[268] COMPUTERIZED ADAPTIVE METHODS FOR SELECTING HEARING AIDS
Donald Dirks, PhD
VA Medical Center, West Los Angeles 90073
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C432-4RA)
No report was received for this issue.
[269] THE ROLE OF IMAGERY IN AUDITORY COMPREHENSION IN BRAIN-DAMAGED ADULTS
Marilyn Selinger, PhD
Denver VAMC, 1055 Clermont Street, Denver, CO
80220
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C818-RA)
PURPOSE--The purpose of this project is to examine the contribution of imagery as an aid to auditory comprehension of connected language in aphasic and right hemisphere lesioned individuals. Specifically, the following questions are being asked: 1) Do imagery laden verbal passages enhance accuracy of comprehension on verbal tasks? 2) Do imagery laden verbal passages increase inter- and intrahemispheric differences in right- and left-brain damaged patients as measured by probe auditory evoked potentials (AEP)?
METHODOLOGY--The PC/Signal Averaging Program, a probe-evoked potential technique, in which a task-irrelevant sensory stimulus is superimposed on ongoing complex tasks, is used to measure intra/interhemispheric response to the task. The paradigm includes a baseline or non-differentiating task and two language or left hemisphere tasks. The baseline task provides a comparative measurement of processing a task that has never been shown to differentially bias one hemisphere over the other. The two language tasks include passages rated as high imagery and passages rated as low imagery by nonimpaired subjects. In addition, multiple choice questions are asked following each language passage. These questions are used as measures of the subjects' comprehension of the material as well as an indicator of their involvement in the task. The questions require both literal and interpretive conclusions to be made about the material. Aphasic patients whose PICA Overall severity levels fall between 55th-85th percentile are included as subjects. Right-hemisphere-damaged patients whose overall scores on the PICA fall between 55-85th percentile serve as pathological comparisons. In addition, a non-brain-damaged control group is included. Each subject undergoes two sessions of electrophysiological testing resulting in evoked potential measures of hemispheric responsivity and test/retest data.
PROGRESS--Nonimpaired controls and left-hemisphere-damaged subjects have been participating in the project for the past 10 months. Background behavioral testing continues as new subjects are being identified. Evoked potential testing of these subjects is ongoing.
FUTURE PLANS--Additional subjects will continue to participate in the study for the oncoming year. Once a small but representative number of subjects from each experimental group completes the paradigm, preliminary data will be analyzed. Final results will not be reported until the running of the experiment is completed for all subjects.
[270] CHANGES IN AUDITORY ABILITIES WITH HEARING AID USE
Lenore Holte, PhD; George Haskell, PhD; Christina Runge,
MA; Cynthia Hogan, PhD
Iowa City VA Medical Center, Iowa City, Iowa 52246;
University of Rochester, Rochester, New York 14642; email:
holte.lenore@iowa-city.va.gov
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C847-RA)
PURPOSE--Recent investigations suggest that the benefit of amplification for speech understanding increases over the first few weeks or months of hearing aid use for some listeners. The purpose of this project is to determine the effects of daily hearing aid use on speech recognition abilities and to test one hypothesis regarding an explanation for any changes in these abilities. Specifically, we will explore whether increases in hearing aid benefit occur over a 6-month period in novice hearing aid users, and, if so, whether these increases in benefit are accompanied by an increased reliance on high-frequency information in speech.
METHODOLOGY--Sixty subjects with hearing impairment will be divided into three groups of 20 and evaluated using the Nonsense Syllable Test (NST). One control group will be comprised of experienced hearing aid users who wear their aids daily. Another control group will be novice hearing aid users who do not wear their aids outside the laboratory during the study. The experimental group will be novice users who wear their hearing aids daily. All hearing aids will be fit binaurally according to the NAL-R prescriptive formula. These groups will be followed over a period of 6 months and tested at monthly intervals. At each visit the following measures will be performed: 1) The NST in the unaided condition, in an aided condition with the NAL-R frequency response, and in an aided condition with a flatter frequency response. Hearing aid benefit will be defined as aided minus unaided performance. 2) Crossover frequency for NST syllables presented at a level that insures audibility of the signal for each listener. This is a measure of the contributions of various frequency regions of speech to speech intelligibility and is the frequency that divides the speech spectrum in half with regard to information used by the listener. 3) Subjectively preferred frequency response (NAL-R vs. a flatter response) while listening to a sample of running speech and a sample of music.
Changes in benefit of amplification to nonsense syllable recognition will be compared for the three subject groups. If daily use of amplification results in an increase in hearing aid benefit, it is expected that an increase in benefit will be demonstrated only for the experimental group using their prescribed frequency response. If increased reliance on high frequency speech cues is responsible for this change in benefit, it is anticipated that crossover frequency will increase for the experimental group over the study period, but not for the two control groups.
PROGRESS--Nonsense syllables have been recorded in eight randomizations on digital audio tape and the test has been adminstered to three pilot subjects with unimpaired hearing to begin to explore the reliability of performance on these lists. The project has recently been moved from the Syracuse VA Medical Center to the Iowa City VA Medical Center. The laboratory has been equipped and subject recruitment is underway. There are no preliminary data at this time.
FUTURE PLANS--Testing will soon begin on subjects with hearing impairment to determine reliability in this population and to establish the optimal signal-to-noise ratio for test administration. Subjects will continue to be recruited into the three study groups. Work will simultaneously begin to bring more of the experimental operations under computer control.
[271] IS THERE AN "ACCLIMATIZATION EFFECT" WITH HEARING AIDS?
Carol A. Sammeth, PhD; Maureen P. Coughlin, MA
Roudebush VA Medical Center, Indianapolis, IN 46202;
Indiana University Medical Center, Indianapolis, IN 46202;
email: sammeth@indianapolis.va.gov
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C840-RA)
PURPOSE--In 1992, Gatehouse coined the term "acclimatization effect" to refer to improvement in speech recognition scores seen over time in a small sample of persons using hearing-aids following their fitting. He believed that the improvements were due to their learning to use amplified speech cues now available to them, rather than depending simply on the stimulation of auditorily deprived ears. Since that time, however, conflicting data have been reported as to whether or not such a phenomenon exists. The present study will examine the acclimatization effect in new users of hearing-aids with a carefully designed study which controls for test-retest variability of the speech recognition tasks, evaluates the effects of changes in user preferred volume wheel setting over time, and uses both linear and nonlinear amplifiers.
METHODOLOGY--Data will be collected on 30 adults with bilaterally symmetrical sensorineural hearing losses ranging from mild to severe, who have not previously used amplification. Subjects will be fit with binaural linear and nonlinear sets of hearing aids via probe-microphone using NAL-r prescribed gain for a 65 dB SPL input signal. Aided speech recognition will be evaluated using the adaptive Hearing-In-Noise-Test (HINT) to determine signal-to-noise ratio (SNR) for 50 percent correct sentence intelligibility and the Speech-Perception-in-Noise (SPIN) test at a fixed SNR of either +4, +8, or +12 dB depending on individual performance, to avoid ceiling and floor effects. A unique aspect of this study is that each subject will serve as his/her own control. There will be several repeat sessions measuring aided speech recognition during a 2-3 week period prior to the subjects wearing one of the hearing aid sets home; this will assess across session test-retest reliability. Subsequently, repeat measurements will be made over a 4-month wearing period. At each of these sessions, testing will be accomplished with both hearing aid sets, and at user-preferred as well as prescribed-gain volume wheel setting for the hearing aid set worn daily. Questionnaire data will be collected over time to examine any changes in perceived hearing aid benefit. Another group of 10 experienced users of hearing-aids will be evaluated as a control group.
PROGRESS--During the past year, the research laboratory was fully equipped and calibrated. Pilot data were collected to establish appropriate measurement parameters, resulting in some minor changes to the research protocol. Data collection is nearly completed on seven subjects, and subject recruitment continues.
PRELIMINARY RESULTS/IMPLICATIONS--There are not enough data at this time to report any preliminary results. Determination of the existence of an acclimatization effect with hearing aids is crucial because such an effect would have significant implications for both clinical practice and research protocols. If there is a period of acclimatization to new amplification, longer trial periods will be necessary prior to measurement of user benefit and decision making regarding changes in electroacoustic parameters.
[272] EFFECT OF LACK OF AMPLIFICATION ON PERSONS WITH UNILATERAL HEARING LOSS
Shlomo Silman, PhD; Carol A. Silverman, PhD; Michele B.
Emmer, MS; John Lutolf, PhD
VA Medical Center, East Orange, New Jersey 07019;
Brooklyn College, CUNY, Brooklyn, NY 11210; Hunter College,
CUNY, New York, NY 10010
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C665-2RA)
PURPOSE--This study was to investigate the effect of presence and absence of amplification on various measures of audition in aided and unaided adults with unilateral sensorineural hearing impairment.
METHODOLOGY--Aided and unaided subjects between 25 and 75 years of age with unilateral, sensorineural hearing impairment were evaluated annually in Years 1, 2, and 3. Measures included puretone thresholds, speech-recognition threshold, W-22 speech-recognition score (SSRS), nonsense syllable test (NST) score, acoustic admittance, brainstem auditory evoked potentials, and middle latency responses.
PROGRESS--This study supports the presence of auditory deprivation in unilateral sensorineural hearing loss ears. Amplification appears to slow or prevent the process of deprivation.
