C O N T I N U O U S V E R S U S P U L S E D P U R E - T O N E A U D I O M E T R Y IN A G R O U P O F S C H O O L - A G E D C H I L D R E N C a th e rin e v a n D ijk * a n d N a e e m a O sm an D e p a r tm e n t o f C o m m u n ic a tio n P athology, U n iversity o f P retoria, S o uth Africa A bstract P u re -to n e te s tin g is th e p rim a ry a u d io lo g ic a l te s t p r o c e d u r e fo rth e d iffe r e n tia ld ia g n o s is o f h e a rin g lossan d h e a rin g disor- d e rs in s c h o o l-a g e d ch ild re n . N o re s e a rc h is c u rre n tly a v a ila b le in te rn a tio n a lly fo rc h ild re n 's re s p o n s e s to c o n tin u o u s versus p ulse d p u re -to n e s . T h e a im o f this e x p lo ra to ry in v e s tig a tio n w as to c o m p a re th e p e rfo rm a n c e o f a g ro u p o f sch o o l-a g e d ch ild re n o n c o n tin u o u s versus p ulsed p u r e -to n e a u d io m e try .T h e aim s w e re to d e te r m in e w h e th e r a th re s h o ld d iffe re n c e existe d b e tw e e n c o n tin u o u s versus p ulse d p u re -to n e s a n d to record w h e th e r a listen er p re fe re n c e existed b e tw e e n con­ tin u o u s versus p u lse d to n e s fo r th e fr e q u e n c y ra n g e o f 125 to 8 0 0 0 H z. E ig h te e n c h ild re n (3 6 ears) a g e d b e tw e e n 8 -1 2 years, p a r tic ip a te d in a h e a rin g e v a lu a tio n as w e ll as in a b r ie f th r e e -q u e s tio n in te rv ie w . D e s c rip tiv e statistics viz. a v e ra g e th re s h o ld , m e a n d iffe r e n c e a n d s ta n d a rd d e v ia tio n o fth re s h o ld s w e re u s e d to a n a ly s e d a ta .L is te n e rs 'p e rc e iv e d p re fe re n c e s w e re c a lc u la te d in p e rc e n ta g e s a n d re a s o n s fo r p re fe r rin g o n e s ig n a lo v e ra n o th e r w a s a n a ly s e d q u a lita tiv e ly .A lth o u g h th e a u to m a tic a lly p u ls e d to n e th r e s h o ld (a v e r a g e d a c r o s s t h e fr e q u e n c ie s t e s te d )w a s lo w e r th a n f o r th e c o n tin u o u s to n e ,th e d iffe re n c e w as o n ly 0.2 dB in th e le ft e a r an d 0.5 dB in th e rig h t ear. This sm all d iffe re n c e is n o t im p o r ta n t in clinical a p p lic a - tio n s fo rw h ic h 5 d B in c re m e n ts a re used in p u r e -to n e a u d io m e tr y .W h e r e a listen er p re fe re n c e w as in d ic a te d , h o w e v e r, th e co n tin u o u s to n e s w e re p re fe rre d o v e r pulse d to n e s b y 5 6 % p e rc e n t o f subjects. T h es e fin d in g s d iffe r fro m sim ilar studies in v o lv in g adults. This re v e a le d th a t ch ild re n m a y y ie ld d iffe re n t p referen ces d u rin g p u r e -to n e te s tin g th a n a d u lts a n d th a t th e s e p referen ces sho u ld b e ta k e n in to co n s id e ra tio n d u rin g te s tin g . K e y w o rd s : a u d io m e try , c o n tin u o u s p u r e -to n e , listen er p re fe re n c e , p u lse d p u re -to n e , p u r e -to n e te s tin g , sc h o o l-a g e d c h ild re n , s tim u lu s signal, th re s h o ld d iffe re n c e . There are two methods o f presentation o f pure-tones, i.e. continuous tones and pulsed tones. Pure-tone threshold measurement is a fundamental diagnostic tool in hearing evaluations (DiGiovanni 8c Repka, 2007). Pure-tone tests o f hearing, using audiometers,have been in use for nearly one hundred years and are quantifiable elec­ tronic extensions o f the same concepts developed in tuning fork tests (M artin 8c Clark, 2 0 0 3 ). I t is the primary audio­ logic test procedure for the differential diagnosis o f hearing loss and hearing disorders in school-aged children (Blandy 8c Lutman, 2 0 0 5 ).j In addition, pure-tone threshold audi­ ometry has become the standard behavioural procedure for describing auditory sensitivity (Margolis 8c Saly, 2007). In 1978, the American Speech-Language-Hearing Association (A SH A ) developed clinical guidelines referred to as Guide­ lines for Manual Pure-Tone Audiometry (A SH A , 1978). In these guidelines they recommended the use o f a one to two second sustained tone for threshold determination. This is referred to as a continuous tone. According to an American, survey o f audiometric practices two decades ago, the ma­ jority o f audiologists surveyed, used a manual presentation o f continuous tones (M artin 8c Sides, 1985). Although the A S H A guidelines (A SH A , 1978) recommend use o f a con­ tinuous tone, a substitution o f pulsed tones is also permitted by A SH A . A pulsed tone can be presented either manu­ ally or can be automatically generated by the audiometer. A manually presented pulsed tone would require the tester to present two consecutive tones with a short quiet interval be­ tween the two tones. This interval should be one second or shorter in duration (A SH A , 1978). Studies on children’s performance on and preference for continuous versus pulsed pure-tone audiometry is lacking. A study conducted by Burk and W iley (2004) on women aged 19 years to 4 4 years showed that pulsed tones are clearly ad­ vantageous in presenting what is perceived as an easier task. For many adult clients the pulsed tone is perceived as easier to hear at near threshold levels (Newby 8c Popelka, 1985). O ther published data suggests that using pulsed tones for adults having tinnitus may result in fewer false-positive re­ sponses (30% less than for continuous tones) and therefore may reduce the number o f