An enhanced standard computer keyboard system for single ... · Rehabilitation Counseling Program, Department of Educational Psy-chology, University of South Carolina, Columbia, SC

VeteransAdministration

Journal of Rehabilitation Researchand Development Vol . 25 No . 4

Pages 17—24

An enhanced standard computer keyboard system forsingle-finger and typing-stick typing

ROBERT A. CHUBON, PhD, and MISCHELE R. HESTER, BSRehabilitation Counseling Program, Department of Educational Psychology, University of South Carolina,Columbia, SC 29208

Abstract—A prototype keyboard system was developed,using off-the-shelf hardware and software, as an inexpen-sive keyboard-based system to facilitate data entry forsingle-finger and typing-stick typists . Evaluation estab-lished that the system can increase entry rate by 50percent or more . The underlying concepts may provide abasis for developing other configurations that accelerateand simplify computer keyboard use for persons with avariety of hand impairments.

Key words : computer keyboard system, hand impair-ments, macros, single finger and typing-stick typing.

INTRODUCTION

The high cost of many potentially useful hightechnology developments results in their being avail-able to a very limited number of persons . Althoughmany developments enable persons with disabilitiesto perform critical functions heretofore not possible(e.g., create written communication), their efficiencyremains limited compared to the performance ofindividuals without disabilities. Consequently, whereproductivity is concerned, persons with disabilitiesare still left with a substantial handicap which mayrender them non-competitive . Thus, it is imperativethat cost and efficiency be given priority in researchand development activities.

Address all correspondence and reprint requests to Dr . Robert Chubon,Rehabilitation Counseling Program, Department of Educational Psy-chology, University of South Carolina, Columbia, SC 29208 .

The personal computer has evolved into one ofthe most useful high technology tools for personswith disabilities . It may now be considered a formof megaprosthesis . New uses are emerging daily intreatment, work, educational, and home settings,and considerable effort is being directed towarddevelopment of interfaces that enable persons withthe spectrum of severe disabilities to benefit fromthese uses (3,4,8,17) . However, the specially-de-signed and relatively costly input systems for per-sons with disabilities appear to be considerably lessefficient than the standard computer keyboard beingused by the typical nonimpaired, 2-handed touchtypist . Thus, in word processing and other applica-tions in which input speed is critical, the person withhand impairment may be functional, but is still leftwith a substantial disadvantage.

The importance of this problem is further under-scored by a report published by the Office forTechnology Assessment (13) which indicates thatthere are more than 2 .5 million people in the UnitedStates with non-paralytic, upper-extremity impair-ment. When the number of people with impairedhand function stemming from high-level spinal cordinjuries and other disabling paralysis is added tothat group, there is an indication of the substantialnumber of people who are impeded in keyboard use.

Because of cost, availability, and related problemswith special input devices, the most frequent ap-proach to utilization of personal computers byhand-impaired individuals involves using a typingstick (either hand- or mouth-held), or single-finger

17

18

Journal of Rehabilitation Research and Development Vol . 25 No . 4 Fall 1988

entry on a standard keyboard . There is an obviousinefficiency in these approaches, although for mostof those who are able to utilize the techniques, entryrates are probably competitive with those attainablewith available alternative input systems . These prob-lems prompted initiation of a project directedtoward reconfiguring the standard keyboard inputsystem to make it more efficient for single-fingerand typing-stick typists, while containing cost withinmore affordable bounds.

This became feasible with the development ofmemory resident keyboard utility software (e .g.,SuperKey ' ), which enables key reassignment with-out special costly hardware or hardware modifica-tion for IBM-PC's and compatible computers . Ad-ditionally, the widespread availability of low costIBM-type computer keyboards contributes to thepracticality of this approach to keyboard adapta-tion.

KEYBOARD DESIGN

Review of keyboard design-related research andobservation of single-finger and typing-stick typistsrevealed 5 significant factors affecting keyboard use:a) the distance the finger or typing stick must travel;b) the speed at which the stick or finger can bemoved; c) the time spent locating the keys to bepressed; d) interruptions evoked by the need tofigure out how to expedite simultaneous keypresscombinations ; and, e) the number of key strokesrequired for some functions (6) . These 5 factorsprovided the focus for reconfiguration efforts.