RESULTS--Of the 83 subjects (41 aided and 42 unaided) seen in Year 1, 48 (27 unaided and 21 aided) completed the study. Comparisons were made between the aided ears and the unaided hearing-impaired ears of the subjects. The analysis of variance showed that in unaided ears, the mean SSRS declined significantly (p=0.0003). There was no significant change in the NST in unaided ears; this is probably related to the initial poor NST in this population and/or to the fact that 3 years were not enough to produce change. There was no significant change in aided ears for both NST and SSRS; however, a trend was noticed toward improvement in both.
FUTURE PLANS/IMPLICATIONS--Further analysis of data will be made, and we plan to publish the data in the near future in several articles.
[273] EFFECT OF PRESENCE VERSUS ABSENCE OF PROLONGED AMPLIFICATION ON AUDITION
Shlomo Silman, PhD; Carol A. Silverman, PhD; Michele B.
Emmer, MS; John Lutolf, PhD; Stanley A. Gelfand, PhD
VA Medical Center, East Orange, New Jersey 07019;
Brooklyn College, CUNY, Brooklyn, NY 11210; Hunter
College, CUNY New York, NY 10010
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C578-2RA)
PURPOSE--This longitudinal study investigates the effect of presence versus absence of amplification on various measures of audition in monaurally aided (MA) and binaurally aided (BA) adults with bilateral, sensorineural hearing impairment (BSHI). These measures have been evaluated annually for the last 4-6 years. Measures included pure-tone thresholds, speech-recognition threshold, W-22 suprathreshold speech-recognition score (SSRS), nonsense syllable test (NST) score, acoustic admittance, brainstem auditory evoked potentials, and middle latency responses.
PROGRESS--This study supports the previously published retrospective studies on the presence of auditory deprivation in the unaided ear of monaurally aided subjects with bilateral sensorineural hearing loss.
RESULTS--Although we are in the process of completing the final stage of the study and writing the final report; we will highlight some important findings.
With subjects who were tested at least 3-6 years, the upper limits (UL) and the lower limits (LL) of the 95 percent critical differences limits (CDL) were used for the SSRS. As for the NST the UL and LL of the 95 percent CDL were established for each individual NST score based on a list of 56 items. The retest NST score was compared with the UL and LL to determine if the retest NST score differs significantly from initial score. A total of 64 subjects completed the study (33 subjects binaurally aided, 31 monaurally aided.) A total of 12 out of 31 monaurally aided ears (37 percent), showed deprivation in speech-recognition score. Interestingly, 3 subjects who showed deprivation on the SSRS in the fourth year, recovered in the sixth year and were included in the 31 undeprived ears. However, the recovery amounts to improvement of approximately 2-4 percent, which caused them to be included within the 95% (CDL). However, the score in the sixth year was remarkably poorer than the initial score.
Five out of 66 ears of the 33 subjects of binaurally aided ears showed deprivation, (3 percent), and the result of the NST indicated that 8 out of unaided 31 ears fell outside the LL, while none of the aided ears did. As for the binaurally aided ears, 10 percent fell outside the LL. Eight subjects changed status from monaural to binaural and are being monitored. No significant changes were obtained for the other measures.
[274] EVALUATION OF NONAUDITORY FACTORS WHICH AFFECT HEARING AID USE IN ELDERLY VETERANS
Bruce Z. Rappaport, PhD; Stephen A. Fausti, PhD
VA Medical Center, Albuquerque, NM 87108; VA Medical
Center, Portland, OR 97207
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C570-2RA)
PURPOSE--The primary goal of this investigation is to identify critical nonauditory factors which affect hearing aid benefit in elderly veterans. Nonauditory factors include mental status, fine motor coordination, visual acuity, social support, and motivation. A secondary goal is to compare traditional auditory measures, and self-report measures of perceived hearing aid benefit. Outcome measures include the Hearing Handicap Inventory-Elderly (HHIE), The Hearing Disability Interview/Hearing Aid Benefit Interview (HDI/HABI), and the Hearing Aid Satisfaction Questionnaire (HAS).
Our specific aims are to evaluate several auditory and nonauditory factors in a group of elderly veterans who are to receive hearing aids through the VA. A multiple regression analysis will be used to assess the predictability of each of the nonauditory factors as well as some of the auditory factors measured during the hearing aid evaluation. Decisions of which veterans could benefit most from hearing aids could be improved by giving the clinician more objective data for predicting outcomes with hearing aid use, and programs of aural rehabilitation could be maximized by attacking those factors that limit hearing aid benefit.
METHODOLOGY--Two samples of subjects will be studied. One sample is composed of outpatient veterans who are scheduled for clinic visits to be evaluated for hearing aids. The second sample consists of veterans in institutionalized settings such as Nursing Home Care Units, Domiciliaries, and Contract Nursing Care Facilities. Each will receive an initial evaluation which will include the following: an assessment of auditory function; assessment of nonauditory factors; and self-assessment of hearing handicap. At 4 months following fitting and dispensing of selected hearing aid(s), subjects will be reevaluated on all auditory and nonauditory measures.
PROGRESS--Complete data are available on over 200 subjects, and analysis of these data shows significant decrease in perceived hearing handicap after 4 months of hearing aid use. Correlation coefficients were computed for each of the variables measured at enter into the study, and the difference score on the total HHIE between entry and follow-up. Eight of the 10 highest correlations were with nonauditory factors. The 10 variables with the largest correlations were then used in a stepwise linear regression. The result was a correlation of 0.52 based upon five variables. Four out of the five variables entered in the stepwise regression are nonauditory variables. These five variables are able to predict approximately 30 percent of the variance of the improvement in the HHIE.
[275] DEVELOPMENT OF AN AUTOMATED TECHNIQUE FOR CLINICAL TINNITUS EVALUATION: A PILOT STUDY
Stephen A. Fausti, PhD; James A. Henry, PhD; Curtin R.
Mitchell, PhD
VA Medical Center, Portland, OR 97207; Department of
Otolaryngology, Oregon Health Sciences University, Portland,
OR
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Pilot Project #C1693-PA)
PURPOSE--Tinnitus refers to the chronic experience of phantom sound. Severe tinnitus is a pathological condition suffered by at least 10 million Americans. Although veterans are particularly impacted by tinnitus, the VAMC system provides minimal clinical management for the condition. The DVA pays approximately $90 million per year in compensation benefits to over 80,000 veterans who are service-connected for tinnitus. This laboratory is addressing the need for standardization of tinnitus management procedures. This study is designed to develop a reliable technique for the measurement of tinnitus, using computer automation.
METHODOLOGY--A computerized psychoacoustical testing system has been developed to run a tinnitus measurement protocol. The system has been programmed to automatically obtain hearing thresholds, and tinnitus loudness and pitch. A video monitor, directed at the patient, displays instructions for responding and serves to receive responses via touch-control with a pen device. Subjects with stable tinnitus are recruited and evaluated with the automated technique. To assess intra- and inter-session reliability of the measures, each measurement is obtained repeatedly within and across sessions for each subject. Reliability of the measures is also assessed between examiners.
PROGRESS--This work has demonstrated the feasibility of using a computer-automated system to obtain reliable measurements of tinnitus loudness and pitch. Preliminary results show consistency of within-subject measurements over time, which is essential for quantification of the disorder and subsequent rehabilitative strategies for remediation.
RESULTS--To date, 20 patients with tinnitus have been evaluated with the automated system. Results of preliminary statistical analyses have indicated reliability of responses: a) within-subject, within-session; b) within-subject, across-session; and c) between-examiner (between groups). Using a manually controlled protocol, an independent group of 20 patients has also been evaluated for tinnitus loudness and pitch. Because the automated testing protocol was patterned after the manual procedure, a comparison of test-retest reliability of the measures between the two techniques was appropriate. An analysis of between-session differences (absolute values) in loudness matches for each subject revealed that the differences were significantly reduced at frequencies ;me2 kHz using the automated procedure.
FUTURE PLANS--This research endeavor is working toward standardization of measurement techniques and rehabilitative procedures associated with the treatment of tinnitus. Future research is designed to: 1) refine the prototype system to optimize test reliability and sensitivity, and minimize testing time; 2) demonstrate that the automated system can achieve test reliability comparable to results obtained when the same testing is conducted manually by trained examiners; and 3) demonstrate between-examiner uniformity of test results in a large group of subjects. Development of an automated system for tinnitus evaluation has the potential to significantly impact the clinical management of veterans, with application beyond the VA health care system. With successful completion of the current project, an automated technique will be fully documented for the clinical assessment of tinnitus. This accomplishment will result in a technique that will minimize between-examiner variability and promote uniformity between clinics. The new methodology can thus facilitate the standardization of tinnitus measurement, provide a means for validating tinnitus claims and make a substantial contribution to tinnitus management and rehabilitation.
[276] EARLY DETECTION OF HEARING LOSS FROM OTOTOXIC AGENTS BY HIGH-FREQUENCY AUDITORY EVALUATION
Stephen A. Fausti, PhD; P. Douglas Noffsinger,
PhD
VA Medical Center (151J), Portland, OR 97207; VA Medical
Center (126), West Los Angeles, CA 90073
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C227-2RA)
PURPOSE--Patients receiving treatment with ototoxic drugs are at risk for acquiring communicatively disabling hearing loss and/or balance disorders. Early detection of ototoxicity allows implementation of intervention strategies that can prevent hearing disability. When intervention is not possible, early detection allows time to prepare the patient and/or family for aural rehabilitation strategies. A multicenter (four-site) study has been established to determine whether loss of hearing sensitivity due to ototoxicity is detectable in the high frequency (9-20 kHz) range of hearing prior to the conventional frequency (0.25-8 kHz) range.
