. https://doi.org/10.1145/3173574.3173772[26] Shaun K. Kane, Meredith Ringel Morris, and Jacob O. Wobbrock. 2013. Touch-plates: low-cost tactile overlays for visually impaired touch screen users. In
ACMSIGACCESS Conference on computers and accessibility . ACM, New York, NY, USA,1–8. https://doi.org/10.1145/2513383.2513442[27] Stacy M. Kelly and Derrick W. Smith. 2011. The impact of assistive technologyon the educational performance of students with visual impairments: A synthesisof the research.
Journal of Visual Impairment & Blindness
Information and Software Technology
51, 1 (2009), 7–15.[29] Barbara Kitchenham and Stuart Charters. 2007. Guidelines for performing sys-tematic literature reviews in software engineering. (2007).[30] Karen Koehler. 2017.
Examining the Conceptual Understandings of GeoscienceConcepts of Students with Visual Impairments: Implications of 3-D Printing . Thesis.[31] Kostiantyn Kucher, Carita Paradis, and Andreas Kerren. 2018. The state of the artin sentiment visualization. In
Computer Graphics Forum , Vol. 37. Wiley OnlineLibrary, 71–96.[32] Jonathan Lazar, Suranjan Chakraborty, Dustin Carroll, Robert Weir, Bryan Size-more, and Haley Henderson. 2013. Development and evaluation of two prototypesfor providing weather map data to blind users through sonification.
Journal ofUsability Studies
8, 4 (2013), 93–110.[33] Lorena C León and Luiz C. Martini. 2015. Education and Social inclusion of Peoplewith Visual Impairment in the Study of Mathematical Functions. In
International echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan
Online Conference on Inclusion and Rehabilitation of Visually Impaired .[34] Claudia Loitsch and Gerhard Weber. 2012. Viable Haptic UML for Blind People. ,509–516 pages.[35] Guiseppe Melfi, Karin M"uller, Thorsten Schwarz, Gerhard Jaworek, and RainerStiefelhagen. 2020. Understanding what you feel: A mobile audio-tactile sys-tem for graphics used at schools with students with visual impairment. In
CHIConference on Human Factors in Computing Systems . ACM, New York, NY, USA.https://doi.org/10.1145/3313831.3376508[36] Cynthia D. Mulrow. 1994. Systematic reviews: rationale for systematic reviews.
Bmj
Guidelines and Standards for TactileGraphics
International Journal of Human-Computer Interaction (2020),1–11. https://doi.org/10.1080/10447318.2020.1752464[39] Hari Prasath Palani and Nicholas A. Giudice. 2017. Principles for DesigningLarge-Format Refreshable Haptic Graphics Using Touchscreen Devices.
ACMTransactions on Accessible Computing
9, 3 (2017), 1–25. https://doi.org/10.1145/3035537[40] Konstantinos Papadopoulos, Eleni Koustriava, and Panagiotis Koukouris. 2018.Orientation and mobility aids for individuals with blindness: Verbal descriptionvs. audio-tactile map.
Assistive Technology
30, 4 (2018), 191–200. https://doi.org/10.1080/10400435.2017.1307879[41] Sabrina Ramsamy-Iranah, Martin Maguire, James Gardner, Satyadev Rosunee,and Naraindra Kistamah. 2015. A comparison of three materials used for tac-tile symbols to communicate colour to children and young people with vi-sual impairments.
British Journal of Visual Impairment
34, 1 (2015), 54–71.https://doi.org/10.1177/0264619615610161[42] Andreas Reichinger, Helena Garcia Carrizosa, and Cornelia Travnicek. [n.d.].Designing an Interactive Tactile Relief of the Meissen Table Fountain. In
ICCHP:International Conference on Computers Helping People with Special Needs , KarlMiesenberger and Georgios Kouroupetroglo (Eds.), Vol. Part II. Springer Interna-tional, Switzerland, 209–216. https://doi.org/10.1007/978-3-319-94274-2_28[43] Shigenobu Shimada, Haruka Murase, Suguru Yamamoto, Yusuke Uchida, MakotoShimojo, and Yutaka Shimizu. 2010. Development of Directly Manipulable TactileGraphic System with Audio Support Function. , 451-458 pages.[44] Mathieu Simmonet, Vieilledent Stéphane, R Daniel Jacobson, and Tisseau Jacques.2011. Comparing tactile maps and haptic digital representations of a maritimeenvironment.
Journal of Visual Impairment and Blindness
SIGACCESS Conference on Computers & Ac-cessibility . ACM, New York, NY, USA, 127–137. https://doi.org/10.1145/2700648.2809854[47] Sarit Felicia Anais Szpiro, Shafeka Hashash, Yuhang Zhao, and Shiri Azenkot.2016. How people with low vision access computing devices: Understanding chal-lenges and opportunities. In
Proceedings of the 18th International ACM SIGACCESSConference on Computers and Accessibility . 171–180.[48] Hiroki Taguchi, Toshimitsu Yamaguchi, and Tetsuya Watanabe. 2012. Develop-ment of a Tactile Star Chart Automated Creation System. In
Universal LearningDesign . 100–104.[49] Brandon Taylor, Anind K. Dey, Dan P. Siewiorek, and Asim Smailagic. 2015.TactileMaps.net: A Web Interface for Generating Customized 3D-Printable TactileMaps. In
ACM SIGACCESS Conference on Computers & Accessibility . ACM, NewYork, NY, USA, 427–428. https://doi.org/10.1145/2700648.2811336[50] Jennifer L. Tennison and Jenna L. Gorlewicz. [n.d.]. Toward Non-visual GraphicsRepresentations on Vibratory Touchscreens: Shape Exploration and Identification.In
EuroHaptics 2016 , F. Bello (Ed.), Vol. Part II. Springer International PublishingSwitzerland, 384–395. https://doi.org/10.1007/978-3-319-42324-138[51] Yoshinori Teshima, Tohru Ogawa, Mamoru Fujiyoshi, Yuji Ikegami, TakeshiKaneko, Susumu Oouchi, Yasunari Watanabe, and Kenji Yamazawa. 2010. Mod-els of Mathematically Defined Curved Surfaces for Tactile Learning. In
ICCHP:International Conference on Computers Helping People with Special Needs , Vol. II.Springer, 515–522.[52] Jenna L. Toennies, Jessica Burgner, Thomas J. Withrow, and Robert J. Webster.2011. Toward haptic/aural touchscreen display of graphical mathematics for theeducation of blind students. https://doi.org/10.1109/WHC.2011.5945515[53] Jakub Wabiński and Albina Mościcka. 2019. Automatic (Tactile) Map Gen-eration—A Systematic Literature Review.