An IBM PC-type keyboard and keyboard utilitysoftware (SuperKey), having a combined cost ofapproximately $100, provided the basis for thesystem. SuperKey was most suitable for meeting thesoftware needs of the project because of its low costand its capacity to support a variety of keyboardinput functions, including complete keyboard layoutreassignment and alteration of simultaneous key-press commands to a sequential mode . A flattened"target-like" arrangement was used as the primaryguideline for the key reorganization . Within thisframework, previous research suggested clusteringthe most frequently used letters in the center, with

1 SuperKey is a product of Borland International, 4585 Scotts ValleyDrive, Scotts Valley, CA 95066 .

others arranged in order of decreasing frequencytoward the periphery (3) . The logic of this approachrests on the concept of a "minimal distance"keyboard or scanning board configuration, whichemanated from efforts to develop more efficientalternative input devices by limiting the distance theinput medium must travel (4,10,11,17).

Additional effort was made to keep letters whichfrequently occur together in close proximity to eachother in accordance with research findings (17) . Thework of Solso and Juel (15), and Solso and King(16) provided the usage frequency data required forreassignment of key position according to theminimal distance concept . Their data were derivedfrom the Kucera and Francis assemblage of words,which comprise a corpus of more than a millionwords selected from written English prose (7) . Thisis the largest sample constructed for languageanalysis and is relatively recent (2,14) . The dataderived from these works, and used in rearrangingthe keys, are contained in Table 1 and Figure 1.

The key caps were physically removed from thekeyboard and relocated to keep their characterlabels consistent with the changes . Based on thedata, the most frequently occurring letter is E, and itwas used as the center of the keyboard arrangement.Although it has been suggested that the space barand ENTER key are actually the most frequentlydepressed (17), they were not reassigned to otherlocations . Because the space bar runs across almostthe entire keyboard, and is in close proximity toseveral keys, there would be little accomplished bymoving it . This assumption is supported by the workof others (6) . The RETURN key was not includedamong those considered for movement, becauseappropriate frequency-of-use data was not found,and because the use of the RETURN or ENTER keyis diminished by most word processing programswhich now include "word wrapping" or automaticreturns when the end of a line is reached.

Ultimately, the layout in Figure 2 evolved. It canbe seen that the letter E is actually located nearer thebottom center of the keyboard . Because there are 4alphanumeric key rows, the alternative would havebeen to locate it somewhere in the upper centerquadrant of the arrangement . This placement wastried, but most users preferred the lower position,and in fact, 2 suggested that near reaches in thebottom 2 rows were easier than far reaches . Consid-eration of this preference eventually led to the

19

CHUBON and HESTER : Enhanced Keyboard System

I

E 1

T 1

0 1

A I

I 1

N

I

S I

R I

H

I

L I

D I

C I

U1El 157636 13861 126780 119057 177325 175885 11393751122849150017 172459 137796 15593

TI 156058 170665 182986 134716 151937 126698 113881218130 1127 115493 124658 I

01 12935 128342 175865 124377 171722 120040 126677 113805 132567 135591 IAl 121352 182189 143063 161307 137057 154323 121577 134575 18378 I

11 1100018160166 135926 131964 139218 127844 132212 14151 I

NI 117103 16327 11377

12870 148547 113841 117952 I115281 113453 17205 14713 15866 126849 I

RI 13832

13989 110420 18943 122696 ►HI 1592 1214 121256 12945 ILI 112440 15631 117035 IDI 188 17723 ICl 110730 1

1

M 1

F 1

P 1

G I

W I

Y

1

B 1

V I

K I

X 1

J 1

Q 1

Z IEI42256 113895 122970 117358 118450 110152

122920 139318 110976 17477 11694 11638 12127 IT11121 13350 13106 1713 12862 17883

1605 16 167 11457 ii I 1147 ID132778 157534 119894 18414 121801 17302

111415 19272 13290 1593 12304 140 1386 1A130368 18694 117573 112983 120897 110069 113941 111914 14899 11587 11269 159 11377 I1123651 117124 18090 115087 114837 11270