METHODOLOGY--Patients hearing sensitivity is monitored prior to, during, and following treatment with the aminoglycoside antibiotics (AMG) amikacin, gentamicin or tobramycin, or the chemotherapeutic agent cisplatin (CDDP). Subject inclusion criteria for study was hearing threshold sensitivity at or below 100 dB SPL at test frequencies up to 9 kHz. Ototoxic change criteria are those recommended by the American Speech-Language-Hearing Association (ASHA) in the national guidelines for ototoxicity monitoring, adopted by ASHA from data provided by this research.
PROGRESS--Currently, 452 subjects (237 AMG and 215 CDDP) have provided data, with 294 of these subjects demonstrating threshold change in one or both ears. From the AMG-treated subjects, 35 percent of initial hearing losses were bilateral and 65 percent were unilateral, while 73 percent of initial losses from CDDP-treated subjects were bilateral and 27 percent were unilateral.
Presenting the data relative to ears rather than subjects, hearing change was seen in 461 ears. Results were analyzed to determine whether changes were initially detected in the high frequency, conventional frequency, or in both high and conventional frequency ranges concurrently. In the AMG-treated ears, 53.3 percent of initial changes occurred in the high frequencies, 29.7 percent in the conventional frequencies and 17.0 percent in both frequency ranges concurrently. In the CDDP-treated ears, 41.2 percent of initial changes occurred in the high frequencies, 24.7 percent in the conventional frequencies, and 34.1 percent in both frequency ranges concurrently.
Preliminary analyses indicated that at frequencies where hearing thresholds were greater than 100 dB SPL, little or no hearing change was seen throughout treatment. Also, in most cases, initial change seemed to appear within very few frequencies. Further analyses revealed a range of five frequencies within which approximately 90 percent of all hearing changes would have been initially detected. This 5 Frequency Range (5FR) is specific to the hearing threshold configuration of each individual, and includes the highest test frequency with hearing threshold sensitivity at or below 100 dB SPL and the next four lower test frequencies. Frequencies above this range accounted for about 2 percent of all changes, while frequencies below this range accounted for about 8 percent of all changes.
IMPLICATIONS--Data from this project have resulted in the development and publication of national guidelines for ototoxicity monitoring. Identification of a five-frequency range that is highly sensitive to initial ototoxic hearing change has implications for testing of ill patients who are capable of giving reliable responses for short periods of time, but are unable to withstand the rigor of behavioral threshold testing at all frequencies (i.e., limited-responsive patients). A software program has been developed that flags the 5FR of each individual and automatically reports any significant changes in hearing thresholds. This program is supported by a miniaturized hospital ward-testing system which reduces testing time and increases patient accessibility without compromising the sensitivity of the monitoring effort.
FUTURE PLANS--In the next phase of this multicenter research effort, we will prospectively explore, in a large sample of patients receiving ototoxic agents, the comparative sensitivity of objective measures of monitoring hearing (ABR; OAE) for early detection of ototoxicity. This research will result in measures that will allow the inclusion of all types of hospital patients (responsive, limited-responsive, and unresponsive) in comprehensive ototoxicity monitoring programs. Secondary goals of this research are planned to explore the relationship between ototoxicity and nephrotoxicity.
[277] MEASUREMENT AND PREDICTION OF BENEFIT FROM AMPLIFICATION
Robyn M. Cox, PhD; Kay M. Pusakulich, MA; Genevieve C.
Alexander, MA; Izel Taylor, MA
Memphis Speech and Hearing Center, Memphis, TN 38105;
email: coxrm@cc.memphis.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C344-4RA)
PURPOSE--Our goals are to describe and develop methods of quantifying and predicting hearing aid benefit and satisfaction for the everyday life of elderly listeners.
METHODOLOGY--To explore the relationship between cognitive variables and objective hearing aid benefit, experienced hearing aid wearers provided data on working memory, speed of mental processing, use of auditory context, and mental flexibility. We employed multiple regression techniques to explore the extent to which cognitive variables add to the prediction of hearing aid benefit.
An evaluation of the Visual Input/Output Locator Algorithm (VIOLA) method for selecting and fitting nonlinear hearing aids was performed. Linear and nonlinear hearing aids were worn by a manikin and subjected to speech inputs at various levels. Sound levels developed at the eardrum were compared to levels predicted by the VIOLA method.
Relationships between several psychological variables and self-assessed hearing aid benefit were explored. Hearing aid wearers provided data on the Abbreviated Profile of Hearing Aid Benefit (APHAB) inventory as well as on measures of extroversion, anxiety, and locus of control. Multiple regression techniques were used to determine whether personality attributes are significantly related to subjective benefit.
Two clinically useful inventories are under development: one to measure post-fitting satisfaction and another to measure pre-fitting expectations of hearing aid performance.
PROGRESS--Forty-five hearing aid wearers provided data on cognitive variables and hearing aid benefit.
To evaluate the VIOLA fitting strategy, data were collected for six commercially available linear and compression behind-the-ear hearing aids fitted to a KEMAR manikin.
Eighty-three elderly hearing aid wearers (successful and unsuccessful) provided data on self-assessed hearing aid benefit and the three psychological variables.
Structured interviews about the important components of hearing aid satisfaction were conducted with 21 hearing aid owners. In addition, a survey of the importance of 14 elements in satisfaction was completed by 160 hearing aid owners. Principal components analyses were used to determine underlying commonalities among the elements. Item selection and testing is underway.
RESULTS--Of the four cognitive variables explored, only mental flexibility made an independent contribution to the prediction of objective hearing aid benefit. It was estimated that mental rigidity costs about 8 points of benefit for a typical person and about 18 points for a rigid individual.
The VIOLA fitting strategy was found to give good predictions of earcanal speech levels when hearing aids were operating linearly and well below pure tone saturation level. Nonlinear hearing aids produced levels lower than the predicted level and the amount of error was proportional to the compression ratio. The VIOLA procedure could be modified to account for these effects.
It was found that certain personality attributes are related to the way that individuals interpret the efficacy of amplification. Outward oriented (extroverted) persons reported more speech communication benefit. Individuals who felt more under the control of other persons tended to display greater negative reactions to environmental sounds.
Four basic factors have been identified that contribute to hearing aid satisfaction: sound merit (benefit and quality), comfort (physical and psychological), cost (personal and financial), and cosmetics.
[278] NOISE REDUCTION FOR HEARING AIDS
James Kates, EE; Gabrielle H. Saunders, PhD
Department of Speech and Hearing, Graduate Center of the
City University of New York, New York, NY 10036
Sponsor: National Institute on Deafness and Communication Disorders
PURPOSE--The main complaint of persons with hearing impairment is difficulty hearing speech in a noisy environment. Our overall goal, therefore, is to improve the performance of hearing aids in noise by using some form of directional processing. More specifically we will develop and evaluate a multi-microphone array for hearing aids that can be worn either on the head (e.g., in a pair of eyeglasses) or used as a hand-held device. Past research has shown that microphone arrays can improve the S/N ratio by suppressing the interfering noise coming from directions other than the desired source. However, past researchers have used computer simulations rather than real rooms for their evaluations, and often have carried out only acoustic measurements, rather than speech intelligibility testing. The array to be evaluated here consists of five omnidirectional microphones uniformly spaced over a unit that is 10 cm long. A previous study in which this array was examined using acoustic measurements showed significant improvements in weighted S/N ratio for superdirective array processing.
METHODOLOGY--The speech and noise signals were processed in four ways, in each of two rooms that differed in reverberation time. Rooms 1 was a non-reverberant office, Room 2 was a reverberant conference room. For each condition, recordings were made in the room of interest and the data processed offline. For the recordings, the microphone system was surrounded by six loudspeakers. The speech signal was played through the speaker at 0° azimuth, noise was played through the speakers at 60, 105, 180, 255, and 300°. In two conditions, the 5-microphone array was used and the recordings obtained were processed using both delay-and-sum beamforming and superdirective processing. In a third condition, a single omnidirectional microphone, and in the forth condition, a single directional (cardiod) microphone was used. The processed speech and noise files were stored on computer and played back to the subjects during testing.
Eighteen hearing impaired male subjects took part in the experiment. They carried out two different measures of speech intelligibility: a speech reception threshold in noise for 50 percent correct performance was measured using spondees and a subjective threshold for 50 percent understanding was obtained using short passages of connected discourse.
PROGRESS--All 18 subjects have been tested with the computer stored stimuli. The results are sufficiently promising that a wearable array will be built and evaluated.
RESULTS--The experimental results showed the superdirective array processing technique to be superior to the other techniques evaluated. It provided improvements of 4 dB S/N over the delay-and-sum beamforming, and 5 dB S/N over the single omnidirectional microphone. The relative performances of the processing techniques were not affected by the type of room, although performance in the non-reverberant room was significantly better than in the reverberant room. The two methods of measuring speech intelligibility produced similar results.
FUTURE PLANS/IMPLICATIONS--The next stage in the study is to build a binaural array that will be tested in the laboratory, following this a wearable binaural array will be built and its efficacy in the real world evaluated.
[279] AN INTERACTIVE VIDEO SYSTEM TO TEST AND TREAT NONLITERAL LANGUAGE DISORDERS
Diana Van Lancker, PhD; Edward Hall, PhD; Muriel
Goldojarb, MA
VA Outpatient Clinic (126), Los Angeles, CA 9001;
Department of Communication Disorders, California State
University at Northridge, Northridge, CA 91330; Audiology
and Speech Pathology Service, Sepulveda VAMC, Sepulveda, CA
91343; email: dianavl@vm.uscedu;
ehall@csun.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C788-RA)
PURPOSE--The goal of this research is to assess and treat deficits in language comprehension in both left and right hemisphere stroke patients utilizing nonliteral language. The objective is the increase in comprehension of nonliteral language, leading to increases in functional communication.