ISPRS International Journal of Geo-Information
8, 7 (2019), 293.[54] Zheshen Wang, Baoxin Li, Terri Hedgpeth, and Teresa Haven. 2009. Instant tactile-audio map: enabling access to digital maps for people with visual impairment. In
Proceedings of the 11th international ACM SIGACCESS conference on Computersand accessibility . 43–50. [55] Limin Zeng, Mei Miao, and Gerhard Weber. 2015. Interactive Audio-haptic MapExplorer on a Tactile Display.
Interacting with Computers
27, 4 (2015), 413–429.[56] Ting Zhang, Bradley S. Duerstock, and Juan P. Wachs. 2017. Multimodal Per-ception of Histological Images for Persons Who Are Blind or Visually Im-paired.
ACM Transactions on Accessible Computing
9, 3 (2017), 1–27. https://doi.org/10.1145/3026794
HI ’21, May 8–13, 2021, Yokohama, Japan Butler et. al.
Ref. Presentation Methods Graphic Tasks Participants Findings[1] Tactile graphic; HT condi-tion: flat surface + handtracking + smartwatch feed-back (audio + vibration) Map Find labelled regions 12 BLV adults With filtering HT condition fasterbut without filtering slower; similarcorrectness; tactile and HT condi-tion with filtering preferred[4] Tactile graphic (swell); tac-tile graphic (embossed); 3Dmodel (printed) Map/ Sym-bol Match symbols (point, line,area) 18 BLV adults More accurate with 3D print thanembossed graphic; faster with 3Dprint than embossed and swell tac-tile graphic[5] Tactile graphic (swell pa-per); touchscreen + overlay+ audio labels Map Memorise map then answerquestions about route, land-mark position, layout; testedboth short-term and long-termrecall 24 BLV adults Learning time less with audio-labels; slight preference for audio-labels; no effect on short or long-term recall;[15] Tactile graphic; 3D model(printed) + audio labels Map Recall of spatial layout and in-formation in labels 24 BLV stu-dents Greater recall of 3D model + audiolabels; similar SUS user satisfactionfor both[16] Tactile graphic; touchscreen+ audio + vibration Bar chart;character;geometricshape Recall of bar chart layout & rel-ative bar height; recognise let-ter; determine shape orienta-tion 12 sighted(blind-folded)+ 3 BLV adults Similar accuracy but slight indica-tion tactile more accurate; tactilefaster[21] Tactile graphic (swell); 3Dmodel (printed) Map/ Sym-bol Flat 2D symbols on tactilegraphic vs volumetric 3D sym-bols on 3D model. Find occur-rence of symbol on map. 40 BLV + 16blindfoldedsighted adults More accurate and faster with3D volumetric symbol than 2Dflat symbol; no difference for 3Dprinted 2D symbol and 2D symbolon swell paper[20] Tactile graphic (thermo-form); 3D model (printed) Map/ Sym-bol Flat 2D symbols on tactilegraphic vs volumetric 3D sym-bols on 3D model. Find occur-rence of symbol on map 30 BLV + 16blindfoldedsighted adults More accurate and faster with 3Dvolumetric symbol than 2D flatsymbol[19] Tactile graphic (thermo-form); 3D model (printed) Map/ Sym-bol Flat 2D symbols on tactilegraphic vs mix 2D and volu-metric 3D symbols on 3D print.Task: Memorize symbols 16 BLV + 6blindfoldedsighted adults Better recall of mix of 2D and 3Dsymbols[22] Touchscreen + overlay +audio labels; touchscreen+ overlay + audio labels+ small mobile robots (dy-namic shape changing dis-play) Bar chart Identify maximum, minimumand trend. 7 BLV adults No difference in accuracy or timebut preference for the robots[23] Tactile graphic; touchscreen+ audio label + vibration Numberline, table,pie chart,bar chart,line graph,and map Education content – unclear 22 BLV chil-dren + adults(5 excluded) No significant difference in accu-racy though indication that tactilegraphic may be more accurate forbar chart and line graph.[24] Tactile graphic (embossed);tactile graphic (writing film) Map Readability of braille and otherdata 21 BLV adults Overall preference for embossed pa-per but writing film for durability[25] Tactile graphic (swell); 3Dmodel (printed) Map Understandability of iconicsymbols; route finding; mem-orability; user preferences; un-derstandability 16 BLV adults 3D print preferred and more under-standable; 3D iconic symbols easierto understand; better short term re-call of 3D print but no differencefor longer term recall
Continued on next page echnology Developments in Touch Graphics 2010-20 CHI ’21, May 8–13, 2021, Yokohama, Japan
Table 14 –
Continued from previous page