16822 117947 16022 11690 1150 1434 12759 INI1169 11944 1228 139395 13598 14555

1176 11603 14484 1101 1374 1191 1102 IS15925 1642 18681 11997 12286 15018

11687 178 13528 13 13 1 120 I

R17058 18849 116628 110627 11659 18786

14999 12218 13561 117 180 I 134 IHI429 183 12999 19791 114883 11701

1209 19 1165 1131 I1 118 118 I141276 14339 110176 12404 11044 117679 18914 11125 11574 120 1 115 1123 ID1614 172 132 11132 1488 12366

1142 1612 142 1 1227 167 1 ►C150 Ill 111 I 123 11478

124 11 16003 1896 1 I 115 IU19138 14224 18832 18098 179 1262

111193 1170 1159 1292 12436 14913 1142 ►MI 1160 18289 1269 173 13036

13544 14 167 I1 1 1 13 1Fl 161 162 165 1315

114 I 1114 119 12 I Ii Ip t 125 194 11020

126 13 161 12395 14 I IGI 119 1632

138 Ii 179 I 1 I 110 1

WI 1340

174 I 1151 113 I1 12 16

Yl 16048 1204 1418 184 12 I 1163 IBI 1214 142 12 1626 I 17VI 12 11 I 11 19 1KI I I 117 111 Ixl I 17 I

JI I I I

41 1ZI

Figure 1.Combined forward and backward bigram frequencies.

placement of more alphabet keys at the bottom ofthe arrangement.

No usage data were available for the numeric,punctuation, and utility keys, but a review of severalsample essays and other examples of general writtencommunication in an English textbook suggestedthat numbers are used relatively infrequently andthat the period, comma, quotation marks, andapostrophe are most common . The semicolon, co-

Ion, hyphen, and question mark seem to fall inbetween . This cursory study provided the basis forplacement as pictured . It is noted that the numerickeys were located around the periphery of the layoutto reflect their low frequency, and the miscellaneouskeys were used to fill in the remaining unusedpositions and appropriately paired to maintainlogical combinations.

A final consideration in determining key place-

20

Journal of Rehabilitation Research and Development Vol . 25 No . 4 Fall 1988

Table 1.Rankings of letters by overall frequency of use and first position use.

Rank Ordered Overall Use Rank by FirstLetter

Frequencies Percentage Position Occurrence

E 577583 12 .62 12

T 427179 9 .33 7

0 350121 7 .65 18

A 348411 7 .61 4

I 336166 7 .34 14

N 325652 7 .11 17

S 297531 6 .5 1

R 281881 6 .15 9

H 252191 5 .51 11

L 188261 4 .11 13

D 181054 3 .95 8

C 142336 3 .11 2

U 124736 2 .72 20

M 116574 2 .54 5*

F 107219 2 .34 10

P 93040 2 .03 3

G 89499 1 .95 15

W 86563 1 .89 16

Y 78749 1 .72 24

B 70714 1 .54 5*

V 45707 .99 19

K 29838 .65 21

X 9032 .19 26

J 7296 .15 22

Q 4936 .10 23

Z 4316 .09 25

*Estimates for these two letters were essentially the same and they were both ranked fifth.

ment was the likelihood that a letter would be thefirst of a word. Since words are usually preceded bya space, it was concluded that there would be anadvantage to placing letters most likely to beginwords near the space bar if consistent with the otherplacement criteria. To determine which letters werecandidates for this placement, the numbers of words

beginning with each was estimated from the respec-tive number of pages in Webster's New WorldDictionary (5).

Another consideration in the enhancement of thekeyboard was to facilitate the visual scanning orkey-locating process . Observation of typing-stickusers revealed that their mode of keyboard entry

21

CHUBON and HESTER : Enhanced Keyboard System

Figure 2.Reassigned key positions .