METHODOLOGY--Left and right hemisphere stroke patients are randomly assigned to a treatment or nontreatment group. All patients receive extensive language testing, along with computerized assessment of nonliteral and literal language comprehension. In the computerized testing, videotaped scenes are presented to subjects along with written and spoken phrases; subjects must decide if the phrases (either literal sentences or nonliteral expressions) match the scenes presented. In the training component, subjects receive additional exposure to computer-presented scenes from popular movies (e.g., North by Northwest); subjects receive feedback about their performance. Assessment of treatment continues throughout, so that incremental progress is measured. Normal age and education-matched controls receive only the testing portion of the protocol and several measures of nonliteral language comprehension for comparison purposes.
PROGRESS--The developmental phase of the project is complete. Two full Nonliteral Language Workstations and one Testing Only Workstation are fully operational. To date, 30 left-hemisphere, 10 right-hemisphere stroke patients and 28 controls are in the process of completing, or have completed, the protocol.
PRELIMINARY RESULTS--Subjects have been found to successfully navigate the computerized program, despite often severe expressive and subjective language impairment.
FUTURE PLANS--Continued running of subjects to reach goals of 60 right- and 60 left-hemisphere subjects is planned. Recruitment of subjects has been extended beyond the original two VA Medical Centers to a third (West Los Angeles VAMC) as well as to nearly all stroke rehabilitation groups in the northern Los Angeles metropolitan area. A prototype of a production version of the workstation for use is in development.
[280] COMPUTER-ASSISTED SPEECH REHABILITATION SYSTEM
James A. Till, PhD
VA Medical Center, Long Beach, CA 90822; University of
California, Irvine College of Medicine, Orange, CA 92668;
email: till.james@forum.va.gov
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C468-4RA)
PURPOSE--This project is to develop, test, and refine a clinical software system to assist in behavioral training and analysis of speech production. The objective is to provide a flexible set of easy-to-use software tools for routine use by speech clinicians with a variety of patients. Ten protocols are being developed to assist self-directed and clinician-directed training in the control of breathing, voice, prosody, and nasality during speech. Acoustic, aerodynamic, nasal accelerometric, and electroglottographic signals are monitored depending on the protocols selected. The software will be compatible with the more than 40 existing work stations developed in previous grant periods for Computer Assisted Speech Evaluation and Rehabilitation (CASPER).
METHODOLOGY--Programs are written in Delphi-Pascal. Algorithms are being developed and tested. Single-subject clinical trials will be completed on 20 subjects.
PROGRESS--A highly flexible system for defining and analyzing patient responses has been incorporated into the software. This considerably expands the general utility of the system and allows the clinician to easily assemble hundreds of custom protocols tailored for individual patients. Raw voltage from transducers, windowed RMS, or time-integral signals may be selected for display and analysis. Tokens within a data sweep are defined in terms of signal amplitude and duration above or below quiescent signal levels. Additional criterial amplitude and duration limits are entered allowing the clinician to require correct responses with mean or peak signal characteristics above, below, or between these limits. Real-time graphics required to implement the analog feedback have also been completed this year. The interface for designing the teaching paradigm has also been improved to allow user-generated sets of audio-visual alerting signals, cues, models, and feedback to be scheduled for cyclical, random, or ratio presentation. The time intervals between all stimulus events, the data collection sweep, feedback events, and restimulation are also controlled by the software.
RESULTS--Five pilot subjects and two clincians have participated in training sessions thus far. Suggestions for software improvements have been incorporated into the system. Informal results suggest that patients find the software useful and enjoyable. Some patients can use the software with only intermittent supervision by a speech clinician. There are not sufficient data to asses treatment efficacy at present.
FUTURE PLANS--An experienced speech clinician will be hired and assist in completing 20 single-subject experiments with speech disordered subjects. In addition, we will provide beta-site versions of the software to selected users. Both activities will result in further refinements of the clinical software and help define the default, predesigned protocols to be included in the software.
[281] USING SELF-MONITORING TO IMPROVE COMMUNICATIVE EFFICIENCY IN APHASIA
Janet L. Whitney, PhD
Audiology and Speech Pathology Service (126), VA Medical
Center, Miami, FL 33125
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C693-RA)
PURPOSE--Our objectives were 1) to determine the efficacy and effectiveness of self-monitoring treatment for subjects with mild and moderate aphasia; 2) to determine the social validity of treatment, that is, how unsophisticated listeners rate the effects of treatment; and 3) to determine the influence of speech, language, cognitive, and demographic variables on treatment effectiveness.
METHODOLOGY--Subjects were selected based on proportion of disfluencies (greater than 10 percent of total words) in pretreatment picture descriptions. They were exposed to four conditions: baseline, training, independent self-monitoring, and follow-up. The treatment task in all conditions was Norman Rockwell picture descriptions. At the end of baseline, subjects were taught to self-monitor: each subject listened to one of his own (audiotaped) baseline picture descriptions and was asked, "What bothers you about your speech?" Subjects identified a target disfluency (multiple repetitions, revisions, or filled pauses): "I'm stuttering," or "It sounds broken up." Subjects were trained to listen to themselves in on-line picture description and to stop at the first occurrence of a disfluency. Proportion of disfluencies in the treatment task and in two generalization tasks constituted the data points for each session. The project used a multiple-baseline design across subjects and across behaviors. Effectiveness of self-monitoring treatment was shown by consistent replication of a treatment effect (reduction in proportion of disfluencies) each time treatment is applied, across subjects and across behaviors (disfluencies) in each subject.
PROGRESS--Four of an initial six subjects completed the treatment program. Subjects 1 and 3 pretests indicated moderate aphasia, Subject 2, mild-moderate, and Subject 5, mild aphasia. Subjects 1, 3, and 5 showed an immediate and dramatic drop in target disfluencies when self-monitoring was introduced. All four subjects showed proportion of overall disfluencies generally at or below lowest baseline levels during independent self-monitoring and in follow-up. Treatment efficacy (impact of picture description) was more robust than treatment effectiveness (generalization tasks). Subject 5 (mild) showed the most stable decrease in disfluencies in both picture description and generalization tasks. Communicative efficiency, defined as Content Information Units/minute, improved in posttreatment picture description for Subjects 1, 3, and 5. Social validity was defined as ability of naive listeners to perceive improvement (significant at <0.05) in subjects' posttreatment speech. For all subjects listeners heard fewer filled pauses. Ratings for Subject 1 showed significant improvement in organization, intelligibility, and naturalness. Subject 2 ratings indicated improvement in stuttering. For Subjects 3 and 5 listeners indicated significantly fewer repetitions, revisions, and improved naturalness; Subject 3 also showed improvement in organization and fewer wrong words. Ratings for Subject 5 indicated perceived improvement in intelligibility.
FUTURE PLANS--The project is completed. All subjects, regardless of severity, were able to identify disfluencies, label them in a personally meaningful way, and monitor themselves on-line for occurrences of disfluencies. All subjects identified at least one important strategy: "slow down," or "think about my speech." It appears essential to train for generalization of self-monitoring, during picture description, personal narrative, and conversation. Subject 5, who improved most, had a more active social life and may have been more motivated to improve. For people with more severe or more long-standing aphasia, group treatment and family involvement may help to increase the stakes, and enhance motivation for better speaking. Self-monitoring resulted in subjects identifying for themselves the need to slow down and think about what they were saying. Future research might compare efficacy, effect, and efficiency of self-monitoring training and pacing, a treatment approach designed specifically to decrease rate, to decrease disfluencies and improve communicative efficiency.
[282] CONNECTED SPEECH DEVIATIONS OF APHASIC AND NON-BRAIN-DAMAGED ADULTS
Robert H. Brookshire; L.E. Nicholas
VA Medical Center, Minneapolis, MN 55417
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C817-RA)
PURPOSE--The overall objective of the proposed research is to determine which characteristics of aphasic adults' connected speech are likely to have the strongest effects on their communicative success in daily life. Specific objectives are: 1) to develop reliable measures of disruptive speech behaviors in the connected speech of aphasic adults; 2) to quantify the frequency and nature of disruptive speech behaviors exhibited by aphasic adults and by non-brain-damaged adults, and to assess how aphasic adults and non-brain-damaged adults differ with regard to these behaviors; and 3) to evaluate the extent to which normal listeners' judgments of the quality and adequacy of aphasic and non-brain-damaged adults' connected speech are related to objectively measured characteristics of the connected speech such as macrostructural integrity, communicative efficiency, and the presence of disruptive speech behaviors.
METHODOLOGY--Connected speech samples from non-brain-damaged adults and adults with aphasia are transcribed and scored for the presence of several specific behaviors, as well as for overall communicative efficiency and informativeness. Audiotape recordings of non-brain-damaged or brain-damaged speakers are then played to nondisabled persons who make subjective judgments about the communicative success of the speakers and the degree to which the speech characteristics cause various subjective reactions in the listeners.
RESULTS--We are currently evaluating the effects of speakers' production of what we call performance deviations (false starts, revisions, unnecessary repetition, producing error words, and so on) on listeners' subjective reactions to the speaker. We have devised a standard, rule-based system for quantifying the occurrence of performance deviations in connected speech, and have analyzed the connected speech of large groups of adults with and without brain damage. We have found that nondisabled speakers and those with aphasia do not meaningfully differ in their productions of several categories of performance deviations, but that large and potentially meaningful differences exist for other categories. We have completed a preliminary study of normal listeners' subjective judgments of ease of comprehension and listening effort as they listen to brain-damaged or nondisabled adults produce spoken narratives, and are preparing a larger study based on the results of the preliminary study.