presentations required to deter­ mine the threshold, resulting in more efficient audiological threshold evaluations (Mineau 8c Schlauch, 1997). There are selected studies that support the use o f pulsed tones rather than continuous tones in manual threshold au­ diometry for adults (Dancer 8c Conn, 1983; Dancer, Ventry 8c H ill, 1976; Gardner, 1947; H ochberg 8cW altzm an, 1972; Hood, 1955; Mineau 8c Schlauch, 1997). Firstly, in compar- Author Contact: Department o f Communication Pathology University o f Pretoria 0002 South Africa Fax: +2712 420 3517 Tel: +2712 420 2357 E-mail: Catherine. VanDijk@up.ac.za TH E S O U T H A F R IC A N J O U R N A L O F C O M M U N IC A T IO N D ISO RD ER S, V O L 55 2 0 0 8 29 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) mailto:VanDijk@up.ac.za C A TH ER INE V A N D D K A N D N A E E M A O S M A N ison to continuous tones, the use o f pulsed tones has been shown to decrease the number o f false-positives in both nor­ mal-hearing listeners and listeners with sensorineural hear­ ing loss with or without tinnitus (Dancer & Conn, 1983; Dancer et al, 1976; Mineau 8c Schlauch, 1997). Secondly, listeners with normal hearing and those with sensorineural hearing loss and tinnitus preferred pulsed tones to continu­ ous tones as a listening task (Gardner, 1947; Hochberg 8c Waltzman, 1972). The use o f pulsed tones in Bekesy audi­ ometry has also provided insight into using pulsed tones in conventional audiometry (Jerger, 1960). In threshold ap­ plications o f Bekesy audiometry, pulsed tones are typically presented to minimize the effects o f auditory adaptation to a continuous tone (Hallpike 8c Hood, 1951; Jerger, 1960; Reger, 1970; Silman 8c Silverman, 1991; Sorensen, 1962). In addition, i f a 200-m illiseconds-on, 200-m illiseconds-off pulsed tone is used for threshold assessment, periodic supra­ threshold presentations o f the test tone acts as a cue for the signal frequency, thereby refreshing the client’s memory for the test stimulus (Mineau 8c Schlauch, 1997). Furthermore, M cC om m ons and Hodge (1969) found that the pulsed tones used in Bekesy audiometry provide more feedback or cues by which the listener can make a decision regarding the presence o f the signal. The slight preference for pulsed tones in normal-hear- ing listeners (Burk 8c Wiley, 2 0 0 4 ), coupled with previous reports demonstrating the benefits o f using pulsed tones in threshold assessment for listeners with sensorineural hear­ ing loss and tinnitus (Hochberg 8c W altzman, 1972; M in e­ au 8c Schlauch, 1997), supports the general use o f pulsed tones in audiometry. However, in practice it may seem that not all audiologists use this technique. From personal ob­ servations, South African audiologists often deviate from standard testing procedures and protocols conveyed to them by their undergraduate training institutions. A diversity o f testing techniques and methods are often applied by au­ diologists in practice and may be due to, amongst others, their employer’s personal preferred method o f testing that has been imparted to them. The type o f test signal (con­ tinuous or pulsed) used in pure-tone audiometry may affect the accuracy and reliability o f pure-tone thresholds. Some data regarding adults’ responses to continuous and pulsed pure-tone audiometry is available (for example: Burk 8c Wiley, 2 0 0 4 ). However, similar data is currently unavailable for children’s responses to continuous versus pulsed tones. Therefore, this study is necessary to explore i f children may respond differently in their preference to adults. Indeed, this study has great relevance given that threshold audiometry by air conduction is a widely applied procedure for obtain­ ing auditory thresholds in children, the validity and vari­ ability o f tonal stimuli presented has not yet been addressed. Indeed, the apparent simplicity o f conducting the test often masks the importance o f investigating specific procedures applied in pure-tone audiometry. Effective pure-tone test procedures are imperative for the timely and accurate diag­ nosis o f hearing loss in children. Hearing loss depending on severity, age o f onset, and a host o f other factors has a mild-to-profound impact on com­ municative functioning (Yoshinaga-Itano, Sedey, Coulter 8c M ehl, 1998). The prevalence o f hearing loss in ch ild ren is such that its diagnosis and management necessitates high prioritisation. I t is estimated that in 2 0 0 2 approximately 72 00 0 school-aged children in the United States o f A m er­ ica (U .S.A .) alone fell in the “hearing impairment category” (United States Office o f Special Education Programs, 2 0 02). In South Africa, a developing context, prevalence o f hearing loss can be expected to be similar or even higher than in the U .S, but to date no statistical data is available for this popu­ lation. School-aged children with undetected hearing loss that have not received early intervention, often experience academic delays. As would be expected, the primary deficit areas for children with hearing loss are in those subjects that are language-based (Johnson, Benson 8c Seaton, 1997). In addition, school-aged children with a hearing loss appear to have increased rates o f grade failures, need for educational assistance, and perceived behavioural issues in the classroom (Johnson, Benson 8c Seaton, 1997). An undetected hear­ ing loss may have a cascading effect