CONTINUOUS_ VERSUS PULSED PURE-TONE AUDIOMETRY IN A GROUP OF SCHOOL-AGED CHILDREN Catherine van Dijk* and Naeema Osman Department of Communication Pathology, University of Pretoria, South Africa

Abstract

Pure-tone testing is the primary audiological test procedure for the differential diagnosis of hearing loss and hearing disorders in school-aged children. No research is currently available internationally for children’s re- sponses to continuous versus pulsed pure-tones. The aim of this exploratory investigation was to compare the performance of a group of school-aged children on continuous versus pulsed pure-tone audiometry. The aims were to determine whether a threshold difference existed between continuous versus pulsed pure- tones and to record whether a listener preference existed between continuous versus pulsed tones for the frequency range of 125 to 8000 Hz. Eighteen children (36 ears) aged between 8-12 years, participated in a hearing evaluation as well as in a brief three-question interview. Descriptive statistics viz. average thresh- old, mean difference and standard deviation of thresholds were used to analyse data. Listeners’ perceived preferences were calculated in percentages and reasons for preferring one signal over another was analysed qualitatively. Although the automatically pulsed tone threshold (averaged across the frequencies tested) was lower than for the continuous tone, the difference was only 0.2 dB in the left ear and 0.5 dB in the right ear. This small difference is not important in clinical applications for which 5 dB increments are used in pure-tone audiometry. Where a listener preference was indicated, however, the continuous tones were preferred over pulsed tones by 56% percent of subjects. These findings differ from similar studies involving adults. This revealed that children may yield different preferences during pure-tone testing than adults and that these preferences should be taken into consideration during testing.

Keywords: audiometry, continuous pure-tone, listener preference, pulsed pure-tone, pure-tone testing, school-aged children, stimulus signal, threshold difference.

here are two methods of presentation of pure- fork tests (Martin & Clark, 2003). It is the primary tones, i.e. continuous tones and pulsed tones. audiologic test procedure for the differential diagno- Pure-tone threshold measurement is a fun- sis of hearing loss and hearing disorders in school- damental diagnostic tool in hearing evaluations (Di- T Correspondence to: Giovanni & Repka, 2007). Pure-tone tests of hear- ing, using audiometers,have been in use for nearly *Telephone number: +27 12 420 2357 one hundred years and are quantifiable electronic ex- Fax number: +27 12 420 3517 tensions of the same concepts developed in tuning Email: [email protected]

40 | DIE SUID-AFRIKAANSE TYDSKRIF VIR KOMMUNIKASIE-AFWYKINGS, VOL. 55, 2008 aged children (Blandy & Lutman, 2005). In addi- Hochberg & Waltzman, 1972; Hood, 1955; Mineau tion, pure-tone threshold audiometry has become the & Schlauch, 1997). Firstly, in comparison to contin- standard behavioural procedure for describing audi- uous tones, the use of pulsed tones has been shown to tory sensitivity (Margolis & Saly, 2007). In 1978, decrease the number of false-positives in both nor- the American Speech-Language-Hearing Associa- mal-hearing listeners and listeners with sensorineural tion (ASHA) developed clinical guidelines referred hearing loss with or without tinnitus (Dancer & to as Guidelines for Manual Pure-Tone Audiometry Conn, 1983; Dancer et al, 1976; Mineau & Schlauch, (ASHA, 1978). In these guidelines they recommend- 1997). Secondly, listeners with normal hearing and ed the use of a one_ to two second sustained tone those with sensorineural hearing loss and tinnitus for threshold determination. This is referred to as a preferred pulsed tones to continuous tones as a lis- continuous tone. According to an American. survey tening task (Gardner, 1947; Hochberg & Waltzman, of audiometric practices two decades ago, the major- 1972). The use of pulsed tones in Bekesy audiometry ity of audiologists surveyed, used a manual presen- has also provided insight into using pulsed tones in tation of continuous tones (Martin & Sides, 1985). conventional audiometry ( Jerger, 1960). In thresh- Although the ASHA guidelines (ASHA, 1978) rec- old applications of Bekesy audiometry, pulsed tones ommend use of a continuous tone, a substitution of are typically presented to minimize the effects of au- pulsed tones is also permitted by ASHA. A pulsed ditory adaptation to a continuous tone (Hallpike & tone can be presented either manually or can be au- Hood, 1951; Jerger, 1960; Reger, 1970; Silman & tomatically generated by the audiometer. A manu- Silverman, 1991; Sorensen, 1962). In addition, if a ally presented pulsed tone would require the tester to 200-milliseconds-on, 200-milliseconds-off pulsed present two consecutive tones with a short quiet in- tone is used for threshold assessment, periodic su- terval between the two tones. This interval should be prathreshold presentations of the test tone acts as a one second or shorter in duration (ASHA, 1978). cue for the signal frequency, thereby refreshing the Studies on children’s performance on and prefer- client’s memory for the test stimulus (Mineau & ence for continuous versus pulsed pure-tone audi- Schlauch, 1997). Furthermore, McCommons and ometry is lacking. A study conducted by Burk and Hodge (1969) found that the pulsed tones used in Wiley (2004) on women aged 19 years to 44 years Bekesy audiometry provides more feedback or cues showed that pulsed tones are clearly advantageous by which the listener can make a decision regarding in presenting what is perceived as an easier task. the presence of the signal. For many adult clients the pulsed tone is perceived The slight preference for pulsed tones in normal- as easier to hear at near threshold levels (Newby & hearing listeners (Burk & Wiley, 2004), coupled with Popelka, 1985). Other published data suggests that previous reports demonstrating the benefits of using using pulsed tones for adults having tinnitus may re- pulsed tones in threshold assessment for listeners sult in fewer false-positive responses (30% less than with sensorineural hearing loss and tinnitus (Hoch- for continuous tones) and therefore may reduce the berg & Waltzman, 1972; Mineau & Schlauch, 1997), number of presentations required to determine the supports the general use of pulsed tones in audiom- threshold, resulting in more efficient audiological etry. However, in practice it may seem that not all threshold evaluations (Mineau & Schlauch, 1997). audiologists use this technique. From personal ob- There are selected studies that support the use of servations, South African audiologists often deviate pulsed tones rather than continuous tones in manu- from standard testing procedures and protocols con- al threshold audiometry for adults (Dancer & Conn, veyed to them by their undergraduate training insti- 1983; Dancer, Ventry & Hill, 1976; Gardner, 1947; tutions. A diversity of testing techniques and meth- THE SOUTH AFRICAN JOURNAL OF COMMUNICATION DISORDERS, VOL. 55, 2008 | 41 Catherine van Dijk and Naeema Osman