IMPLICATIONS--The results of this research promise to affect clinical management of aphasic patients. Sensitive and reliable methods for quantifying the disruptive speech behaviors of aphasic adults will provide for databased decisions about the effectiveness of specific treatment approaches, and enable clinicians to formulate treatment procedures and objectives based upon reliable measures of communicative anomalies. Knowledge of which disruptive speech behaviors have the greatest effects on normal listeners' subjective judgments of communicative success will enable clinicians to focus treatment on aspects of connected speech that have the greatest potential for improving daily-life communication.
[283] APHASIC NAMING DEFICITS: EFFECTS OF DEEP- AND SURFACE-LEVEL TREATMENTS
Robert C. Marshall, PhD; Donald B. Freed, PhD
VA Medical Center, Portland, OR 97207; University of
Rhode Island, Kingston, RI 02881; California State
University, Fresno, CA 93740; email:
rcmarsh@uriacc.uri.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C789-RA)
PURPOSE--This research compared the effects of an associative learning procedure, personalized cueing with a traditional surface-level training procedure, phonological cueing, on the long-term naming of realistic stimuli by subjects with aphasia and without brain damage.
METHODOLOGY--Thirty-nine adults with aphasia (APH) and 40 adults with no brain damage (NBD) meeting certain selection criteria participated in the study. All subjects were randomly assigned to personalized cueing (PERS) or phonological cueing (PHON) conditions. Experimental stimuli consisted of 40 photographs of subordinate members of the semantic categories dogs and birds that subjects were unable to name on pretesting. Subjects were taught the names of 20 of the dogs via PERS or PHON cueing procedures. Ten additional dog pictures and 10 bird pictures served as untrained, semantically related, and untrained semantically unrelated, control items, respectively. In the PERS condition, the examiner verbally provided individualized personalized cues created by each subject in a pretraining procedure before asking the subject to name the dog in the photograph. In the PHON condition, the examiner provided surface-level phonological information in the form of the first sound of the dog name and the number of syllables in the word. For the 20 untrained stimuli, the examiner presented the picture of the dog or bird and named the item for the subject. Subjects completed 12 training sessions. During each session all trained and untrained items were presented twice in random order. Naming probes, which involved presentation of stimulus pictures without cues or feedback from the examiner, were administered 1 week, 30 days, and 6 months after training was completed.
RESULTS--Both groups of subjects attained a high degree of accuracy during training with the NBD group outperforming the APH group by approximately 20 percent. To ascertain the effects of training, separate three-way ANOVAs with repeated measures were performed for APH and NBD subjects. These examined differences among conditions (PERS vs. PHON cueing) stimuli (trained dogs, untrained dogs, untrained birds), and naming probes (1 week, 30 days, 6 months). Results failed to support the superiority of PERS cueing shown in prior investigations using "novel" training stimuli for either group. Naming probe scores for both APH and NBD groups declined from probe-to-probe suggesting that items learned in training were forgotten without training. Naming probe scores (f percentage of items correctly named) for APH subjects for the 1 week (M=53; SD=29), 1 month (M=40; SD=27), and 6 month (M=20; SD=18) probes were all significantly different from each other (P<0.001). APH subjects performed significantly better on the trained dogs (M=44 correct; SD=28) than the untrained dogs (M=32; SD=26), and the untrained birds (M=37; SD=31). Analyses of results are presently considered in their preliminary stages; many additional analyses are forthcoming.
[284] ANALYSIS AND TREATMENT OF APRAXIC SOUND ERRORS
Julie Wambaugh, PhD; Patrick J. Doyle, PhD
Audiology & Speech Pathology, Research Service (151R),
Highland Drive VA Medical Center, Pittsburgh, PA 15206;
email: pdoyle@pitt.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C692-4RA)
PURPOSE--The primary objectives of this investigation are to 1) determine the stimulus generalization effects of a minimal contrast treatment for apraxia of speech; 2) evaluate the effects of sequential modification of treatment in terms of additional stimulus generalization; 3) study the effects of treatment on general measures of verbal production; 4) perform acoustic comparisons of correctly and incorrectly articulated sounds, as produced by apraxic patients pre- and post-treatment, in order to examine the relation among such sounds; and 5) determine the relation between acoustic measures of apraxic speech and raters' direct magnitude estimates of degree of accuracy of sound production.
METHODOLOGY--Combined single subject and group experimental designs are being utilized to examine the effects of treatment. Multiple baseline designs across subjects, behaviors, and conditions are being used to evaluate the effects of treatment on production of specific sounds in treated and untreated words across several stimulus conditions within individual subjects. Concurrently, a repeated measures group design is being employed to evaluate the effects of treatment on additional aspects of speech production.
Sixteen apraxic/aphasic subjects will receive treatment, applied individually in a manner consistent with a multiple baseline design. Continuous baseline, treatment, and maintenance probes will be conducted for each individual subject to evaluate production of specific sounds. Additional measures of speech production (e.g., percent consonants correct on consonant inventory, PICA verbal score) will be obtained from each subject at pre- and post-treatment intervals in order to determine group effects of treatment.
In addition to the preceding pre- and post-treatment measures, samples of incorrectly and correctly produced sounds are being collected and analyzed by acoustic means. Pre-treatment word lists are prepared for each subject on the basis of performance on a consonant inventory probe. These lists are designed to elicit incorrect productions of sounds that individual subjects misarticulate. Following acquisition of incorrectly produced sounds, word lists are prepared to elicit correct productions of homonymous words. Non-brain-damaged speakers are being matched for age, sex, and regional dialect to each apraxic subject. Correct productions of each of the words utilized in the apraxic speaker's sample are being elicited from the normal speakers for purposes of comparison. Acoustic analyses will be specific to selected sounds and the Computerized Speech Lab is being utilized for the measurements.
PROGRESS--Five apraxic/aphasic subjects have been entered into treatment to date. Treatment has been completed for three of the subjects, as have corresponding acoustic analyses.
PRELIMINARY RESULTS--All of the subjects have demonstrated positive acquisition and response generalization effects of treatment for all sounds trained, with the exception of one subject whose positive results were limited to two of the three trained sounds. Unlike previous investigations, across-sound generalization has been observed for three of the subjects. To date, sequential modification of treatment across stimulus contexts has resulted in positive stimulus generalization effects for subjects who have reached that phase of treatment.
Acoustic analyses of incorrect and correct sound productions have been completed for three subjects. Results indicate that for two of the three, sound errors appeared to be characterized as distortions rather than substitutions. Findings for the third subject were inconclusive.
[285] SMITH-KETTLEWELL EYE RESEARCH INSTITUTE: SELECTED PROJECTS
John A. Brabyn, PhD
Rehabilitation Engineering Research Center, The
Smith-Kettlewell Eye Research Institute, San Francisco, CA
94115-1821; email: rerc@skivs.ski.org; web:
http://www.ski.org/rerc
PURPOSE--The Smith-Kettlewell Eye Research Institute is dedicated to research on human vision. It was founded to encourage a productive collaboration between the medical clinic and the scientific laboratory. The research staff, numbering some fifty professionals, is drawn from diverse scientific and medical backgrounds, including ophthalmology, neurology, experimental psychology, engineering, physics, optometry, biophysics, and audiology.
Here are some highlights of only a few of the many vision research projects currently being undertaken at Smith-Kettlewell.
Development and Evaluation of Active VEP Electrodes For Pediatric Vision Research
Our program of vision impairment assessment for infants and young children is influenced by the convenience with which electrodes to measure visual evoked potentials (VEPs) can be applied and removed. This is especially true of the multihandicapped population targeted for our present 5-year research plan. The availability of suitable electrodes, easy to apply and remove and without the need for electrode paste, would provide a quantum leap in practicality and convenience in addressing the vision problems of this and other target groups. Accordingly, we have initiated collaboration with Wen H. Ko & Associates of Cleveland to explore new electrode technologies.
We have evaluated initial prototype multigigohm active electrodes previously fabricated by this firm for an EKG application. We designed and built an interface specifically for our VEP application, and obtained pilot data in our RERC Pediatric Vision lab. These data were supplied to the manufacturer for the development of new active electrodes specifically for the VEP application. These electrodes incorporate a semiconductor amplifier on the electrode itself and require only water to enhance contact with the skin. We believe this device has a high likelihood of providing enhanced practicality in the vision screening of infants and neonates, especially those with combined vision, cognitive, hearing, and physical disabilities.
Talking Signs® Development, Applications and Evaluation
The Talking Signs system is our principal project on orientation and mobility for blind and visually impaired individuals. The system consists of infrared transmitters placed at the locations of signs, and a pocket-sized receiver carried by the blind user. The receiver decodes the message from the transmitter and verbally informs the user of its message.
We have developed a new and upgraded system of Talking Signs transmitters for use in crosswalks on controlled intersections. The new transmitters utilize improved infrared LEDs to reach across streets up to 100 feet wide, and provide the following information to the user: the name of the street about to be crossed, the block in which the user is located, the direction in which the user is walking, and the crosswalk signal condition. The transmitter is designed so that the walk/don't walk message fades out when users veer off the crosswalk, providing them with guidance for remaining within the designated boundaries of the crossing. The prototype transmitters have been installed on the intersection of 5th and Market Streets in San Francisco in collaboration with the San Francisco Department of Parking and Traffic. This is a complex 5-way intersection which borders on the entrance to the Powell Street BART/Muni transit station (also comprehensively outfitted with Talking Signs) and thus provides an ideal test site. User evaluations have been extremely positive.