in terms o f habilitation. Various effects are insurmountable - such as economic sta­ tus, education, psychosocial stigma and eventually potential employment opportunities as an adult. Fortunately, there is a continued emphasis nationally and internationally on ear­ ly identification and monitoring o f hearing loss. Therefore, it is vitally important to obtain optimal thresholds for chil­ dren as these results have direct consequences on identifica­ tion, diagnosis and management o f a hearing loss. Early de­ tection followed by appropriate intervention maximises the benefits to the child, family, and society (Diefendorf, 2 0 0 2 ; Pappas, 1998). Part o f the intervention process includes the appropriate fitting o f hearing aids. However, i f optimal and reliable hearing thresholds are not obtained, over or under amplification may occur. Although more objective tests have been developed as part o f the test battery to obtain auditory thresholds, one cannot move away from standard pure-tone audiometry (Hall, 2 0 0 7 ). Electrophysiological tests like A E P (audito­ ry evoked potentials) and E C o c h G (electrocochleography) are used to estimate hearing thresholds. However, these test procedures often require sedation, may have artefact con-' tamination, and are not suitable for all types o f losses, e.g. conductive hearing losses or mild sensori-neural hearing losses (Hall, 2 0 0 7 ). Electrophysiological tests cannot sub- j stitute the use o f pure-tone audiometry where reliable be- ! havioural responses can be obtained to determine frequency ! specific information. D ata regarding the pure-tone testing o f school-aged chil­ dren is o f significant importance as this population is most­ ly evaluated with pure-tone audiometry. School-aged chil­ dren often pose unique challenges to the audiologist during pure-tone testing (Johnson, et al, 1997). These ch allen g-, es include assessment o f children with developmental de­ lays in, for example, receptive language, which reduces their comprehension and following o f test instructions resulting in troublesome test behaviour. Some children will not toler­ ate earphones or ear inserts for long periods o f time as this makes them uncomfortable. In comparison to adults, chil­ dren have a limited attention span and cannot remain still or concentrate for long periods during testing. This may cause them to become restless and/or uncooperative. False-posi­ tive responses are p rev alen t in this population an d failure 30 | DIE SU ID -AFR IK A A N SE TY D SK R IF VIR K O M M U N IK A S IE -A F W Y K IN G S ,V O L . 5 5 ,2 0 0 8 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) C O N T I N U O U S V E R S U S P U L S E D P U R E - T O N E A U D I O M E T R Y IN A G R O U P O F S C H O O L - A G E D C H IL D R E N to respond to stimuli close to their hearing thresholds are quite common. In addition, children’s reliability o f respons­ es tends to decrease as the time o f testing increases (Jo h n ­ son et al, 1997). Therefore, threshold data comparisons for continuous tones and pulsed tones in school-aged children are justified because it may be, for example, that one stimulus tone is perceived as an easier task to respond to, resulting in a shorter testing time. A shorter testing time may address assessment challenges such as reduced attention span and may increase the reliability o f responses. Comparisons o f continuous and pulsed pure-tone thresholds may yield findings that indicate better thresholds obtained for one o f these methods. These findings may guide audiologists when deciding which stimulus tone to use when testing children with pure-tone stimuli. A review o f available professional literature has clearly revealed a lack o f data to enable comparisons o f continuous tones and pulsed tones in children using the manual pure- tone technique (as recommended by A S H A , 1978). This exploratory investigation aimed to compare threshold data and listener preference for continuous and pulsed tones on a small group o f children. Therefore, the main aim o f this study was to compare continuous and pulsed pure-tone audiometry in a group o f school-aged children. The sub­ aims o f the study were: • to determine whether a threshold difference existed between continuous versus pulsed pure-tone audiometry for the frequency range o f 125 H z to 8 000 H z; and • to determine listener preference for continuous ver­ sus pulsed tones for the frequency range o f 125 H z to 8000 Hz. Method Design This exploratory study was mainly quantitative in na­ ture consisting o f pure-tone audiometric testing and a brief, three-question interview. A comparative based study em­ ploying a quasi-exp’erimental design o f research was used as not all variables could be controlled for (Struwig 8c Stead, 2 0 01). The main aim o f this study was to compare continuous and pulsed pure-tone audiometry in a group o f school-aged children. Participants • Ethical issues ! E thical clearance was obtained from the departmental research committee o f the University o f Pretoria before the investigation was carried out. W ritten informed consent and assent was obtained from the participants and their guard­ ians. Participants and their guardians were informed that their participation was voluntary and that they could with­ draw without consequence at any stage during the study. All identifying information was kept confidential and