ods are often applied by audiologists in practice and based ( Johnson, Benson & Seaton, 1997). In ad- may be due to, amongst others, their employer’s per- dition, school-aged children with a hearing loss ap- sonal preferred method of testing that has been im- pear to have increased rates of grade failures, need parted to them. The type of test signal (continuous for educational assistance, and perceived behavioural or pulsed) used in pure-tone audiometry may affect issues in the classroom ( Johnson, Benson & Seaton, the accuracy and reliability of pure-tone thresholds. 1997). An undetected hearing loss may have a cas- Some data regarding adults’ responses to continuous cading effect in terms of habilitation. Various effects and pulsed pure-tone audiometry is available (for ex- are insurmountable – such as economic status, edu- ample: Burk & Wiley, 2004). However, similar data is cation, psychosocial stigma and eventually potential currently unavailable for children’s responses to con- employment opportunities as an adult. Fortunately, tinuous versus pulsed tones. . Therefore, this study is there is a continued emphasis nationally and inter- necessary to explore if children may respond differ- nationally on early identification and monitoring of ently in their preference to adults. Indeed, this study hearing loss. Therefore, it is vitally important to ob- has great relevance given that threshold audiometry tain optimal thresholds for children as these results by air conduction is a widely applied procedure for have direct consequences on identification, diagnosis obtaining auditory thresholds in children, the valid- and management of a hearing loss. Early detection ity and variability of tonal stimuli presented has not followed by appropriate intervention maximises the yet been addressed. Indeed, the apparent simplicity benefits to the child, family, and society (Diefendorf, of conducting the test often masks the importance 2002; Pappas, 1998). Part of the intervention proc- of investigating specific procedures applied in pure- ess includes the appropriate fitting of hearing aids. tone audiometry. Effective pure-tone test procedures However, if optimal and reliable hearing thresholds are imperative for the timely and accurate diagnosis are not obtained, over_ or under amplification may of hearing loss in children. occur. Hearing loss depending on severity, age of onset, Although more objective tests have been devel- and a host of other factors has a mild-to-profound oped as part of the test battery to obtain auditory impact on communicative functioning (Yoshinaga- thresholds, one cannot move away from standard Itano, Sedey, Coulter & Mehl, 1998). The prevalence pure-tone audiometry (Hall, 2007). Electrophysi- of hearing loss in children is such that its diagno- ological tests like AEP (auditory evoked potentials) sis and management necessitates high prioritisation. and ECochG (electrocochleography) are used to es- It is estimated that in 2002 approximately 72 000 timate hearing thresholds. However, these test pro- school-aged children in the United States of Amer- cedures often require sedation, may have artefact ica (U.S.A.) alone fell in the “hearing impairment contamination, and are not suitable for all types of category” (United States Office of Special Educa- losses, e.g. conductive hearing losses or mild senso- tion Programs, 2002). In South Africa, a developing ri-neural hearing losses (Hall, 2007). Electrophysi- context, prevalence of hearing loss can be expected ological tests cannot substitute the use of pure-tone to be similar or even higher than in the U.S, but to audiometry where reliable behavioural responses can date no statistical data is available for this popula- be obtained to determine frequency specific infor- tion. School-aged children with undetected hear- mation. ing loss that have not received early intervention, Data regarding the pure-tone testing of school- often experience academic delays. As would be ex- aged children is of significant importance as this pected, the primary deficit areas for children with population is mostly evaluated with pure-tone au- hearing loss are in those subjects that are language- diometry. School-aged children often pose unique 42 | DIE SUID-AFRIKAANSE TYDSKRIF VIR KOMMUNIKASIE-AFWYKINGS, VOL. 55, 2008 CONTINUOUS VERSUS PULSED PURE-TONE AUDIOMETRY IN A GROUP OF SCHOOL-AGED CHILDREN