With supplementary funding from Project ACTION, and the collaboration of the San Francisco bus system (Muni), we have conducted pilot studies of Talking Signs needs for buses and bus stops. A focus group composed of eight research subjects was used to identify the perceived needs of users. In addition to the San Francisco Muni, we have also worked with several other transit authorities.
A pilot study in the Powell Street BART/Muni station to determine the usefulness of Talking Signs to those with cognitive impairments has been completed. Results were extremely encouraging and indicated that Talking Signs have a definite role to play in assisting the cognitively impaired gain improved access to public facilities.
Among many new Talking Signs installations are San Francisco's new public toilets, a city office building, the new Main Public Library, and the new Muni light-rail stops being installed on the Embarcadero. Recent installations elsewhere include The Lighthouse, New York, Washington Metro (one station as a trial), and the Texas School for the Blind.
Access to Visual Displays, Laboratory Instruments, and other Consumer and Vocational Products
We are undertaking several projects aimed at improving access to a wide variety of instruments and displays. One example is the possibility of making modern laboratory instruments such as digital oscilloscopes accessible to the blind. Our earlier Auditory Oscilloscope project made regular analog oscilloscopes usable by blind individuals, using variable-pitch tones to represent vertical displacement of the trace on the screen as a cursor was scanned across from left to right. With the advent of digital instruments with RS-232 or other interfaces, the opportunity exists to adapt these earlier solutions for a wider range of instrumentation. We are presently working with a sample oscilloscope with both RS-232 and IEEE bus connections to devise a practical approach to the accessibility problem. A recurring problem for persons without sight is the need to conduct simple measurements. We have developed a device to duplicate a carpenter's steel measuring tape.
We and others have often proposed the development of a standard for infrared links to and from products with visual displays, based on our experience with Talking Signs technology. The concept would require the information normally sent to a visual display to be routed through an infrared LED whose transmissions could be monitored by a suitable transceiver with speech and other output. This approach appears to offer the hope of a practical, low-cost solution if the needed standards and protocols can be devised and adopted. Accordingly, we are now participating in a new initiative spearheaded by the Trace RERC known as the Universal Disability Infrared Link Protocol, to develop suitable standards to benefit all persons with impairments.
Dexter, the Robotic Fingerspelling Hand
Dexter, our robotic fingerspelling hand project is undergoing development to improve its overall performance as well as interfacing with a TDD to enable telecommunications reception. New electro-mechanical components have been designed to improve the hand's physical performance, and an ALVA braille display configured in preparation to develop the software for the proposed TDD interface. Upstart Robots plans to have prototype hands completed in the near future for field testing. Results of earlier tests have provided useful information for necessary upgrades in the hand's functions.
The SKILL Card: Measuring Hidden Visual Impairments
Based on our findings with the SKILL Card in an earlier pilot study of 15 "recovered" optic neuritis (ON) patients, we embarked on a collaborative study with an ON study group that has been following a group of 448 individuals with acute demyelinating ON. The original purpose of this multicenter clinical trial was to determine the efficacy of corticosteroids in the treatment of ON. The SKILL Card has been incorporated in the vision test battery being used to track the course of recovery and/or recurrence of vision loss in these patients. The patients tested with the Card had their initial attack of ON a minimum of 5 years earlier. Nonetheless, though 72 percent had normal (better than 20/20) visual acuity, and 47 percent had normal contrast sensitivity, only 29 percent had normal SKILL score, indicating that this score is sufficiently sensitive to detect long-term (permanent) subtle abnormalities. The results confirm the ability of the SKILL Card to measure the real-world vision impairments of which many patients complain, even though their visual acuity as conventionally measured may be normal.
The SKILL Card has been requested by and sent to more than 50 persons and/or research groups. It is being employed in a variety of settings including driving studies, drug studies, glaucoma and diabetes studies, and aging studies. In fact, the Card has been shown to be a very sensitive indicator of the presence of amblyopia.
Compendium of Technology
We have developed a compendium of technology to act as a resource guide and "idea stimulator" for blind persons and rehabilitation professionals involved in placing blind individuals in employment situations. Although the main focus of the compendium is on vocational instruments, other areas of device applications are also included. It is hoped that the existence and distribution of this compendium will help to stimulate employment of blind individuals in less stereotyped areas, and stimulate demand for specialized vocational instruments: increasing such demand would ease the way toward having such devices manufactured commercially. To date, we have completed the braille, computer disk, and cassette versions of the compendium; the illustrated large print version will be completed shortly.
[286] VISUAL CORRELATES OF MOBILITY IN THE VISUALLY IMPAIRED
Thomas K. Kuyk, PhD; Jeffry L. Elliott, MA; Patti W.
Fuhr, OD, PhD; John Biehl, MA
Birmingham VA Medical Center, Birmingham, AL 35233;
email: kuyk@vision.vsrc.uab.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C775-RA)
PURPOSE--The purpose of this project is to investigate relationships between different aspects of visual function and independent ambulation (mobility performance) in individuals with mild to severe visual impairments.
METHODOLOGY--Subjects are veterans with varying degrees of vision loss from mild to severe and divided by type of vision loss into three groups: acuity loss (primarily age-related macular degeneration and glaucoma); field restriction (primarily retinitis pigmentosa); combination acuity loss and field restriction. Each subject is administered a mobility behaviors questionnaire and the following vision tests: high and low contrast letter acuity, glare sensitivity, Pelli-Robson contrast sensitivity, color confusion (d-15), stereopsis, scanning reaction time, figure-ground, motion sensitivity, spatial contrast sensitivity, and Goldmann visual field. In an indoor experiment, a high-density obstacle course and an open hallway course are completed under high and low illumination conditions. In an outdoor experiment, the obstacle course and a 4 block outdoor mobility route are completed under the two illumination levels. Performance measures are time to complete a course and number of errors made.
PROGRESS--The indoor experiment is complete, with 88 subjects tested. The outdoor experiment is nearly complete with 65 subjects tested.
RESULTS--Both time and errors on the mobility tasks are equally effective measures of performance. The following comparisons are based solely on the error data. Performance on the indoor tasks was modeled by considering the quality of visual information at three levels of the visual system: type of vision loss, sensory visual function, and perceptual visual function (information processing). Each level is significantly correlated with performance. However, combining the sensory and perceptual levels provides the best predictive model of performance accounting for 46, 36, 25, and 30 percent of the performance variance on the high and low illumination obstacle course and high and low illumination hallway course, respectively. Adding type of vision loss to the model does not improve prediction. Subjects' self-ratings of their travel abilities were also correlated with errors and adding self-ratings as a level in the model in combination with the sensory and perceptual levels improved prediction modestly, but significantly to 48, 41, 27, and 37 percent for the tasks in the order listed above. Other findings include: 1) a significant increase in errors (roughly double) as a result of decreasing light level; 2) step-over and low contrast objects are contacted most frequently on the obstacle course; 3) searching with hand or foot and stops are the most common errors on the hallway course, but tripping is the behavior most adversely affected by reducing illumination; 4) subjects without a peripheral field restriction perform significantly better than those with one; and 5) self-report results indicate the subjects in this study had their greatest difficulties with single steps (up or down), adapting to illumination changes, poor illumination, small objects in the travel path, unfamiliar environments, and glare.
FUTURE PLANS--One of our future goals is to continue developing models of mobility in the visually impaired, with specific emphasis on the inclusion of nonvisual factors such as cognitive function, problem solving, and personality. We also plan to examine travel-related injuries/incidents, such as falls, and whether or not current variables in our model predict these. Finally, we plan to determine if training on visual information processing tasks such as visual search can improve mobility performance.
[287] AGE VARIANCE IN NYSTAGMUS SUPPRESSION: A PILOT STUDY
Frederick E. Cobb, PhD
VA Medical Center, Bay Pines, FL 33504
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Pilot Project #C1830-PA)
PURPOSE--The purpose of the research is to evaluate the need for age specific normative values for an electronystagmography subtest called fixation suppression. A continuum of subject age, with the sixth decade of life being the dividing point, serves as the comparison for this single test procedure. The long-term goal is to enhance electronystagmography test sensitivity and specificity.
METHODOLOGY--This between-groups pilot study is evaluating a total of 80 subjects, who are 20-49 and 60-90 years of age, with a single test protocol. Nonimpaired volunteers who meet inclusionary criteria will, upon informed consent, provide the following data sets: otoscopic, medical history, acoustic immittance, basic audiometrics, smooth horizontal visual pursuit, and vestibulo-ocular reflex induced nystagmus with an index of nystagmus reduction produced by visual fixation during caloric irrigations of the external ears. All data are saved using commercially available computers and allow for electro-oculographic eye movement recordings with automated eye movement data analysis. Testing takes about an hour.
PROGRESS--Currently the project is in the data collection phase with 100 percent of volunteers meeting the inclusionary criterion. At this writing, 28 subjects have been tested with 24 in the younger age group and 4 in the elder. Plans for public awareness are in place to increase the harvest of subjects in the elder group. Currently, statistical analysis is not possible.
PRELIMINARY RESULTS--All 28 subjects were able to complete the protocol without adverse reaction. Mean and standard deviation data for horizontal smooth visual pursuit demonstrate consistent findings among subjects, and in comparison with the literature, with a gain factor near one for horizontal pursuit tasks and without evidence of vestibular weakness, direction preponderance, nor failure of fixation suppression. The raw data are currently being placed into a spreadsheet. It is anticipated that the elder age group will provide horizontal pursuit gains of between 0.8 and 1 (80 to 100 percent), produce normal caloric responses in terms of vestibular weakness and directional preponderance, and finally provide evidence that failure fixation suppression index data set are separate and distinct from the younger group. Lastly, a methodological change has been requested to retest a limited subgroup to allow an evaluation of validity.
