Abel & Lam (2004)
|
Sixteen adults ages from 21 to 53 years (8 men and 8 women) with normal hearing.
|
The Indoor and Outdoor E-A-R plugs manufactured by the Aero Company were used. The hearing thresholds were measured on nine one-third-octave noise bands (i.e., 0.125, 0,25, 0,5, 1, 2, 3.15, 4, 6.3, and 8 kHz) for three conditions (i.e., unoccluded, indoor plug, and outdoor plug)
|
Repeated measures
|
Attenuation
|
Amount of attenuation (dB).
|
The results of ANOVA with repeated measures showed significant effects of ear condition, frequency, and ear by frequency (p < 0.0001). Post hoc comparison showed that the differences in attenuation for the two devices were significant for all testing frequencies (p < 0.001 or better).
|
The attenuation values in the indoor plug increased from 21 dB at 0.125 kHz to 40 dB at 8 kHz. Also, the results for the outdoor plug attenuated the thresholds by 5 and 14 dB at 0.5 kHz and 1 kHz.
|
Abel et al. (2002)
|
Twenty-four young adults with normal hearing (age under 40 years), twenty-four older adults with normal hearing (age over 40 years), and twenty-four older adults with hearing loss
|
A total of five ear condition were adjusted: 1) ears with unoccluded, 2) ears with Class A muffs, 3) ears with muffs on hard hat and air-purifying half-mask respirators 4) ears with muffs on hard hat and safety glasses, and 5) ears with muffs on hard hat, safety glasses, and respirators. Attenuation measurements were conducted for eight one-third octave noise band centered from 0.25 to 8 kHz.
|
Repeated measures
|
Attenuation
|
Amount of attenuation (dB SPL)
|
The results of ANOVA with repeated measures showed significant effects for group [F(5,61) = 4.3, p < 0.002], protector condition [F(3,183) = 104.2, p < 0.0001], and frequency [F(7,427) = 387.7, p < 0.0001]. The interaction of the protector condition by group [F(35,427) = 1.8, p < 0.006], protector condition by group [F(21,1281) = 12.2, p < 0.0001], protector condition by frequency by group [F(105,1281) = 6.8, p < 0.04]. The interaction with protector condition by group was not significant.
|
Averaged across protector conditions, attenuation showed significant increases at frequency increased from 0.25 to 1 kHz and then remained constant, except for a dip at 6.3 kHz. Averaged across groups and frequencies, the least attenuation was achieved with the muff on hard hat in combination with the glasses and respirator. The muff on hard hat alone condition showed highest attenuation. The range in attenuation in order to the conditions was greater at 0.25 and 0.5 kHz (9 dB) and least at 2 and 3.15 kHz (3–4 dB).
|
Casali et al. (2004)
|
Ten adults with normal hearing age ranging from 18 to 49 years.
|
The signal stimulus was a standard vehicle back-up alarm at a presentation level of 85 and 100 dBA. The experimental conditions consisted of four different ear conditions (i.e., earplug, passive earmuff, ANR earmuff, and unoccluded ear).
|
Repeated measures
|
Attenuation
|
Mean masked thresholds (dBA)
|
In comparisons of HPD analyzed by ANOVA with repeated measures, interactions included ear condition ⅹ noise level [F(2,18) = 19.38, p = 0.0001], ear condition noise spectrum [F(2,18) = 8.51, p = 0.0053] were significant.
In comparisons of HPDs and an unoccluded condition, there were no statistically significant interactions for ear condition ⅹ noise spectrum. However, the main effects of the ear condition at 85 dBA [F(3,27) = 7.91, p = 0.0006] was significant.
|
Given 100 dBA noise, earplugs showed significantly lower masked thresholds (91.9 dBA) than either the ANR muff (93.8 dBA) or the passive earmuff (95.5 dBA). Also, the ANR muff produced significantly lower masked thresholds than the passive muff did.
|
de Faria & Suzuki (2008)
|
Thirty adults with normal hearing (age rang 20 to 58 years).
|
Two HPDs with 21 and 17 dB NRR, respectively, were used to measure attenuation.
|
Repeated measures
|
Attenuation
|
Noise attenuation (decibels)
|
The mean noise attenuation for 60 ears at all frequencies was lower than the manufacturer’s rating. No ear achieved the expected attenuation, except at 500 Hz (35 dBNA).
|
The observed noise attenuation measurements at each frequency were : 1) 500 Hz (mean: 22.8 dB, SD: 3.8), 2) 1000 Hz (mean: 23.4 dB, SD: 3.7), 3) 2000 Hz (mean: 27.3 dB, SD: 4.5), and 4) 4000 Hz (mean: 29.4 dB, SD: 4.3).
|
Tufts et al. (2012)
|
Thirty adults with normal hearing (mean age: 21 years, SD: 2.7).
|
Two types of HPDs were used (i.e., custom-molded and non-custom earplugs). For five measurements, the results of trained and untrained group were compared.
|
Repeated measures
|
Attenuation
|
Mean per-frequency attenuation (dB)
|
Repeated-measures of ANOVA conducted on mean PAR showed the training to be statistically significant [F(1,29) = 11.77, p = 0.002], but neither earplug type [F(1,29) = 1,91, p = 0.178] nor the interaction of earplug type ⅹ training [F(1,29) = 0.29, p = 0.596] was statistically significant.