challenges to the audiologist during pure-tone test- existed between continuous versus pulsed pure-tone ing ( Johnson, et al, 1997). These challenges include audiometry for the frequency range of 125 Hz to assessment of children with developmental delays in, 8000 Hz; and for example, receptive language, which reduces their • to determine listener preference for continu- comprehension and following of test instructions re- ous versus pulsed tones for the frequency range of sulting in troublesome test behaviour. Some children 125 Hz to 8000 Hz. will not tolerate earphones or ear inserts for long pe- METHOD riods of time as this makes them uncomfortable. In Design comparison to adults, children have a limited atten- This exploratory study was mainly quantitative in tion span and cannot remain still or concentrate for nature consisting of pure-tone audiometric testing long periods during testing. This may cause them to and a brief, three-question interview. A compara- become restless and/or uncooperative. False-positive tive based study employing a quasi-experimental de- responses are prevalent in this population and failure sign of research was used as not all variables could be to respond to stimuli close to their hearing thresholds controlled for (Struwig & Stead, 2001). The main are quite common. In addition, children’s reliability aim of this study was to compare continuous_ and of responses tends to decrease as the time of testing pulsed pure-tone audiometry in a group of school- increases ( Johnson et al, 1997). Therefore, threshold aged children. data comparisons for continuous tones and pulsed Subjects tones in school-aged children are justified because Ethical issues it may be, for example, that one stimulus tone is Ethical clearance was obtained from the depart- perceived as an easier task to respond to, resulting mental research committee of the University of Pre- in a shorter testing time. A shorter testing time toria before the investigation was carried out. Writ- may address assessment challenges such as reduced ten informed consent and assent was obtained from attention span and may increase the reliability of the subjects and their guardians. Subjects and their responses. Comparisons of continuous and pulsed guardians were informed that their participation was pure-tone thresholds may yield findings that indicate voluntary and that they could withdraw without con- better thresholds obtained for one of these methods. sequence at any stage during the study. All identify- These findings may guide audiologists when deciding ing information was kept confidential and subjects which stimulus tone to use when testing children were not subjected to harmful or unjust procedures. with pure-tone stimuli. Subject selection criteria A review of available professional literature has • Subjects were required to respond reliably clearly revealed a lack of data enable comparisons of and had to concentrate for long periods (40 minutes continuous tones and pulsed tones in children using or longer). For this reason, children between the the manual pure-tone technique (as recommended by ages of 8 years to 12 years were selected. It was felt ASHA, 1978). This exploratory investigation aimed that children in this age group were more able than to compare threshold data and listener preference younger children to perform these tasks consistently. for continuous and pulsed tones on a small group • Subjects had to present with normal hear- of children. Therefore, the main aim of this study ing as the presence of a hearing loss may influence was to compare continuous and pulsed pure-tone the ability to distinguish between the continuous audiometry in a group of school-aged children. The and pulsed pure-tone test signal. Furthermore, most sub-aims of the study were: researchers are in agreement that when investigat- • to determine whether a threshold difference ing the benefits of any audiological test procedure, THE SOUTH AFRICAN JOURNAL OF COMMUNICATION DISORDERS, VOL. 55, 2008 | 43 Catherine van Dijk and Naeema Osman