FUTURE PLANS--Should the research hypothesis be accepted as true, there will be recruitment of a larger group of subjects to add to this data set (including sixth decade data) in order to construct age-specific normative values for the failure fixation suppression index (FFSI). Further, we wish to evaluate additional experimental factors related to the FFSI such as room light and color, subject-to-target distance, fixation target size and lumen requirements, and visual background effects along the age continuum.
[288] DEVELOPMENT OF A DATABASE OF CANE TECHNIQUES
Bruce B. Blasch, PhD; William R. De l'Aune, PhD
VA Rehabilitation Research and Development Center
(151R), Decatur, GA 30033
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C813-2RA)
PURPOSE--The goal of the 3-year study is to construct a database of cane techniques used by blind individuals. This database will be used to establish norms and to evaluate cane techniques over time. This information will be applied to problems of evaluation, development, and design of mobility techniques, accessibility standards and Electronic Travel Aids (ETAs).
METHODOLOGY--A mobility client profile has been developed in conjunction with the orientation and mobility staffs of five of the VA Blind Rehabilitation Centers (BRCs). Data in the profile is obtained from the mobility instructor, the client, and through site visits to the home environment of the client. Information about travel patterns and use or disuse of cane techniques is recorded before, during, and after training at the BRCs and through follow-up with the client after discharge. Included in this profile is information contained in the RoboCaneTM software package.
RESULTS--The profile was successfully developed in the first year of the project. However, due to clinical staff cutbacks at the BRCS, gathering of information on the full mobility client profile has proved to be very problematic. In order to facilitate data gathering, the investigators have expanded the scope of the project to include data gathering and analysis of national Blind Rehabilitation Service information on all programs, not simply mobility. By doing this, we have broadened the scope of our data gathering to include mobility in the context of the full rehabilitation experience. We have also expanded our sample size to include not simply the clients of the original five BRCs. We are now recording and have access data on clients seen at all of the VA BRCs plus all of the clients seen by the Blind Rehabilitation Outpatient Service (BROS) and Visual Impairment Service Team (VIST) programs. All data gathering procedures will be in place by October 1996.
IMPLICATIONS--The mobility profile not only includes measurable individual variables and environmental characteristics, but also can be used to generate demographic norms for visually impaired individuals in different environments. This data set will be invaluable as a universal learning tool for mobility students, practicing mobility specialists, patients, and their families. These objective measures provide documentation for increased accountability, which results in improved quality assurance, increased capacity for the training and monitoring of professional staff, and diminished threat of liability. In addition, the use of the RoboCane software modeling package provides a unique opportunity to analyze the techniques that visually impaired individuals may have adapted that may be superior and safer than the traditional techniques currently being taught.
[289] A STUDY OF ILLUMINATION SOURCES FOR LOW VISION INDIVIDUALS
Gale R. Watson, MAED
Rehabilitation Research and Development Center, Atlanta
VA Medical Center, Decatur, GA 30033; email:
gwi@delphi.com
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C815-2RA)
PURPOSE--The objective of this 2-year project is to determine whether specific recommendations for appropriate illumination sources and levels of illumination can be derived for low vision individuals when reading. This project proposes to study illumination sources in order to provide low vision professionals an objective decision-making rationale for lighting recommendations based on type of visual impairment, type of lighting, and amount of light.
The research is designed to answer four questions about visual performance of individuals with low vision performing near tasks:
METHODOLOGY--Eighty subjects with low vision are currently being recruited in the following categories: 20 with macular loss, 20 with restricted fields, 20 with cataract or other media opacity, and 20 with general amblyopia with no field loss. Each subject who agrees to participate is tested without low vision devices for acuities, field of view, and contrast sensitivity.
The participants are then evaluated by reading a shortened version of the Pepper Visual Skills for Reading Test with a 5× spectacle magnifier under each of the following lighting sources: an incandescent lighting in a flex arm lamp, fluorescent lighting in a flex arm lamp, halogen lighting in a flex arm lamp, sodium vapor lighting in a flex arm lamp, full spectrum lighting in a flex arm lamp, a krypton spectacle mounted lighting device, an electroluminescent lighting device (developed by researchers at this Center), and a 5× Eschenbach illuminated hand-held magnifier.
After reading the Pepper Test under each lighting condition, participants then complete a visual fatigue diary to indicate a variety of physiological reactions to the type of lighting and to indicate their subjective preference for each lighting device. The data will be analyzed to ascertain whether individuals with some pathologies require certain types of lighting for best performance, and to determine whether any relationship exists between specific pathologies and illumination preference.
PROGRESS--To date we have tested 38 participants: 18 in the macular loss category, 10 in the restricted fields category, 8 in the cataract category, and 2 in the amblyopic category. Current recruitment is ongoing for additional VA and non-VA participants.
RESULTS--Using data from the macular loss category, we ran means and found the following trends. For accuracy, the lights were ranked in order as follows: Krypton, Fluorescent, Incandescent, Halogen, Sodium Vapor, ELVIS, Hand-held Magnifier, and Full-Spectrum. The lights were ranked for reading-rate as follows: Krypton, Fluorescent, Incandescent, Halogen, Sodium Vapor, ELVIS, Hand-held Magnifier, and Full-Spectrum. We found that the self-reported preferences were ranked as follows: Halogen, Full-Spectrum, Incandescent, Hand-held Magnifier, Sodium Vapor, Fluorescent, Krypton, and ELVIS. This information is just preliminary, and has no statistical significance at this point, but this does indicate certain trends in the data.
[290] MEASURING LOW VISION READING ASSESSMENTS USING A SCANNING LASER OPHTHALMOSCOPE
Gale R. Watson, MAEd; Ronald Schuchard, PhD
Rehabilitation Research and Development Center, VA
Medical Center, Decatur, GA 30033; University of
Missouri-Kansas City School of Medicine Department of
Ophthalmology Kansas City, MO 64108; email: gw1@delphi.com;
rschuchard@vaxl.umkc.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C849-2RA)
PURPOSE--The objective of our research is to evaluate the efficiency and efficacy of available low vision reading rehabilitation assessment tools using the scanning laser ophthalmoscope (SLO). We seek to discover the answers to the following: 1) Do rate, accuracy, and error scores on a visual skills for reading test relate to preferred retinal locus (PRL) ability and characteristics as measured by the SLO? 2) Do reading training techniques efficiently and effectively assist readers with macular loss in improving their visual skills for reading? 3) Can the effectiveness of reading training be measured by the improvement in visual skills such as PRL ability/characteristics and/or improvements in reading assessment scores? 4) Are improvements in reading performance related to improvements in different PRL abilities/characteristics or improvements in reading assessment scores which might allow the prediction of the type of reading training that is most appropriate for individual low vision patients?
METHODOLOGY--Software for the SLO will be developed for low vision reading rehabilitation and assessment, so that the Pepper VSRT, the Morgan LVRCA, and the MNRead reading tests can be presented in the SLO. This capability will allow the investigators to see the visual stimuli (letters and words) on the retina that the patients see when looking into the instrument, and thus determine the interaction between the macular scotoma and PRL. Eighty subjects with macular degeneration and an interest in reading with low vision devices will be recruited from the Eye Clinic of the Atlanta VA Medical Center. Before rehabilitation intervention, subjects will be assessed with standard clinical testing instruments (e.g., acuity, contrast sensitivity, and glare) and specialized clinical testing, (e.g., SLO macular perimetry, binocular retinal correspondence, and preferred retinal locus testing including fixation stability, pursuit ability, saccade ability, and word retinal location). Reading performance will be assessed by the Pepper VSRT, the Morgan LVRCA, and the MNRead Continuous Text Test. These reading tests will be performed both in their standard method of presentation and in the SLO (using the software developed by this project). Reading rehabilitation will then be provided at the Atlanta VA Medical Center using normal accepted training strategies that assist the subjects in maximizing their visual skills for reading. After the reading rehabilitation is complete, the standard clinical testing, specialized clinical testing, and reading performance assessment will be repeated.
RESULTS--Programming for the input of the reading tests on the SLO has been completed. The MNRead test, the Morgan Low Vision Reading Comprehension Test, and the Pepper Visual Skills for Reading Tests can now be administered via SLO.
FUTURE PLANS/IMPLICATIONS--Subject testing will begin when the software has been installed and debugged. Following the methods described above will provide this research project with a database of reading performance information as it relates to visual function characteristics of patients. We will be able to determine whether any of the visual function characteristics have an influence on reading performance, training strategies or outcomes. For example, the error patterns within the Pepper test can be related to macular scotoma and PRL characteristics. This database of reading performance information will also allow the investigators to propose detailed clinical procedures for the optimal training strategies and outcome assessment tools for each individual low vision patient's reading capabilities. The final results of this project will lead to a better understanding of the reading skills and abilities of individuals with macular loss.
[291] DESIGN AND EVALUATION OF LIQUID CRYSTAL (LC) DARK-ADAPTING EYEGLASSES FOR PERSONS WITH LOW VISION
David A. Ross, MSEE, MEd; Gary L Mancil, OD
Atlanta VA Medical Center, Rehabilitation R&D (151-R),
Decatur, Georgia 30033; Salisbury VA Medical Center,
Surgical Service (112), Salisbury, NC 28144
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C776-RA)
PURPOSE--The purpose of this 3-year project is to develop and evaluate liquid crystal (LC) light/dark-adapting eyewear for individuals with low vision. Many persons with low vision function best under restricted lighting conditions. The purpose of this project is first to develop LC sunglasses that very quickly and precisely control the amount of illumination reaching the eyes, and second, to test the usefulness and practicality of these LC sunglasses in actual use by persons with low vision.