321 V U P 6 7 8+4 Q M I T S C K Z 9:

5 J G N R E H B Y X 0/ ' F 0 A D L W,

----------------------------------------------------

C

was heavily dependent upon visual scanning, unliketouch-typing, which is predominantly dependentupon kinesthetic feedback (1) . Measures were imple-mented to assist the user in developing a scanningpattern consistent with the target-like arrangement.Generally, English reading/writing people learn toscan from left to right . However, with this targetarrangement, it was felt that it would be advanta-geous if users maintained focus on the center of thekeyboard and learned to use this as their referencepoint, since the layout was designed to concentratethe keyboard activity in this area . To this end, thenumbers were arranged to curve around from nearthe top center of the keyboard in order of occur-rence. Thus, as can be seen in Figure 2, the numbers1 through 5 run counterclockwise, and 6 through 0are arranged in clockwise fashion.

The second measure taken to facilitate scanningand location of the keys was to color code them to

highlight the overall pattern, and to provide adiscriminable characteristic in addition to the labelson the keys. The "bull's eye" letter E was coveredwith a red transparency to make it stand out . Thenumbers were made blue, punctuation marks green,and the miscellaneous letters yellow . The tilde (— )key, which serves as a special function key, wasrelocated in the upper left corner and made orange.Color coding was accomplished using commercially-produced, transparent, stick-on key top covers,which allowed the character labels to show through.The resultant target-like appearance can be seen inFigure 3.

Two approaches were taken to eliminating key-strokes. First, the use of macros, that is, theassignment of words, phrases or other multiplecharacter strings to a defined and relatively limitedkeying sequence, was incorporated into the scheme.The software provides for the creation of these aids.

Figure 3.Reconfigured keyboard .

22

Journal of Rehabilitation Research and Development Vol . 25 No . 4 Fall 1988

Usually macro strings are derived from frequently-typed material, such as a person's signature, whichcan be assigned to a keying combination like Alt S,and thereafter typed in entirety by entering thislimited keystroke combination . Because this projectwas directed at more general usage, a set of macrowords was derived from the Kucera and Francisword corpus analysis (7) . The frequencies of the 200most-used words were multiplied by their length toobtain products that reflect the relative number ofstrokes required by their use when typing substantialamounts of ordinary written English . The 5 wordsyielding the largest products were utilized for mac-ros . Thus, the following words were assigned akeying sequence consisting of the Alt key and analphabet character (or the Ctrl key and the alphabetcharacter to create capitalized macros) : that, from,with, this, and which.

A second means of eliminating keystrokes wasderived from the macro concept . Because spacing isusually required after punctuation marks, the macroconcept was used to configure the system to auto-matically insert 1 space following the comma andsemicolon . Upon pressing the appropriate punctua-tion key, 2 spaces are automatically inserted follow-ing the period, question mark, and colon . Theapostrophe and quotation mark keypress functionswere not changed because of their frequent usewithout trailing spaces.

If it becomes necessary to use these punctuationmarks without trailing spaces, one possible solutionis to backspace to the desired position after enteringthe reconfigured punctuation marks ; however, thesoftware creates an "undo" key (the orange-coloredtilde key mentioned above) which returns the key-board to its nonmacro state for 1 keypress, and thisprovides an effective solution.

The final enhancement to the keyboard inputsystem was to utilize the software's "single-finger"mode . This enables entering keyboard commands,which normally would require simultaneous, multi-ple keypresses (e .g ., shift and a letter key to typeupper case), to be entered as sequential commands.

EVALUATION

The reconfigured keyboard input system wasevaluated in a research study that followed a formalexperimental paradigm . Control and experimental

groups, each comprised of 3 adults having little orno typing experience, engaged in two 90-minutesingle-finger typing sessions . Subjects who wereinexperienced in typing were used to eliminate theintrusion of negative transfer as a confoundingvariable (12). The control group was instructed topractice single-finger typing, using a standard IBM-PC keyboard and a word processing program . Theexperimental group was oriented to the features ofthe reconfigured keyboard system, and instructed topractice single-finger typing with the altered key-board and the same word processing program usedby the control group . Additionally, all participantswere provided with written instructions for theirsystems to use as needed.