normative data should be recorded before similar in- riodically in accordance with American National vestigations can be conducted on hearing-impaired Standards Institute (ANSI) 1996 specifications and populations (Hall, 2007). listening checks of the equipment were performed • In order to ascertain hearing status, all sub- prior to testing. Results were recorded on a standard jects were subjected to a test battery prior to the audiogram. study. The test battery included otoscopy, tympan- Apparatus and material for main study ometry, reflex testing, oto-acoustic emission screen- Pulsed and continuous hearing thresholds were ing and comprehensive pure-tone bone_ and air con- obtained using the same sound-insulated booth and duction testing. diagnostic audiometer employed for the pre-selec- Sampling tion of subjects. An adaptation of a standard audio- Subjects were purposefully selected by requesting gram was used to record the results. The symbol that final year students to supply contact details of fam- was used for both ears to indicate continuous hearing ily members or acquaintances that had children be- threshold level was an outline of a circle and a solid tween the ages of 8 years and 12 years. The subjects circle indicated pulsed hearing thresholds. A brief were expected to visit the Hearing Clinic at the Uni- three-question interview schedule was used after versity of Pretoria and therefore only children from completion of the hearing test to obtain information the Pretoria region were targeted. about listener preference for the type of test signal. Description of participants Procedures Eighteen children (36 ears) with normal hearing Pure-tones were presented through earphones to thresholds for all seven tested frequencies were in- assess air conduction thresholds. The results of the cluded. According to Katz (2002) and Martin and air conduction tests in both test conditions (continu- Clark (2003) a range of 0 dB to 20 dB can be re- ous and pulsed) were plotted on an audiogram from garded as normal hearing for school-aged children. which the results were compared. The ascending/de- Eight of the subjects were male and ten were female. scending method developed by Carhart and Jerger The age distributions were as follows: three were (1959) was used to obtain pure-tone thresholds bi- eight years old, one was nine years old, four were laterally at 125 Hz, 250 Hz, 500 Hz, 1000 Hz, 2000 ten years old, six were eleven years old and four were Hz, 4000 Hz, and 8000 Hz. Octave frequencies of twelve years old. 3000 Hz and 6000 Hz were excluded to reduce test- Apparatus and Materials ing time. However, these octaves were included in Apparatus and material for pre-selection the initial selection procedures to determine normal Otoscopic examinations were conducted us- criteria. Steps of 5 dB were used to confirm hearing ing a Welch-Allyn otoscope. Acoustic immittance threshold levels. Subjects took part in one testing measures were obtained using a calibrated (accord- session with two alternating signal conditions. The ing to SABS 0154-1/2, 0182) tympanometer (GSI signal conditions were a manually presented contin- Tympstar). A Scout Oto-Acoustic Emission ma- uous tone for one second to two seconds and two chine with 2 kHz to 6 kHz (4 dB to 6 dB for a pass) short pulses (rise-fall time of 35 ms with a duration screening protocol was used to perform oto-acoustic of 200 ms onset to offset) automatically generated by emission screening. Pure-tone air and bone conduc- the audiometer. The test signals were presented in an tion thresholds were obtained using a twin channel alternating manner to ensure that the same test sig- diagnostic audiometer, the GSI 61. Subjects were nal (e.g. a pulsed pure tone) was not always presented seated in a sound-insulated booth (Industrial Acous- in the same order to the subject. For example, the tics Company, Inc.). Calibration is conducted pe- subject would hear, at 125 Hz, first a pulsed_ then a 44 | DIE SUID-AFRIKAANSE TYDSKRIF VIR KOMMUNIKASIE-AFWYKINGS, VOL. 55, 2008 CONTINUOUS VERSUS PULSED PURE-TONE AUDIOMETRY IN A GROUP OF SCHOOL-AGED CHILDREN

continuous tone. At 250 Hz, he/she would first hear diometry a continuous then a pulsed tone. This would random- The first sub-aim of the study was to determine ly vary from subject to subject. This presentation en- whether a threshold difference exists between con- sured reliability and validity of results and yielded tinuous versus pulsed pure-tone audiometry in two sets of thresholds (continuous and pulsed tones) school-aged children. These results included hear- for each ear. The first signal (either automatically ing threshold levels obtained for continuous as well pulsed or a continuous tone) was randomly chosen as pulsed tones at frequencies 125 Hz to 8000 Hz in and followed alternately by the other signal. After each ear of each participant. The difference between the hearing test, a brief three-question interview fol- the average of the continuous hearing thresholds and lowed the pure-tone testing. Subjects were asked if the average of the pulsed hearing thresholds for each they preferred one test signal above the other, and, ear and specific frequencies were determined. These if so, which signal they preferred and why. Their an- differences were calculated by subtracting the av- swers were recorded by the researcher on a space pro- eraged pulsed tones from the averaged continuous vided on the audiogram. Testing and recording was tones and are depicted in Figure 1. done by a final year audiology student. A qualified Figures (1 to 5) audiologist supervised the student during testing of 2.0 the first three subjects to ensure that test procedures 1.5 1.0 were valid and reliable. The supervisor was available 0.5 for consultation for the duration of the testing. 0.0 Data analysis -0.5

125Hz 250Hz 500Hz Data was organised and processed using “Micro- 1000Hz 2000Hz 4000Hz 8000Hz -1.0

soft Office Excel” software. A univariate procedure averaged pulsed thresholds -1.5 was used in order to draw comparisons between con- Left Ear -2.0