partici­ pants were not subjected to harmful or unjust procedures. Participant selection criteria Participants were required to respond reliably and had to concentrate for long periods (40 minutes or longer). For this reason, children between the ages o f 8 years to 12 years were selected. I t was felt that children in this age group were more able than younger children to perform these tasks consistently. Participants had to present with normal hearing as the presence o f a hearing loss may influence the ability to distinguish between the continuous and pulsed pure-tone test signal. Furthermore, most researchers are in agreement that when investigating the benefits o f any audiological test procedure, normative data should be recorded before similar investigations can be conducted on hearing-impaired popu­ lations (Hall, 2007). In order to ascertain hearing status, all participants were subjected to a test battery prior to the study. The test battery included otoscopy, tympanometry, reflex testing, oto-acoustic emission screening and comprehensive pure- tone bone and air conduction testing. Sampling Participants were purposefully selected by requesting fi­ nal year students to supply contact details o f family mem­ bers or acquaintances that had children between the ages o f 8 years and 12 years. The participants were expected to visit the H earing C linic at the University o f Pretoria and there­ fore only children from the Pretoria region were targeted. D escription o f participants Eighteen children (36 ears) with normal hearing thresh­ olds for all seven tested frequencies were included. Accord­ ing to Katz (2002) and M artin and Clark (2003) a range o f 0 dB to 2 0 dB can be regarded as normal hearing for school-aged children. E ig h t o f the participants were male and ten were female. The age distributions were as follows: three were eight years old, one was nine years old, four were ten years old, six were eleven years old and four were twelve years old. Apparatus and M aterials Apparatus an d m aterial f o r pre-selection O toscopic examinations were conducted using a W elch- Allyn otoscope. Acoustic immittance measures were ob­ tained using a calibrated (according to S A B S 0154-1/2, 0182) tympanometer (G SITym pstar). A Scout O to-A cous­ tic Emission machine with 2 kH z to 6 kH z (4 dB to 6 dB for a pass) screening protocol was used to perform oto- acoustic emission screening. Pure-tone air and bone con­ duction thresholds were obtained using a twin channel di­ agnostic audiometer, the G S I 61. Participants were seated in a sound-insulated booth (Industrial Acoustics Com pa­ ny, In c.). Calibration is conducted periodically in accord­ ance with American National Standards Institute (A N SI) 1996 specifications and listening checks o f the equipment were performed prior to testing. Results were recorded on a standard audiogram. Apparatus and material for main study Pulsed and continuous hearing thresholds were obtained using the same sound-insulated booth and diagnostic audi­ ometer employed for the pre-selection o f subjects. A n ad­ aptation o f a standard audiogram was used to record the results. The symbol that was used for both ears to indicate continuous hearing threshold level was an outline o f a cir­ cle and a solid circle indicated pulsed hearing thresholds. A brief three-question interview schedule was used after com ­ pletion o f the hearing test to obtain information about lis­ tener preference for the type o f test signal. Procedures Pure-tones were presented through earphones to assess conduction thresholds. The results o f the air conduc-air T H E S O U T H A FR IC A N J O U R N A L OF C O M M U N IC A T IO N D ISO R D ER S, VO L . 55 2 0 0 8 | 31 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) tion tests in both test conditions (continuous and pulsed) were plotted on an audiogram from which the results were compared. The ascending/descending method developed by Carhart and Jerger (1959) was used to obtain pure-tone thresholds bilaterally at 125 H z, 25 0 H z, 50 0 H z, 1000 Hz, 2 0 0 0 H z, 4 0 0 0 H z, and 8000 H z. Octave frequencies o f 3000 H z and 6 0 0 0 H z were excluded to reduce testing time. However, these octaves were included in the initial selec­ tion procedures to determine normal criteria. Steps o f 5 dB were used to confirm hearing threshold levels. Participants took part in one testing session with two alternating sig­ nal conditions. The signal conditions were a manually pre­ sented continuous tone for one second to two seconds and two short pulses (rise-fall time o f 35 ms with a duration o f 20 0 ms onset to offset) automatically generated by the au­ diometer. The test signals were presented in an alternating manner to ensure that the same test signal (e.g. a pulsed pure tone) was not always presented in the same order to the particpant. For example, the participant would hear, at 125 H z, first a pulsed then a continuous tone. A t 2 5 0 Hz, he/she would first hear a continuous then a pulsed tone.This would randomly vary from participant to participant. This presentation ensured reliability and validity o f results and yielded two sets o f thresholds (continuous and pulsed tones) for each ear. The first signal (either automatically pulsed or a continuous tone) was randomly chosen and followed al­ ternately by the other signal. A fter the hearing test, a brief three-question interview