Pure-tone Audiogram
Measuring Auditory Sensitivity over the Age
João Paulo Teixeira
1
and Paula Odete Fernandes
1,2
1
Polytechnic Institute of Bragança, Bragança, Portugal
2
NECE-UBI, Covilhã, Portugal
Keywords: Pure-tone Audiogram, Hearing Sensitivity, Ear.
Abstract: In this study a pure tone audiogram was developed under the Matlab® mathematical software. Audiogram
measurements were performed to 35 subjects belonging to the female and male and aged between 10 and 88
years old. Some of the subjects with more advanced age had hearing problems over the course of age,
however, none of them was carrying any type of hearing aid. The threshold of the Sound Pressure Level
(SPL) was recorded under 12 pure tones between 125 Hz and 15 kHz. The developed pure-tone audiogram
confirmed its ability to produce auditory brainstem responses (ABRs). Statistical analysis of the SPL
threshold shows no differences between genders and confirms the correlation between age and loss of
sensitivity, more accentuated for higher frequency tones. A strong loss of sensitivity was observed after the
decade of 60 years old.
1 INTRODUCTION
The ear is a receiver associated with the
procurement, conduct, modification, amplification
and analysis of sound waves. The sound waves are
due to molecular vibrations in the air and
characterized by a given frequency and a given
intensity.
At present there are some problems associated
with normal operation of the ear which may result in
a hearing impairment. Deafness, in particular, is one
of the most common symptoms around the world,
reaching all ages, from newborns to the elderly,
through all the existing races, both gender, with a
huge response from the point of view of language
and communication, including, family, cultural,
professional, emotional and even psychological,
with large varied causes (Fransen et al., 2003).
The measure of hearing loss is performed by an
audiogram that submit the subject to stimuli with
different energy or amplitude and records the
threshold of energy necessary to be perceived by
subjects. However, the hearing sensitivity varies
along the frequency. The human hearing apparatus is
more sensitive of the bandwidth between 1 and
5kHz and less sensitive for frequencies below 100
Hz and above 10 kHz.
The literature refer different types of stimuli used
in audiograms, such as clicks, noise masking and
recorded sounds (Stapells and Oates, 1997).
Considering a flat cochlear hearing loss a stimulus
of 1000 Hz filtered clicks was also used (Conijn et
al., 1990). But the pure tone with different
frequencies also has been used by Stapells and
Oates, (Stapells and Oates, 1997).
In this context, we propose small pack software
to produce the pure tone waves and a program for
easily submit the subject to the audiogram test
searching for the threshold of energy of the sound
wave to be listen to. We also present the results of
several subjects and its analysis along the age and
gender considering always the different frequency
ranges.
2 SOUND AND HEARING
The sine wave, which is used in this work are sound
waves that have only one frequency component.
Therefore a sound with the shape of a sine is known
as a pure tone.
In relation to sound waves, they are associated
with changes in pressure called sound pressure and
is commonly expressed as Sound Pressure Level
(SPL) and is measured in decibel (dB). Sound waves
have wavelengths or frequencies of sound, measured
293
Teixeira J. and Fernandes P..
Pure-tone Audiogram - Measuring Auditory Sensitivity over the Age.
DOI: 10.5220/0004328802930296
In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2013), pages 293-296
ISBN: 978-989-8565-34-1
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
in cycles per second or Hertz (Hz).
The human ear is sensitive to sound frequencies
between 20 Hz and 20 kHz. The frequencies below
20 Hz are called infra-sound, whilst for frequencies
over 20 kHz, are called the ultrasound (often used in
medicine, more precisely on ultrasound). The
equation 1 presents the measure of the SPL in dB as
a relation between the pressure (P) and the reference
pressure (P
0
). This reference corresponds to the
threshold of hearing for young persons (aged from
18 to 25 year) at the pure tone frequency of 1000
Hz. This reference corresponds to 20 Pa. For a
sound of pressure P equal to P
0
the SPL is 0 dB.
10
0
20 log ( ) (dB)
P
SPL
P
(1)
Fig. 1, represents the relation between SPL over the
frequency, and pressure measurement in dB causing
the same sense of sound volume in young
individuals, in the presence of pure tones (ISO
226:2003).
Figure 1: Equal-loudness contour.
3 AUDIOGRAM
This work has focused on measuring the auditory
sensitivity along frequency and over the years. For
this purpose it was developed an audiogram in
software using the mathematical software
MATLAB®.
The objectives was measuring the threshold of
hearing at 12 different frequencies, ranging from
125Hz down to the 15000Hz, performing the
graphic corresponding to the relation frequency /
amplitude similarly as the graph presented in Fig.1.
For each frequency a pure tone wave produced
by a pure sine wave is created with initial amplitude.