Difference between averaged continuous and Right Ear tinuous and pulsed threshold values at discreet fre- -2.5 quencies. The average threshold, mean difference Frequencies and standard deviation of the two thresholds ob- Figure 1. Difference between average hearing threshold levels of continuous tained with the two different types of stimuli at each and pulsed tones in each ear (n=36) frequency tested for each ear, was calculated. Listen- Figure 1. Difference between average hearing ers’ preferences indicated in the interview were cal- threshold levels of continuous and pulsed tones in each culated in percentages to obtain the most preferred ear (n=18) signal. Reasons for preferring one signal to anoth- er was analysed qualitatively. During data analysis A difference value of zero, which was obtained at the results for each ear were not combined as litera- 250 Hz in the left ear and at 1000 Hz in the right ear, ture suggests that ear differences occur (Hochberg indicated no threshold difference between the two & Waltzman, 1972). The afore-mentioned statistical stimulus types as presented in Figure 1. Negative procedures used in the study were deemed appropri- values indicated that lower thresholds were obtained ate for the small sample size and provided sufficient for continuous tones. Continuous tones yielded bet- data in order to answer the research question com- ter thresholds than pulsed tones at 500 Hz in the left prehensively. ear and 2000 Hz in the right ear. Lower thresholds RESULTS were obtained for pulsed tones at all other frequen- Determining whether a threshold difference ex- cies for both ears. Therefore, pulsed tones yielded ists between continuous versus pulsed pure-tone au- lower thresholds than continuous tones for the ma- THE SOUTH AFRICAN JOURNAL OF COMMUNICATION DISORDERS, VOL. 55, 2008 | 45 Catherine van Dijk and Naeema Osman

jority of frequencies tested. the right ear, follows in Figure 3.

In order to determine whether the signal 12 StDev (Continuous) type yielded a threshold difference at specific fre- StDev (Pulsed) 10 Average (Continuous) quencies, findings per frequency for the respective Average (Pulsed)

ears were analysed. The standard deviation and the 8 mean average of the hearing threshold levels were 6 obtained at each frequency tested, for each stimulus type, for the left ear, follows in Figure 2. 4 12 StDev (Continuous)

continuous and pulsed thresholds pulsed and continuous 2 StDev (Pulsed) 10 Average (Continuous) Mean average and standard deviations of Average (Pulsed) 0 125Hz 250Hz 500Hz 1000Hz 2000Hz 4000Hz 8000Hz 8 Frequencies

6 Figure 3. Mean averages and standard deviation for continuous and pulsed tones in the right ear (n=36). 4 Insert here: Figure 3. Mean averages and standard deviation for continuous and pulsed tones in the right 2 continuous and pulsed thresholds pulsed and continuous ear (n=18). Mean average and standard deviations of 0 125Hz 250Hz 500Hz 1000Hz 2000Hz 4000Hz 8000Hz Frequencies The average of the standard deviations for contin-

Figure 2. Mean average and standard deviation for continuous and pulsed uous tones was 7.3 dB compared to 7.5 dB for pulsed tonesFigure in the left ear2. (n=36)Mean average and standard deviation for tones in the right ear. Continuous tones presented continuous and pulsed tones in the left ear (n=18) less variation in all thresholds except at 1000 Hz and 4000 Hz in the right ear. Absolute differences in av- The average standard deviation for continuous eraged thresholds at each test frequency ranged from tones was 8 dB in the left ear. The average standard 0 dB to 1.4 dB between 125 Hz and 8000 Hz. Aver- deviation for pulsed tones was 7.9 dB in the left ear. aged pulsed tones yielded slightly lower thresholds at Continuous tones presented less variation in thresh- all frequencies except at 1000 Hz where thresholds olds at 125 Hz, 500 Hz, 4000 Hz and 8000 Hz in were equal and 2000 Hz where the continuous tones the left ear while pulsed tones presented less varia- obtained slightly lower thresholds. tion in thresholds at 250 Hz, 1000 Hz and 2000 Hz. When results of both ears are considered, thresh- Absolute differences in averaged thresholds at each old standard deviations (averaged over all frequen- test frequency ranged from 0 dB to 1.9 dB between cies tested) showed that the pulsed presentation 125 Hz and 8000 Hz. Averaged pulsed tones yielded method yielded slightly less variation (0.005 dB) in slightly lower thresholds at all frequencies except at thresholds, than the continuous presentation meth- 250 Hz where thresholds were equal and at 500 Hz od. On average, pulsed tones yielded slightly lower where the continuous tones revealed slightly lower thresholds where the average pulsed tone threshold thresholds. was 0.4 dB lower than for the averaged continuous Similar to the left ear, findings per frequency were tone stimulus. These small differences in averaged analysed for the right ear to determine whether the thresholds and standard deviations obtained for both signal type yielded a threshold difference at specific test tones reinforce the equality of the two stimu- frequencies. The standard deviation and the mean lus signals. Based on the aforementioned, the over- average of the hearing threshold levels were obtained all differences would not be sufficient to recommend at each frequency tested, for each stimulus type, for one signal above the other. 46 | DIE SUID-AFRIKAANSE TYDSKRIF VIR KOMMUNIKASIE-AFWYKINGS, VOL. 55, 2008 CONTINUOUS VERSUS PULSED PURE-TONE AUDIOMETRY IN A GROUP OF SCHOOL-AGED CHILDREN