followed the pure-tone testing. Subjects were asked i f they preferred one test signal above the other, and, i f so, which signal they preferred and why. Their answers were recorded by the researcher on a space provided on the audiogram. Testing and recording was done by a final year audiology student. A qualified audiologist su­ pervised the student during testing o f the first three subjects to ensure that test procedures were valid and reliable. The supervisor was available for consultation for the duration o f the testing. D a ta analysis Data was organised and processed using “M icrosoft O f­ fice Excel” software. A univariate procedure was used in order to draw comparisons between continuous and pulsed threshold values at discreet frequencies. The average thresh­ old, mean difference and standard deviation o f the two thresholds obtained with the two different types o f stimuli at each frequency tested for each ear, was calculated. L is ­ teners’preferences indicated in the interview were calculated in percentages to obtain the most preferred signal. Reasons for preferring one signal to another was analysed qualita­ tively. During data analysis the results for each ear were not combined as literature suggests that ear differences oc­ cur (Hochberg &. Waltzman, 1972). The afore-mentioned statistical procedures used in the study were deemed appro­ priate for the small sample size and provided sufficient data in order to answer the research question comprehensively. Results D eterm ining whether a threshold difference exists be­ tween continuous versus pulsed pure-tone audiometry The first sub-aim o f the study was to determine wheth­ er a threshold difference exists between continuous ver­ sus pulsed pure-tone audiometry in school-aged children. C A TH ER INE VA N DIJK A N D N A E E M A O S M A N These results included hearing threshold levels obtained for continuous as well as pulsed tones at frequencies 125 H z to 8000 H z in each ear o f each participant. The difference between the average o f the continuous hearing thresholds and the average o f the pulsed hearing thresholds for each ear and specific frequencies were determined. These dif­ ferences were calculated by subtracting the averaged pulsed tones from the averaged continuous tones and are depicted in Figure 1. Frequencies Figure 1: Difference between average hearing threshold levels o f continuous andpulsed tones in each ear (n=18) A difference value o f zero, which was obtained at 25 0 H z in the left ear and at 1000 H z in the right ear, indicated no threshold difference between the two stimulus types as presented in Figure 1. Negative values indicated that lower thresholds were obtained for continuous tones. Continuous tones yielded better thresholds than pulsed tones at 50 0 H z in the left ear and 2 0 0 0 H z in the right ear. Lower thresh­ olds were obtained for pulsed tones at all other frequencies for both ears. Therefore, pulsed tones yielded lower thresh­ olds than continuous tones for the majority o f frequencies tested. In order to determine whether the signal type yielded a threshold difference at specific frequencies, findings per frequency for the respective ears were analysed. The stand­ ard deviation and the mean average o f the hearing threshold levels obtained at each frequency tested, for each stimulus Figure 2 : M ean average an d standard deviation f o r continu­ ous andpulsed tones in the left ear (n=18) The average standard deviation for continuous tones was 8 dB in the left ear. The average standard deviation for pulsed tones was 7.9 dB in the left ear. Continuous tones 32 | DIE SU ID -AFR IKA AN SE TYDSKRIF VIR K O M M U N IK A S IE -A FW Y K IN G S , VOL. 5 5 ,2 0 0 8 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) C O N T I N U O U S V E R S U S P U L S E D P U R E - T O N E A U D I O M E T R Y IN A G R O U P O F S C H O O L - A G E D C H IL D R E N presented less variation in thresholds at 125 H z, 50 0 Hz, 4 0 0 0 H z and 8 0 0 0 H z in the left ear while pulsed tones presented less variation in thresholds at 2 5 0 H z, 1000 H z and 2 0 0 0 Hz. Absolute differences in averaged thresholds at each test frequency ranged from 0 dB to 1.9 dB between 125 H z and 8 000 H z. Averaged pulsed tones yielded slight­ ly lower thresholds at all frequencies except at 2 5 0 H z where thresholds were equal and at 50 0 H z where the continuous tones revealed slightly lower thresholds. Similar to the left ear, findings per frequency were ana­ lysed for the right ear to determine whether the signal type yielded a threshold difference at specific frequencies. The standard deviation and the mean average o f the hearing threshold levels were obtained at each frequency tested, for each stimulus type, for the right ear, follows in Figure 3. Figure 3 : M ean averages an d standard deviation f o r continu­ ous an d pulsed tones in the right ear (n-18). The average o f the standard deviations for continuous tones was 7.3 dB compared to 7.5 dB for pulsed tones in the right ear. Continuous tones presented less variation in all thresholds except at 1000 H z and 4 0 0 0 H z in the right ear. Absolute differences in averaged thresholds at each test frequency ranged from 0 dB to 1.4 dB between 125 H z and 8 000 H z. i Averaged pulsed tones yielded slightly lower thresholds at all frequencies except at 1000 H z where thresholds were equal and 2 0 0 0 H z where the continuous