The sessions, scheduled a week apart, involvedtyping practice exercises contained in a typinginstruction textbook (9). During the last 5 minutesof each session, 5-minute proficiency tests wereadministered to the participants . Following thesecond session and proficiency test, a debriefinginterview was conducted with each participant, andeach member of the experimental group was alsoasked to complete a rating scale developed to obtaintheir opinions about the value of the enhancements.

RESULTS

The data from the typing tests conducted at theend of the second practice sessions were subjected tostatistical analysis . The number of errors was sub-tracted from each subject's total number of keypressentries to obtain proficiency scores . The meancalculated from the scores of the standard keyboardgroup was 264.7 (sd = 27 .4), and the mean for thereconfigured keyboard group was 409 .7 (sd = 29 .7).Review of individual scores revealed that all 3subjects in the reconfigured keyboard group scoredhigher than the highest score in the standardkeyboard group . A t test supports the conclusionthat the difference between the means was signifi-cant (p< .005, t = 5 .69, df = 4). From another per-spective, the experimental group rate was approxi-mately 55 percent faster than that of the control orstandard keyboard group.

Responses from the debriefing interviews alsorevealed a notable difference . All 3 subjects in thecontrol group (standard QWERTY-arranged key-board) reported experiencing finger fatigue during

23

CHUBON and HESTER : Enhanced Keyboard System

practice, while none of the experimental groupmembers cited this as a problem. Examination ofthe rating scale data obtained from the reconfiguredkeyboard group suggested that all of the enhance-ments contributed positively to their performance,including both speed and accuracy, except the macrowords . They attributed problems in using the mac-ros to their inability to quickly learn and recall thekeystroke sequences, and suggested that the se-quences were "unnatural" or not compatible withthe flow of thought and concentration commandedby the typing activity.

DISCUSSION

This limited study suggests that the standardcomputer keyboard entry system can be enhanced toimprove the proficiency of hand-impaired typists.Although the study does not indicate the maximumentry rates which may be attainable by single-fingerand typing-stick typists using this reconfiguredkeyboard system, substantial gains may be realized.Moreover, the study confirms that relatively inex-pensive avenues to enhancement do exist.

It is also noted that this keyboard arrangementmay not be of benefit to persons using other than asingle-finger or typing-stick entry technique. Forexample, persons using 2 typing sticks or 2 fingersmay actually find the arrangement to be a hin-drance. Because the activity is concentrated in themiddle of the keyboard, 2 or more entry media maycollide and interfere with each other . On the otherhand, the project outcome supports the notion thatit may be possible to discern and create an optimalkeyboard arrangement to accommodate any hand-impaired individual and his or her unique residualfunction . The low cost of keyboards and keyboardutility software, and the flexibility inherent in thesoftware approach to reconfiguration, make the ideaplausible.

The key to widespread customizing of keyboardentry systems may lie in the development of analyti-cal algorithms that enable rapid evaluation ofvarious configurations as they are being tried bypotential users . Similarly, software designed toanalyze unique kinds of material that users will betyping may provide a basis for optimal, individually-tailored key arrangements and macros . Programsthat could analyze an individual's text files to reveal

their unique usage patterns (including word andletter frequencies), would likely provide a betterbasis for keyboard rearrangement than generalusage material . It is noted that some of these issuesare being addressed at Tufts University (11).

Attempts to develop analytical algorithms must bepursued carefully, however . This study, as well asothers cited, suggest that to be most effective,reconfiguration must take into consideration a num-ber of human factors that can vary substantiallyfrom individual to individual . Included are matterssuch as visual scanning rate, extremity/digit strength,stamina, dexterity, alternating extremity/digitresponse time and sequential response time, andcognitive factors such as memory and attentioncapabilities . Consequently, to be effective, analyticalalgorithms will have to be able to produce muchmore than arithmetically-determined key arrange-ments based on typing or other entry materialcontent and the number of fingers or typing aidsused. Each additional critical factor may cause thecomplexity of the required algorithm to increasegeometrically . The most feasible approach may usea moderately encompassing design algorithm, usertrial, and an algorithm to analyze system perfor-mance.