Findings for both ears at all frequencies were differs from the study done on adults where lower plotted to compare overall threshold differences due thresholds (averaged across all frequencies) were ob- to different signals. These results are presented in tained for continuous tones than for automatically Figure 4. pulsed tones, where the difference was 1.0 dB (Burk & Wiley, 2004). The obtained minor 0.3 dB differ- 11 Continuous Left Ear 10 Pulsed Left Ear ence would not result in a substantial difference in Continuous Right Ear 9 Pulsed Right Ear hearing threshold level, since these measurements 8 are made in 5 dB increments. 7

6 Determining whether a listener preference is per- 5 ceived between continuous versus pulsed tones

Mean averages 4 The second aim of the study was to determine 3

2 whether subjects had a listener’s preference for one 1 of the two signals. The percentage of subjects pre- 0 ferring one signal over the other based on the three 125Hz 250Hz 500Hz 1000Hz 2000Hz 4000Hz 8000Hz Frequencies questions asked, was calculated. These results are Figure 4. A comparison of the average hearing threshold levels obtained for represented Figure 5. continuous and pulsed tones in each ear (n=36) (note that a lower bar graph Figure 4. A comparison of the average hearing Continuous indicates a better hearing threshold) thresholds obtained for continuous and pulsed tones in (Beep was easier) Pulsed each ear (n=18). (Beep-beep was easier)

As may be noted in Figure 4, the differences in 8 thresholds between the two stimuli were relatively (44%) 10 small. This correlates with results obtained by Hoch- (56%) berg and Waltzman (1972), which indicated that both types of signals obtained comparable threshold levels at all audiometric frequencies tested, and did not vary by more than 1.6 dB. Across all frequencies Listeners’ preference for one stimulus type over the other (n=36) tested in the left ear, an average of 3.8 dB for contin- Figure 5. uous tones and 3.6 dB for pulsed tones was obtained. Similarly, the average hearing thresholds obtained in Figure 5. Listeners’ preference for one stimulus type the right ear were 5.7 dB for continuous tones and over the other (n=18). 5.2 dB for pulsed tones. Although the automatically pulsed tone threshold (averaged across the frequen- The difference in the percentage of perceived cies tested) was lower than for the continuous tone, preference was 11%, with 56% preferring continu- the average difference was only 0.2 dB in the left ear ous tones and 44% preferring pulsed tones. This dif- and 0.5 dB in the right ear. This small difference is fers from the findings of a study (Burk & Wiley, not important in clinical applications for which 5 dB 2004) done on 24 adults, where it was found that increments are used in pure-tone audiometry. 62% of subjects preferred the pulsed tones and 38% Taken as a whole, the pulsed tone stimuli produced preferred the continuous tones. Over two-thirds slightly lower thresholds at all the test frequencies of normal and tinnitus subjects, in a study done by where the average pulsed tone threshold was 0.3 dB Hochberg and Waltzman (1972), preferred to listen lower than for the continuous tone stimulus. This to pulsed tones as opposed to continuous tones dur- THE SOUTH AFRICAN JOURNAL OF COMMUNICATION DISORDERS, VOL. 55, 2008 | 47 Catherine van Dijk and Naeema Osman