tones obtained slightly lower thresholds. W h en results o f both ears are considered, threshold standard deviations (averaged over all frequencies tested) showed that the pulsed presentation method yielded slight­ ly less variation (0.005 dB) in thresholds, than the continu­ ous presentation method. O n average, pulsed tones yield­ ed slightly lower thresholds where the average pulsed tone threshold was 0 .4 dB lower than for the averaged contin­ uous tone stimulus. These small differences in averaged thresholds and standard deviations obtained for both test tones reinforce the equality o f the two stimulus signals. Based on the aforementioned, the overall differences would not be sufficient to recommend one signal above the other. Findings for both ears at all frequencies were plotted to compare overall threshold differences due to different sig­ nals. These results are presented in Figure 4. Figure 4: A comparison o f the average hearing thresholds ob­ tain ed for continuous andpulsed tones in each ear (n-18). As may be noted in Figure 4, the differences in thresh­ olds between the two stimuli were relatively small. This cor­ relates with results obtained by Hochberg and Waltzman (1 9 7 2 ), which indicated that both types o f signals obtained comparable threshold levels at all audiometric frequen­ cies tested, and did not vary by more than 1.6 dB. Across all frequencies tested in the left ear, an average o f 3 .8 dB for continuous tones and 3.6 dB for pulsed tones was ob­ tained. Similarly, the average hearing thresholds obtained in the right ear were 5 .7 dB for continuous tones and 5 .2 dB for pulsed tones. Although the automatically pulsed tone threshold (averaged across the frequencies tested) was low­ er than for the continuous tone, the average 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. Taken as a whole, the pulsed tone stimuli produced slightly lower thresholds at all the test frequencies where the average pulsed tone threshold was 0.3 dB lower than for the continuous tone stimulus. This differs from the study done on adults where lower thresholds (averaged across all frequencies) were obtained for continuous tones than for automatically pulsed tones, where the difference was 1.0 dB (Burk &. W iley, 2 0 04). The obtained minor 0.3 dB differ­ ence would not result in a substantial difference in hearing threshold level, since these measurements are made in 5 dB increments. D eterm ining whether a listener preference is perceived between continuous versus pulsed tones The second aim o f the study was to determine whether participants had a listener’s preference for one o f the two signals. The percentage o f participants preferring one signal over the other based on the three questions asked, was cal­ culated. These results are represented Figure 5. TH E S O U T H A F R IC A N J O U R N A L OF C O M M U N IC A T IO N D ISO RD ER S, V O L. 55 2 0 0 8 | 3 3 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) C A TH ER INE V A N DIJK A N D N A E E M A O S M A N 0 Continuous (Beep w as easier) Z Pulsed (Beep-beep was easier)___________ F ig u re 5 : L isten ers’ p referen ce f o r one stimulus type ov er the other (n=18). The difference in the percentage o f perceived preference was 11%, with 56% preferring continuous tones and 44% preferring pulsed tones. This differs from the findings o f a study (Burk 8c Wiley, 20 0 4 ) done on 2 4 adults, where it was found that 62% o f subjects preferred the pulsed tones and 38% preferred the continuous tones. O ver two-thirds o f normal and tinnitus subjects, in a study done by Hochberg and Waltzman (1972), preferred to listen to pulsed tones as opposed to continuous tones during threshold determina­ tion. The difference in the sample population group in the present study, which was done with children, may contribute to the disparity in the findings. Results indicated that more participants preferred the continuous tone, although, the post-test comments across the signal types were very similar. This differs from pre­ vious investigations where persons experiencing tinnitus commented that continuous tones are more confusing than pulsed tones (M ineau 8c Schlauch, 1997). Findings revealed that the most common explanation given by those partici­ pants preferring the continuous tones was that it was louder and clearer and that it did not have interruptions. The most common reason given by those participants preferring the pulsed tones was that one could hear it a second time if one was unsure o f hearing the first presentation. A direct comparison provided further findings regard­ ing the correlation between the perceived listener’s prefer­ ence and the stimulus signal that produced the lower hear­ ing threshold level. Seven subjects obtained lower hearing thresholds in both ears for the stimulus signal that was not their perceived preference. A further seven subjects per­ ceived preference correlated with a lower hearing threshold in one ear only. O nly four participants obtained better hear­ ing thresholds in both ears for the stimulus signal that they perceived as their preference. Therefore, it would seem that listeners’ perceived preference did not correlate with a lower hearing threshold for the preferred stimulus signal. Discussion and Conclusion The aim o f this study was to compare continuous and pulsed pure-tone audiometry in a group o f normal-hearing school-aged children. This