PROGRESS--Investigators have completed an evaluation of LC materials, constructed an initial prototype based on the evaluation, completed subject testing of the initial design prototype, revised the LC sunwear design based on the results of the initial subject tests, re-evaluated LC materials, constructed improved prototypes, established indoor testing protocols simulating outdoor mobility tests, and are currently testing subjects. The project should be completed by fall 1996.
[292] LOW VISION ENHANCEMENT SYSTEM (LVES)
Rex A. Ballinger, OD; Peter A. Lalle, OD; Joseph H.
Maino, OD; Joan Stelmack, OD; Kristen Tallman, OD; Richard
Wacker, OD; Brendal Waiss, OD
VA Medical Center, Baltimore, MD 21201; VA Medical
Center, Kansas City, MO 64128; VA Medical Center, Hines, IL
60141; VA Medical Center, West Haven, CT 06516; VA Medical
Center, Palo Alto, CA 94304; VA Medical Center, Northport,
NY 11768
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C638-2DA)
PURPOSE--The purpose of this research is to evaluate the effectiveness of the Low Vision Enhancement System (LVES) device on the visual rehabilitation and improvement of quality of life in low vision patients. The goal is to improve the ability of the patient to function in daily living tasks and reduce dependence on multiple low vision aids.
METHODOLOGY--Subjects with subnormal or low vision (including those who are legally blind) were fitted and dispensed manual Beta-1 or autofocus Beta-2 LVES units. The device consists of a head-mounted series of video cameras, video displays, and integrated optical elements. The headset is connected to and powered by a cable connected to a self-contained battery control box which is mounted as a beltpack. The subject adjusts the controls on the beltpack as necessary for optimal viewing of the image. Subjects were evaluated before and after fitting and dispensing with regard to visual acuity and contrast sensitivity.
PROGRESS--We have entered 117 patients into the study thus far from two VA blind rehabilitation centers, 3 VICTORS (Visual Impairment Center To Optimize Remaining Sight) centers and one outpatient clinic. Recruitment is ongoing.
PRELIMINARY RESULTS--Of the 117 patients in the study, 63 percent suffered from macular degeneration, 11 percent from diabetic retinopathy, and 26 percent from various other diseases (including glaucoma, histoplasmosis, and Stargardts disease) Of the patients fitted with the LVES, 96 percent (112) had improvement in visual acuity over their best spectacle correction. (Median visual acuity with spectacle correction was 20/200 OU and median acuity using the LVES was 20/40 OU. Mode visual acuity was 20/200 and 20/20 respectively). Further, 93 patients were evaluated for change in contrast sensitivity using the LVES. Of these patients 78 percent demonstrated improved contrast sensitivity of several log units with the LVES.
FUTURE PLANS--Recruitment will continue up until the end of the study. Further testing is underway to evaluate the impact of the LVES on reading speed and time, visual-motor tasks, and activities of daily living. New developments are underway to improve the autofocus response times, and to assess the impact of the device on the visual field.
[293] DEVELOPMENT OF SCANNING LASER OPTHALMOSCOPE FOR LOW VISION REHABILITATION
Joseph H. Maino, OD
VA Medical Center, Kansas City, MO 64128
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Project #C838-RA)
No report was received for this issue.
[294] IDENTIFICATION OF SKILLS AND KNOWLEDGE NECESSARY FOR PEOPLE WITH VISUAL IMPAIRMENTS BEGINNING JOBS AFTER GRADUATING FROM POSTSECONDARY INSTITUTIONS
Lynn W. McBroom, PhD
Rehabilitation Research and Training Center on Blindness
and Low Vision, Mississippi State University, Mississippi
State, MS 39762; email: lwm3@ra.msstate.edu
Sponsor: National Institute on Disability and Rehabilitation Research, Washington, DC 22202; Mississippi State University, Mississippi State, MS 39762
PURPOSE--The purpose of this project is to identify skills and knowledge necessary for people with visual impairments to successfully make the transition from college to the workplace. This project builds on work completed in a previous project on transition from high school to college.
METHODOLOGY--Using contacts with employees, state directors of vocational rehabilitation programs, and college administrators, telephone interviews were completed with 45 employees with visual impairments and 27 of their employers. Questionnaire responses were analyzed using frequency analysis, correlation analysis, and factor analysis. The format of the surveys also encouraged respondents to provide lengthy answers to certain questions. Responses to these open-ended questions were analyzed and categorized.
RESULTS--The project is complete. Employees stated that the following were the most important in obtaining a job: making a career choice, developing a resume, locating transportation, communicating with others about the type of job desired, communicating with employers about accommodation needs, practicing being interviewed, visiting job sites, and making housing arrangements. The most common problems experienced by employees at work included the following: having enough money; locating transportation; being discriminated against because of the visual impairment; assessing books, written materials, diagrams, and charts; being lonely; and managing time.
The most frequently provided work accommodations by employers were travel instruction or orientation to the worksite, computer access devices or CCTVs, and alternative print media. The majority of employers reassigned some job duties, provided readers or transportation, allowed flex-time scheduling, and removed architectural barriers. Employers advised potential workers to not expect special considerations for impairments, to obtain skills needed to perform the job, to emphasize abilities and assets during job interviews, to select jobs that match abilities and skills, to discuss accommodations and provide information about obtaining accommodations, and to educate employers and coworkers about the impairment.
FUTURE PLANS--Results from this study and a related study on transition from high school to college will be used to write a brochure identifying the skills, knowledge, and steps necessary for students with visual impairments to successfully make the transition from high school to college and from college to the workplace. These materials will be disseminated to high school students, counselors, and parents; and to college students, advisors, and rehabilitation personnel to help increase the number of students with visual impairments completing college and beginning work.
[295] TACTILE AND HAPTIC VISUALIZATION PROJECTS
Kenneth E. Barner, PhD; Richard Foulds, PhD; Thomas Way,
BS; Jason Fritz, BS
SEM Tactile and Haptic Interface Project, Applied
Science and Engineering Laboratories, University of
Delaware/A.I. duPont Institute, Wilmington, DE 19899; email:
sem-info@asel.udel.edu;
barner@asel.udel.edu
Sponsor: National Science Foundation, Arlington, VA 22230; Nemours Foundation, A.I. duPont Institute, Wilmington, DE 19899
PURPOSE--Individuals with visual disabilities face challenges that can be insurmountable in interfacing to the ever more prevalent graphical computer interface environments. Not surprisingly, these individuals are underrepresented in science, engineering, and math (SEM) academic programs and professions, which often rely on the latest computer technology. As a part of the SEM Project (Engaging, Recruiting, and Retaining Students with Disabilities in Science, Engineering, and Math), the Tactile and Haptic Interface Project addresses the needs of the person with visual impairment through basic human factors research and display system design and development. Our research investigates and quantifies viable methods for the rendering of traditionally visual information such as photographs, graphs of mathematical functions, physics and chemistry experiments, and other images in a tactual or haptic fashion which can subsequently be understood by the student. Development of a number of software and hardware systems is underway, and includes using devices for the tactile and haptic display of such visual information, and combining various existing technologies in new ways to achieve the goal of tactile accessibility.
METHODOLOGY--Understanding the human factors involved in the basic human ability to recognize physically represented visual information is the first goal of our research. Through measurement of factors, including the tactile resolution of the fingertip under various real world constraints, tactile image discernment tasks such as distinguishing various tactile shapes from one another, and abilities of blind and deaf-blind persons to comprehend tactile and haptic information, we will gain insight into how the mind perceives what the human tactile and haptic systems sense.
Concurrently with this basic research, and subsequently relying on its results, will be the development of a number of software and hardware systems to provide blind computer users access to visual information. Using combinations of products and devices, such as capsule paper, the Optacon, the PHANToM, screen reading software, touch screen technology, and others, together with the appropriate software, either commercial or custom written, we will create tactile and haptic access systems. While some of this research will involve the use of new and somewhat expensive technology, most of it will emphasize the use of affordable equipment and software to ensure that the end product will be as widely available as possible to help the greatest number of individuals with the need for it.
PROGRESS--The haptic visualization project has made a great deal of progress in the past year. Since the PHANToM haptic interface did not include a software library or programming interface, one was created using Microsoft's Visual C++ running under Windows NT. Several methods of haptic modeling of different types of scientific data have been created, as well as a graphical display on a Silicon Graphics machine for demonstration and debugging. Models for haptic enhancements, such as friction and texture, have also been created to help an unsighted person understand the data that he or she is feeling. Speech output of information is also possible. Several persons with visual impairment have had the opportunity to feel different haptic simulations, and the feedback has been positive.
Materials for conducting basic two-point and point-line tactile experiments have been produced using isotropically etched glass slides, and these will be supplemented with more complex shapes portrayed on capsule paper in experimentation into the human factors of the sense of touch. An image-processing system for simplifying and generating a tactile representation of photographs, drawings and other images is in the early stages of development.
FUTURE PLANS--Tactile studies will be conducted and will be extended to the haptic sense system. We will continue to develop the software interface for haptic visualization with the PHANToM, as well as conduct more human factors studies of the algorithms for haptic rendering of data. We will need to determine if an unsighted person can extract enough information about the rendered data using the developed algorithms. The image simplifcation system will be completed, as will systems that make use of touch screen technology and tactile overlays.
Go to TOP.
Last revised Wed 05/26/1999