ing threshold determination. The difference in the rately for each ear at the various test frequencies. sample population group in the present study, which In summary, the findings obtained for the aver- was done with children, may contribute to the dis- aged hearing threshold differences at each frequen- parity in the findings. cy when comparing continuous to pulsed hearing Results indicated that more subjects preferred the threshold levels were: continuous tone, although, the post-test comments • Pulsed tones yielded slightly lower thresh- across the signal types were very similar. This differs olds in both ears at 125 Hz. from previous investigations where persons experi- • At 250 Hz, no difference in hearing thresh- encing tinnitus commented that continuous tones old levels were obtained in the left ear and pulsed are more confusing than pulsed tones (Mineau & tones yielded slightly lower thresholds in the right Schlauch, 1997). Findings revealed that the most ear. common explanation given by those subjects prefer- • At 500 Hz, continuous tones yielded slight- ring the continuous tones was that it was louder and ly lower thresholds in the left ear and pulsed tones clearer and that it did not have interruptions. The yielded slightly lower thresholds in the right ear. most common reason given by those subjects pre- • At 1000 Hz, pulsed tones yielded slightly ferring the pulsed tones was that one could hear it lower thresholds in the left ear and no difference in a second time if one was unsure of hearing the first hearing threshold levels were obtained in the right presentation. ear. A direct comparison provided further findings re- • At 2000 Hz, pulsed tones yielded slightly garding the correlation between the perceived listen- lower thresholds in the left ear and continuous tones er’s preference and the stimulus signal that produced yielded slightly lower thresholds in the right ear. the lower hearing threshold level. Seven subjects ob- • Pulsed tones yielded slightly lower thresh- tained lower hearing thresholds in both ears for the olds in both ears at 4000 Hz. stimulus signal that was not their perceived prefer- • Pulsed tones yielded slightly lower thresh- ence. A further seven subjects perceived preference olds in both ears at 8000 Hz.§ correlated with a lower hearing threshold in one The above-mentioned findings revealed that the ear only. Only four subjects obtained better hear- pulsed tone stimuli produced slightly lower thresh- ing thresholds in both ears for the stimulus signal olds (0.3 dB) on average at all the test frequencies that they perceived as their preference. Therefore, it than for the continuous tone stimulus. Literature would seem that listeners’ perceived preference did does not provide an explanation for why audiomet- not correlate with a lower hearing threshold for the ric pulsed tones yield slightly lower thresholds, ex- preferred stimulus signal. cept that subjects tend to report that a pulsed tone DISCUSSION AND CONCLUSION seems to be more alerting than a continuous tone The aim of this study was to compare continuous (McCommons & Hodge, 1969). Inter-aural differ- and pulsed pure-tone audiometry in a group of nor- ences were apparent at some test frequencies in this mal-hearing school-aged children. This was done by investigation and are also documented (Hochberg & comparing hearing threshold levels obtained for both Waltzman, 1972). The ear that produced the- bet stimulus signals and determining if a listener’s pref- ter pulsed tone thresholds varied within the same erence existed for one tone over the other. The value subject, which makes the theory of ear domination of this study, was that data, unlike previous studies invalid for these cases. Current literature does not of the same nature (Burk & Wiley, 2004; Mineau aid in providing an explanation nor are the results & Schlauch, 1997), was recorded and analysed sepa- of this study useful to explain why these inter-aural 48 | DIE SUID-AFRIKAANSE TYDSKRIF VIR KOMMUNIKASIE-AFWYKINGS, VOL. 55, 2008 CONTINUOUS VERSUS PULSED PURE-TONE AUDIOMETRY IN A GROUP OF SCHOOL-AGED CHILDREN

differences are found for continuous and pulsed tone duration of the test procedure. She indicated that, thresholds. However, differences are statistically so although some subjects showed obvious signs of small that it does not warrant speculation as to why boredom, this resulted in only a few inconsistent re- they occur. sponses and not more than one would expect dur- It may be concluded that the use of pulsed tones ing behavioural testing. Despite these shortcomings, for audiometric threshold measures in normal-hear- limited research is available to date that compares ing children had no clinically significant effect on outcomes of different test signals in pure-tone au- obtained hearing thresholds as 5 dB increments are diometry in children and this exploratory study pro- used in pure-tone audiometric testing. This differs vides preliminary data with some clinical considera- from previous findings demonstrating the benefits of tions for audiologists in practice. using pulsed tones in threshold assessment for adults CLINICAL IMPLICATIONS with sensorineural hearing loss and persons with tin- The benefits of using pulsed tones over continu- nitus (Burk & Wiley, 2004; Hochberg & Waltzman, ous tones have been documented with adults with 1972; Mineau & Schlauch, 1997). These prelimi- hearing-loss as well as with adults suffering from nary findings may indicate that school-aged children tinnitus. However, in this exploratory study it would yield different responses during pure-tone testing seem that using either continuous or pulsed pure- than adults. tone signals during testing with normal-hearing However, continuous tones showed a slight ad- children does not yield significant threshold differ- vantage in terms of presenting what is perceived as ences. Conversely, audiologists might consider ask- an easier task without adversely affecting the test’s ing their young clients during testing which signal outcome. A slight difference was obtained, with more they preferred. As revealed in this study, this may subjects preferring continuous tones than pulsed not result in obtaining better threshold levels. Nev- tones. The slight preference for continuous tones in ertheless, taking their preference into consideration normal-hearing children also differs from the study may improve cooperation and make the testing ex- done on adults, where the preference was for pulsed perience less strenuous for the young client. tones (Burk & Wiley, 2004). In this study, it was also RECOMMENDATIONS found that, overall, listeners’ perceived preference did A future study could include a larger sample and not correlate with a lower hearing threshold for the smaller age intervals with more subjects per age in- preferred stimulus signal. Reasons for these findings terval to ensure more generalizability of findings. are unexplained in literature to date. This would also allow for inter variable analysis on LIMITATIONS the influence of gender and auditory maturation on The study presented limitations in terms of sam- results. In addition, a larger sample may yield results ple size and ages included in the selected age range. that are more directly comparable to similar interna- In addition, as with most quasi-experimental studies, tional studies on adults. not all variables could be controlled for such as test- er-bias and the subjects’ attention span. Although, References the person who conducted the testing was the same American National Standards Institute. (1996). Specifica- person who analysed the data, the researcher ensured tions for Audiometers, S3.6. New York, ANSI. that data was recorded and analysed as accurately American Speech-Language-Hearing Association. (1978). and objectively as possible through supervision. The Guidelines for Manual Pure-tone Audiometry. Asha, person conducting the testing reported that all sub- 20, 279-301. jects demonstrated sufficient attention span for the THE SOUTH AFRICAN JOURNAL OF COMMUNICATION DISORDERS, VOL. 55, 2008 | 49 Catherine van Dijk and Naeema Osman