was done by comparing hear­ ing threshold levels obtained for both stimulus signals and determining i f a listener’s preference existed for one tone over the other. The value o f this study, was that data, unlike previous studies o f the same nature (Burk 8c Wiley, 2004; M ineau 8c Schlauch, 1997), was recorded and analysed sep­ arately for each ear at the various test frequencies. In summary, the findings obtained for the averaged hear­ ing threshold differences at each frequency when comparing continuous to pulsed hearing threshold levels were: Pulsed tones yielded slightly lower thresholds in both ears at 125 Hz. A t 2 5 0 H z, no difference in hearing threshold lev­ els were obtained in the left ear and pulsed tones yielded slightly lower thresholds in the right ear. A t 5 0 0 H z, continuous tones yielded slightly lower thresholds in the left ear and pulsed tones yielded slightly lower thresholds in the right ear. A t 1000 Hz, pulsed tones yielded slightly low­ er thresholds in the left ear and no difference in hearing threshold levels were obtained in the right ear. A t 2 0 0 0 H z, pulsed tones yielded slightly low­ er thresholds in the left ear and continuous tones yielded slightly lower thresholds in the right ear. Pulsed tones yielded slightly lower thresholds in both ears at 4 0 0 0 Hz. Pulsed tones yielded slightly lower thresholds in both ears at 8 000 Hz. The above-mentioned findings revealed that the pulsed tone stimuli produced slightly lower thresholds (0.3 dB) on average at all the test frequencies than for the continu­ ous tone stimulus. Literature does not provide an explana­ tion for why audiometric pulsed tones yield slightly lower thresholds, except that subjects tend to report that a pulsed tone seems to be more alerting than a continuous tone (M c - Commons 8c Hodge, 1969). Inter-aural differences were apparent at some test frequencies in this investigation and are also documented (Hochberg 8c Waltzman, 1972). The ear that produced the better pulsed tone thresholds var­ ied within the same subject, which makes the theory o f ear domination invalid for these cases. Current literature does not aid in providing an explanation nor are the results o f this study useful to explain why these inter-aural differences are found for continuous and pulsed tone thresholds. However, differences are statistically so small that it does not warrant speculation as to why they occur. I t may be concluded that the use o f pulsed tones for audi­ ometric threshold measures in normal-hearing children had no clinically significant effect on obtained hearing thresh­ olds as 5 dB increments are used in pure-tone audiometric testing. This differs from previous findings demonstrating the benefits o f using pulsed tones in threshold assessment for adults with sensorineural hearing loss and persons with tinnitus (Burk 8c Wiley, 2 0 0 4 ; Hochberg 8c Waltzman, 1972; M ineau 8c Schlauch, 1997). These preliminary find­ ings may indicate that school-aged children yield different responses during pure-tone testing than adults. However, continuous tones showed a slight advantage in terms o f presenting what is perceived as an easier task with­ out adversely affecting the test’s outcome. A slight difference was obtained, with more participants preferring continuous tones than pulsed tones. The slight preference for continu­ ous tones in normal-hearing children also differs from the study done on adults, where the preference was for pulsed tones (Burk 8c Wiley, 2 0 0 4 ). In this study, it was also found 34 | DIE SU ID -AFR IKA AN SE TYDSKRIF VIR K O M M U N IK A S IE -A FW Y K IN G S , VO L. 5 5 ,2 0 0 8 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) C O N T I N U O U S V E R S U S P U L S E D P U R E - T O N E A U D I O M E T R Y IN A G R O U P O F S C H O O L - A G E D C H IL D R E N that, overall, listeners’ perceived preference did not correlate with a lower hearing threshold for the preferred stimulus signal. Reasons for these findings are unexplained in litera­ ture to date. Limitations The study presented limitations in terms o f sample size and ages included in the selected age range. In addition, as with most quasi-experimental studies, not all variables could be controlled for such as tester-bias and the participants’ at­ tention span. Although, the person who conducted the test­ ing was the same person who analysed the data, the research­ er ensured that data was recorded and analysed as accurately and objectively as possible through supervision. The person conducting the testing reported that all participants dem­ onstrated sufficient attention span for the duration o f the test procedure. She indicated that, although some subjects showed obvious signs o f boredom, this resulted in only a few inconsistent responses and not more than one would expect during behavioural testing. Despite these shortcomings, limited research is available to date that compares outcomes o f different test signals in pure-tone audiometry in children and this exploratory study provides preliminary data with some clinical considerations for audiologists in practice. Clinical Implications The benefits o f using pulsed tones over continuous tones have been documented with adults with hearing-loss as well as with adults suffering from tinnitus. However, in this ex­ ploratory study it would seem that using either continuous or pulsed pure-tone signals during