Then, the amplitude is gradually adjusted until the
subject couldn’t perceive sound. The last amplitude
is registered and converted to SPL in dB.
A simple program as developed to produce the
pure tones along the selected frequencies to be used
during the audiogram test. The program intends to
facilitate the task of selecting the SPL to be
presented to the subject until find the threshold of
hearing.
No modulation of amplitude was performed in
contrast with the work presented by Canale in
(Canale et al., 2006). The program asks for the SPL,
then converts the SPL in dB to a linear scale and
produces a 2 second long sine wave signal with the
frequency and the amplitude introduced by the user.
The signal is sent to the audio system to be
presented to the subject. The subject should inform
if the sound was listen in an interactive way to
record the threshold of SPL. The procedure should
be repeated for every frequency tone.
The tests were performed with 35 subjects
belonging to the female and male gender, aged
between 10 and 88 years. Some subjects with more
advanced age had hearing problems over the course
of age, however, none of them was carrying any type
of hearing aid. Importantly, all measures were
performed on privileged and quite rooms and with a
good sound isolation. The subjects were always
placed in front of the computer. The measures were
made on the same computer, so this way all the tests
were performed with the same consistency in the
magnitude of sound waves. Therefore there were no
significant changes in the conditions to collect data
along subject.
The absolute SPL presented to the subject cannot
be determined once the sound card of the computer
has an unknown amplification. In order to deal is
this situation the value of 1 dB was used as a
reference for the threshold of hearing at 125 Hz. We
assumed that the audio system of the computer have
an equally gain along the frequency range (the
frequency response of the computer audio system
must be measured and considered in future
developments). Therefore the absolute values
presented in following section cannot be compared
with the absolute values of Fig 1, only the variation
of sensitivity along frequencies can be compared, by
the shape of the curves of sensitivity.
The same volume level in the computer was kept
during all measures.
BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices
294
4 RESULTS
First a comparison of each subject within his decade
age was performed for the decades of tens to decade
of eighties. Then a comparison between the averages
of subject of each decade is presented in Fig. 2. The
figure presents the average minimum SPL for each
frequency perceived by the subject of each decade.
As higher is the curve more SPL in need to be
perceived by listener that means less sensitivity.
Therefore, a higher value of SPL in the graphic may
represent a hearing loss at that frequency. Anyhow,
the variation of sensitivity along frequency as
presented in Fig. 1 has to be considered normal.
Figure 2: Hearing sensitivity of each decade.
The decades with lower SPL threshold are the
first four lower decades, ie concerning the decades
10, 20, 40, and finally concerning the decade at 50
years. These first four decades have a hearing
sensitivity pattern similar to the pattern presented in
Figure 1, meaning that the eventual hearing loss may
not be significant. Moreover, the latest age
corresponding to the 60, 70 and 80 decades illustrate
that the higher decades are correlated with a
significant loss of sensitivity. Thus, it is
recommended to the people belonging to this ages
and who have very significant hearing loss, the
possible use of hearing aids in order to aid the
process of hearing (Bento et al., 2010), (Cheng et al.,
2011).
5 STATISTICAL ANALYSIS OF
RESULTS
For all results of inferential analysis a 5% of
significance level was assumed.
Table 1 lists some characteristics of the subjects.
The original sample consisted in 35 subjects, 77,1%
female and 22,9% male. The 6 subjects with clinical
diagnosis of hearing pathology 83,3% are female,
and those who are not a clinical diagnosis of the
pathology 75,9% are female.
Table 1: General Characteristics of Sample (N=35).
subjects
No.
%
Gender
Female
27 77,1
Male
8 22,9
Age
Decade 10 7 20,0
Decade 20 7 20,0
Decade 30 1 2,9
Decade 40 5 14,3
Decade 50 3 8,6
Decade 60 2 5,7
Decade 70 5 14,3
Decade 80 5 14,3
Clinical diagnosis of Hearing
pathology
With 6 17,1
Without 29 82,9
In order to evaluate differences of SPL among
the eight age decades in each Frequency tones, a
Kruskal-Wallis test was conducted. The Kruskal-
Wallis Test is the nonparametric test equivalent to
the one-way ANOVA and is used when the
assumption of the parametric test have been too
grossly violated (Green and Salkind, 2008). It is
used to test the null hypothesis that all populations
have identical distribution functions against the
alternative hypothesis that at least two of the
samples differ only with respect to location
(median), if at all. Chi-square and p-value have been
used for the comparison. Results are not statistically
significant for the frequency tones of SPL at 125 and
500 Hz (p-value>0.05), it indicates that the groups
are not different from the others. The results of the
analysis for the remaining frequency tones indicates
that there is a significant difference in the SPL
medians, it is the same that at least one of the groups
is different from the others. It can therefore be
concluded that there are differences in the SPL
threshold at frequency tones among the group age
decades.