Blandy, S. & Lutman, M. (2005). Hearing threshold levels Johnson, C.D., Benson, P.V., & Seaton, J.B. (1997). Educa- and speech recognition in noise in 7-year-olds. Interna- tional Audiology Handbook. San Diego, Singular. tional Journal of Audiology, 44, 435-443. Katz, J. (2002). Handbook of Clinical Audiology. (5th Ed.). Burk, M.H. & Wiley, T.L. (2004). Continuous versus Baltimore, Lippincott Williams & Wilkins. pulsed tones. American Journal of Audiology, 13, 54-61. McCommons, R.B. & Hodge, D.C. (1969). Comparison Carhart, R. & Jerger, J. (1959). Preferred methods of deter- of continuous and pulsed tones for determining Beke- mination of pure-tone threshold. In J.W. Hall & H.G. sy threshold measurements. The Journal of the Acoustical Meuller (Eds.). Audiologists’ Desk Reference. San Diego, Society of America, 45, 1499-1504. Singular. Margolis, R.H. & Saly, G.L. (2007). Towards a standard Dancer, J.E. & Conn, M. (1983). Effects of two procedural description of hearing loss. International Journal of Au- modifications of the frequency of false-alarm responses diology, 46, 746-758. during pure-tone threshold determination. Journal of Martin, F.N. & Clark, J.G. (2003). Introduction to Audiol- Auditory Research, 23, 215-219. ogy. (8th Ed.). Boston, Pearson Education. Dancer, J., Ventry, I.M. & Hill, W. (1976). Effects of stim- Martin, F. & Sides, D.G. (1985). Survey of current audio- ulus presentation and instructions on pure-tone thresh- metric practice. Asha, 27, 29-36. olds and false-alarm responses. Journal of Speech and Mineau, S.M. & Schlauch, R.S. (1997). Threshold meas- Hearing Disorders, 41, 315-324. urement for clients with tinnitus: Pulsed or continuous Diefendorf, A.O. (2002). Detection and assessment of tones. American Journal of Audiology, 6, 52-56. hearing loss in infants and children. In J. Katz (ed.). Newby, H.A. & Popelka, G.R. (1985). Audiology. (5th Ed.). Handbook of Clinical Audiology (5th Ed.). Baltimore, New Jersey, Prentice-Hall. Lippincott Williams & Wilkins. Pappas, D.G. (1998). Diagnosis and Treatment of Hearing DiGiovanni, J.J. & Repka, J.N. (2007). Response method Impairment in Children. (2nd Ed.). San Diego, London, in audiometry. American Journal of Audiology, 16, 145- Singular. 148. Reger, S.N. (1970). Bekesy audiometry and the method of Gardner, M.B. (1947). A pulsed-tone technique for clini- limits. International Audiology, 9, 24-29. cal audiometric threshold measurements. The Journal of Silman, S. & Silverman, C.A. (1991). Auditory Diagno- the Acoustical Society of America, 19, 592-599. sis: Principles and Applications. San Diego, Academic Hall, J.W. (2007). New handbook of auditory evoked respons- Press. es. Boston, Pearson Edcuation. Sorensen, H. (1962). Clinical application of continuous Hallpike, C.S. & Hood, J.D. (1951). Some recent work on threshold recording. Acta Otolaryngology, 54, 403-422. auditory adaptation and its relationship to the loudness Struwig, F.W. & Stead, G.B. (2001). Planning, Designing recruitment phenomenon. The Journal of the Acoustical and Reporting Research. South Africa, Pearson Educa- Society of America, 23, 270-274. tion. Hochberg, I. & Waltzman, S. (1972). Comparison of United States Office of Special Education Programs. pulsed and continuous tone thresholds in clients with (2002). Child count: Number served (ages 6-21), by dis- tinnitus. Audiology, 11, 337-342. ability and state. Retrieved July 22, 2006, from http:// Hood, J.D. (1955). Auditory fatigue and adaptation in the www.ideadata.org/tables26th/ar_aa3.htm. differential diagnosis of end-organ disease. Annals of Yoshinaga-Itano, C., Sedey, A.L., Coultier, D.K., & Mehl, Otology, Rhinology and Laryngology, 64, 507-518. A.L. (1998). Language of early- and later- identified Jerger, J. (1960). Bekesy Audiometry in analysis of audi- children with hearing loss. Pediatrics, 102, 1161-1171. tory disorders. Journal of Speech and Hearing Research, 3, 275-287. 50 | DIE SUID-AFRIKAANSE TYDSKRIF VIR KOMMUNIKASIE-AFWYKINGS, VOL. 55, 2008