testing with normal- hearing children does not yield significant threshold differ­ ences. Conversely, audiologists might consider asking their young clients during testing which signal they preferred. As revealed in this study, this may not result in obtaining better threshold levels. Nevertheless, taking their preference into consideration may improve cooperation and make the test­ ing experience less strenuous for the young client. Recommendations A future study could include a larger sample and smaller age intervals with more subjects per age interval to ensure more generalizability o f findings. This would also allow for inter variable analysis on the influence o f gender and audi­ tory maturation onj results. In addition, a larger sample may yield results that are more directly comparable to similar in­ ternational studies on adults. References American National Standards Institute. (1996). Specifications fo r Audiometers, S3.6. New York: ANSI. American Speech-Language-Hearing Association. (1978). Guide­ lines for Manual Pure-tone Audiometry. Asha, 2 0 ,279-301. Blandy, S. & Lutman, M. (2005). Hearing threshold levels and speech recognition in noise in 7-year-olds. International Journal o f Audiology, 44, 435-443. Burk, M.H. & Wiley, T.L. (2004). Continuous versus pulsed tones. American Journal o f Audiology, 1 3 ,54-61. Carhart, R. &Jerger, J. (1959). Preferred methods of determination of pure-tone threshold. In J.W. Hall & H.G. Meuller (Eds.). Audiologists’ Desk Reference. San Diego: Singular. Dancer, J.E . & Conn, M. (1983). Effects o f two procedural modi­ fications of the frequency of false-alarm responses during pure- tone threshold determination. Journal o f Auditory Research, 23, 215-219. DancerJ., Ventry, I.M. & Hill, W. (1976). Effects of stimulus pres­ entation and instructions on pure-tone thresholds and false- alarm responses. Journal o f Speech and Hearing Disorders, 41, 315-324. Diefendorf, A.O. (2002). Detection and assessment of hearing loss in infants and children. In J. Katz (ed.). Handbook o f Clinical Au­ diology (5 th Ed.). Baltimore: Lippincott Williams & Wilkins. DiGiovanni,J.J. &Repka,J.N. (2007). Response method in audi­ ometry. American Journal ofAudiology, 1 6 ,145-148. Gardner, M.B. (1947). A pulsed-tone technique for clinical audio­ metric threshold measurements. The Journal o f the Acoustical So­ ciety o f America, 1 9 ,592-599. Hall, J.W. (2007). New handbook o f auditory evoked responses. Bos­ ton: Pearson Edcuation. Hallpike, C.S. & Hood, J.D. (1951). Some recent work on audi­ tory adaptation and its relationship to the loudness recruitment phenomenon. The Journal o f the Acoustical Society o f America, 23, 270-274. Hochberg, I. & Waltzman, S. (1972). Comparison of pulsed and continuous tone thresholds in clients with tinnitus. Audiology, 1 1 ,337-342. Hood, J.D . (1955). Auditory fatigue and adaptation in the differ­ ential diagnosis of end-organ diseas e. Annals o f Otology, Rhinol- ogy and Laryngology, 6 4 ,507-518. Jerger, J. (1960). Bekesy Audiometry in analysis of auditory disor­ ders. Journal o f Speech and Hearing Research, J , 275-287. Johnson, C.D., Benson, P.V., & Seaton, J.B. (1997). Educational Audiology Handbook. San Diego: Singular. Katz, J. (2002). Handbook o f Clinical Audiology. (5th Ed.). Baltimore: Lippincott Williams & Wilkins. McCommons, R.B. & Hodge, D.C. (1969). Comparison of con­ tinuous and pulsed tones for determining Bekesy threshold measurements. The Journal o f the Acoustical Society o f America, 4 5 ,1499-1504. Margolis, R.H. & Saly, G.L. (2007). Towards a standard descrip­ tion of hearing loss. International Journal o f Audiology, 46, 746- 758. Martin, F.N. & Clark, J.G . (2003). Introduction to Audiology. (8th Ed.). Boston: Pearson Education. Martin, F. & Sides, D.G. (1985). Survey of current audiometric practice. Asha, 2 7 ,29-36. Mineau, S.M. & Schlauch, R.S. (1997). Threshold measurement for clients with tinnitus: Pulsed or continuous tones. American Journal o f Audiology, 6 ,52-56. Newby, H.A. & Popelka, G.R. (1985). Audiology. (5,h Ed.). New Jersey: Prentice-Hall. Pappas, D.G. (1998). Diagnosis and Treatment o f Hearing Impair­ ment in Children. (2nd Ed.). San Diego: Singular. Reger, S.N. (1970). Bekesy audiometry and the method of limits. International Audiology, 9 ,24-29. Silman, S. & Silverman, C.A. (1991). Auditory Diagnosis: Principles and Applications. San Diego: Academic Press. Sorensen, H. (1962). Clinical application of continuous threshold recording. Acta Otolaryngology, 5 4 ,403-422. Struwig, F.W. & Stead, G.B. (2001). Planning Designing and Re­ porting Research. South Africa: Pearson Education. United States Office of Special Education Programs. (2002). T H E S O U T H A F R IC A N J O U R N A L O F C O M M U N IC A T IO N D ISO RD ER S, V O L. 55 2 0 0 8 | 3 5 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Child count: Number served (ages 6-21), by disability and state. Retrieved July 22, 2006, from http://www.ideadata.org/ta- bles26th/ar_aa3.htm. CATH ER INE V A N D D K A N D N A E E M A O S M A N Yoshinaga-Itano, C., Sedey, A.L., Coultier, D.K., & Mehl, A.L. (1998). Language of early- and later- identified children with hearing loss. Pediatrics, 1 0 2 ,1161-1171. 36 | DIE SUID -AFR IKA AN SE TYDSKRIF VIR K O M M U N IK A S IE -A FW Y K IN G S , VOL. 5 5 ,2 0 0 8 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) http://www.ideadata.org/ta-