To evaluate if the median performance of the
SPL along the frequency tones differ for female and
male, the pairwise comparisons was conducted using
the Mann-Whitney U test (nonparametric test),
which yields identical results with the Kruskal-
Wallis test for two independent samples (Green and
Salkind, 2008). Based on the results at the =0.05
level of significance, the results does not provide
Pure-toneAudiogram-MeasuringAuditorySensitivityovertheAge
295
statistically significant evidence of a difference
between female and male in median of SPL at all
frequency tones used (p-value>0.05).
The Pearson's correlation was used to find a
correlation between the variables Age decades and
SPL threshold at the frequency tones. According to
the provided results it is possible to say that the
strength association between these two continuous
variables is very high and that the correlation
coefficient is very highly significantly different from
zero (p-value<0.05). For all results produced were
obtained positive coefficients of correlation which
indicate that for every frequency tones the older age
higher is the SPL threshold of hearing. Additionally,
as higher is the frequency tone higher is the Pearson
correlation meaning that for higher frequencies more
correlated are the decade ages with SPL threshold.
This last observation leads to the conclusion that the
lost of sensitivity along age is more effective for
higher frequency tones.
In order to complete the inferential analysis and
see if there is an association between the variables
Clinical Diagnosis of Hearing Pathology and Age
Decade Group (considering two groups: more than
50 years and less than or equal to 50 years), the
Fisher's exact test took place, recommended when
we have two nominal variables (Green and Salkind,
2008). The result indicated that the null hypothesis
can be rejected; so it was possible to say that the
variables are associated.
6 CONCLUSIONS
A pure tone computer software device to produce
auditory brainstem responses was developed and
used to measure the threshold of SPL perceived by
35 subjects in a scale a different frequency tones. A
small group of six individuals had a clinical
diagnosis of hearing pathology, but none of them use
any hearing device.
Despite the small number of individuals used in
the test it was clear that the lost of sensitivity was
higher in older individuals and became more
effective after the decade of 60 years old.
A statistical analysis using SPSS software was
performed. It may be concluded that no significant
differences exist in the frequency tones of 125 and
500 Hz, but there are significant differences in other
frequency tones (250 Hz, 1 kHz, 2 kHz, 3 kHz, 4
kHz, 6 kHz, 8 kHz, 10 kHz, 12 kHz and 15 kHz).
There are no statistically significant evidence of a
difference between female and male of SPL at all
frequency tones used. It was also concluded that
there is a high correlation between SPL threshold
and age within all frequency tones. Despite this
obvious conclusion it was confirmed for all
frequency tones. The statistical analysis also leads us
to the conclusion that as higher is the frequency tone
more correlation there is between sensitivity loss and
age.
Finally, the statistical analysis confirmed a high
correlation between subjects with clinical diagnosis
of hearing loss and the threshold of SPL.
REFERENCES
Acoustics 2003 - Normal equal-loudness-level contours -
ISO 226:2003 Acoustics. International Organization
for Standardization (ISO) 2nd edition.
Bento, D., Ferreira, S.; Magalhães, B., Rocha, D.,
Teixeira, J. P., 2010. Auditory system rehabilitation
available technologies. Biomedical Engineering and
Informatics, Vol.5, pp.1806-1810.
Canale Andrea, Lacilla Michelangelo, Cavalot Andrea
Luigi and Albera Roberto. 2006. Auditory steady-state
responses and clinical applications, European Archives
of Oto-Rhino-Laryngology, Volume 263, Number 6,
Pages 499-503.
Chen, Zhangli; Hu, Guangshu; Glasberg, Brian R.; Moore,
Brian C. J., 2011. A new model for calculating
auditory excitation patterns and loudness for cases of
cochlear hearing loss. Hearing Research, Vol.282(1),
pp.69-80.
Conijn E. A., Brocaar M. P., van Zanten G. A., 1990.
Frequency specificity of the auditory brainstem
response elicited by 1,000-Hz filtered clicks.
Audiology. 1990;29(4):181-95.
Fransen, Erik; Lemkens, Nele; Laer, Lut Van; Camp, Guy
Van., 2003. Age-related hearing impairment (ARHI):
environmental risk factors and genetic prospects.
Experimental Gerontology, 2003, Vol.38(4), pp.353-
359.
Green, S. B., & Salkind, N. J., 2008. Using SPSS for
Window and Macintosh: Analyzing and understanding
data. (5th ed.). Upper Saddle River, NJ: Pearson
Prentice Hall.
Stapells D. R., Oates P. 1997. Estimation of the pure-tone
audiogram by the auditory brainstem response: a
review. Audiol Neurootol. 2(5):257-80.
BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices
296