In sum, the experience gained from this projectindicates that there are numerous relatively inexpen-sive, off-the-shelf productivity enhancements thathold potential benefit for persons with disabilities.However, their optimal use may rest with thedevelopment of new assessment and design tools.

REFERENCES

1. Alden DG, Daniels RW, Kanarick AF: Keyboard designand operation: A review of major issues . Human Factors14 :275-293, 1972.

2. Cress CJ, Vanderheiden GC : Applications of multiplewordstudy analysis in development of vocabularies foraugmentative communication systems . In Proceedings ofthe 9th Annual Conference on Rehabilitation Technology6 :351-53 . Washington, DC : Association for the Advance-ment of Rehabilitation Technology, 1986.

3. Getschow CO, Rosen CJ, Goodenough-Trepagnier C : Asystematic approach to design of a minimum distancealphabetical keyboard . In Proceedings of the 9th AnnualConference on Rehabilitation Technology 6 :396-398.Washington, DC : Association for the Advancement ofRehabilitation Technology, 1986 .

24

Journal of Rehabilitation Research and Development Vol . 25 No. 4 Fall 1988

4. Goodenough-Trepagnier C, Levine S: An optimal, inex-pensive text entry system for the orthopedically andneurologically disabled . Rehabilitation R & D Progress

11.Reports 22(4) :134 . Washington, DC : Office of Technol-ogy Transfer, Veterans Administration, 1985.

5. Guralnik DB (Ed.) : Webster's New World Dictionary,Second College Edition, New York: World PublishingCo., 1970 .

12.6. Kinkead R : Typing speed, keying rates, and optimal

keyboard layouts . In Proceedings of the Human Factors

13.Society 19th Annual Meeting, 159-161 . Santa Monica,CA: Human Factors Society, 1975.

7. Kucera J, Francis WN : Computational Analysis of

14.Present-Day American English, Providence : Brown Uni-versity Press, 1967.

8. Lee CC, Vanderheiden GC, Rasmussen A : One finger

15.operation of the IBM family of personal computers . InProceedings of the 9th Annual Conference on Rehabilita-tion Technology 6 :265-266 . Washington, DC : Association

16.for the Advancement of Rehabilitation Technology, 1986.

9. Lloyd AC, Rowe JG, Winger FE: Typewriting forColleges, New York : McGraw-Hill, 1964 .

17.10. Levine SH, Goodenough-Trepagnier C, Rosen MJ,

Getschow CO : Adaptive technique for customized inter-face design with applications to nonverbal communica-tion . In Proceedings of the 9th Annual Conference onRehabilitation Technology 6 :399-401 . Washington, DC :

Association for the Advancement of Rehabilitation Tech-nology, 1986.Minneman SL : Keyboard optimization technique to im-prove output rate of disabled individuals . In Proceedingsof the 9th Annual Conference on Rehabilitation Technol-ogy 6 :402-404 . Washington, DC : Association for theAdvancement of Rehabilitation Technology, 1986.Noyes J : The QWERTY keyboard : A review . Interna-tional Journal for Man-Machine Studies 18 :265-281, 1983.Office of Technology Assessment . Technology and Hand-icapped People, Washington, DC: U.S . GovernmentPrinting Office, 1982.Shapiro SI, Gordon GP : Contemporary norms of wordand phonetic frequencies . Journal of Verbal Learning andVerbal Behavior 10 :92-94, 1971.Solso RL, Juel CL : Positional frequency and versatility ofbigrams for two- through nine-letter words . BehavioralResearch Methods and Instrumentation 12 :297-343, 1980.Solso RL, King JF : Frequency and versatility of letters inthe English language . Behavior Research Methods &Instrumentation 8 :283-286, 1976.VerVoort S, Winningham D, Swarts A, Krouskop T: Acomputer keyboard design for people who use typing-sticktyping . In Proceedings of the 9th Annual Conference onRehabilitation Technology 6 :394-395 . Washington, DC:Association for the Advancement of Rehabilitation Tech-nology, 1986.