|
Mean per-frequency attenuation and mean PAR for each earplug type and training condition showed that the mean PAR of untrained non-custom plug produced 3.2 dB less attenuation than the trained non-custom plug condition. For the custom plug, the untrained group showed 4.1 dB less mean PAR than the trained group. For the frequency comparison, the trained group showed higher attenuation values than the untrained group did for both custom and non-custom plugs.
|
Abel & Paik (2005)
|
Twelve young adults (age rang 18 to 30 years) and twelve older adults (age rang 40 to 55 years) with normal hearing.
|
Three ear conditions (i.e., with and without ANR earmuff operational and unoccluded ear) was conducted. A total of eight speakers formed eight azimuth angles (i.e., 15, 75, 105, 165, 195, 255, 285, and 345 degrees). For the sound source identification task, three different signals, such as 0.5 kHz, 4 kHz, and broadband noise, were presented.
|
Repeated measures
|
Sound localization
|
Response time (ms) for sound source identification
|
An ANOVA indicated that there were significant effects for ear condition [F(2,40) = 5.6, p < 0.007] and stimulus frequency [F(2,40) = 9.0, p < 0.001].
|
Averaged across the four groups and three stimuli, the mean median response times ranged from 712 ms for ANR On to 805 ms in the unoccluded condition. Averaged across the four groups and three ear conditions, the mean median response times ranged from 667 ms for the broadband noise to 803 ms for the 0.5 kHz stimulus.
|
Alali&Casali (2011)
|
Twelve adults (older than 18 years) with normal hearing.
|
Within-subject design (8 HPDs x 2 signals x 2 presentation levels) were adjusted. The HPDs consisted of: 1) unoccluded, 2) foam earplug, 3) pre-molded earplug, 4) flat attenuation HPD, 5) level-dependent HPD, 6) passive earmuff, 7) dichotic sound transmission earmuff, and 8) custom-made diotic sound transmission earmuff. The signals consisted of standard backup alarm and spectrally-modified one. Presentation levels of pink noise were both 60 dBA and 90 dBA.
|
Repeated measures
|
Sound localization
|
Percentage correct (%) of sound identification.
|
The ANOVA revealed the significant main effects of HPD for a percentage correct localization [F(7,77) = 59.2, p < 0.0001], percentage of right-left localization errors [F(7,77) = 42.78, p < 0.0001], percentage of front-rear localization errors [F(7,77) = 43.03, p < 0.0001], and localization absolute deviation in degrees [F(7,77) = 92.67, p < 0.0001]. The interaction between HPD and background noise level on percentage correct localization response also significant [F(7,77) = 9.51, p < 0.0001].
|
The mean values of percent correct localization showed that flat attenuation HPD (83.9%) was the highest percent correct. Unoccluded (82.2%), pre-molded earplug (81.5%), and passive earmuff followed, but were not significant. The custom-made diotic sound transmission earmuff (66.3%) and foam earplug (64.6%) showed similar percentage correct. The dichotic sound transmission earmuff (15.8%) showed the lowest percentage correct for localization.
|
Bolia et al. (2001)
|
A total of six adults age 18 to 34 years with normal hearing.
|
Three ear conditions such as unoccluded, earplugs and earmuff and three source elevation conditions (i.e., upper hemisphere, peri-hemisphere, and lower hemisphere) were adjusted.
|
Repeated measures
|
Sound localization
|
Azimuth error (°), front-back confusions (%), and elevation error(°)
|
Repeated ANOVA revealed significant main effects of hearing protector [F(2,10) = 29.94, p < 0.05], and source elevation [F(2,10) = 19.96, p < 0.05]. Also, the interaction between hearing protector and source elevation [F(4,20) = 10.86, p < 0.05] was significant.
|
While the earplugs and earmuffs condition showed similar mean azimuth error, the unoccluded condition showed lower mean azimuth error for all source elevation conditions. In addition, for the front-back confusions and elevation error, the unoccluded condition showed significantly lower confusion and errors than the other HPDs conditions did.
|
Brown et al. (2015)
|
Ten normal hearing adults (mean age: 29.5 years).
|
Four HPDs and one control (unoccluded) were used in the experiment: 1) passive HPD, 2) active HPD, 3) hybrid HPD, and 4) ShotShields HPD.
|
Repeated measures
|
Sound localization
|
Response angle (°)
|
The ANOVA indicated a significant main effect of device condition [F(4,36) = 13.64, p < 0.001] and target angle [F(11,99) = 5.67, p < 0.001]. A significant interaction between device and target angle [F(44,396) = 5.67, p < 0.001] was revealed.
|
All tested HPDs significantly degraded localization performance relative to control (unoccluded) condition. However, post-hoc analysis of HPD-related distortions of sound source localization cues demonstrated that the ShotShields produced both the best behavioral performance and the greatest preservation of sound source cues.
|
Carmichel et al. (2007)
|
Eight normal hearing listeners (age 20 to 40 years).
|
Four stimuli (i.e., two successive transient clicks of a semiautomatic handgun being loaded, the double ring of a telephone with a mechanical ringer, and electronically generated FM tone bursts at 0.5 and 4 kHz) and four HPDs (i.e., electronic earmuffs, dynamic level compression protectors, action ear sport, and unoccluded) were used.
|
Repeated measures
|
Sound localization
|
Localization performance (%) and response time (sec)
|
Differences in localization among the conditions were significant for the firearm [F(3,21) = 15.30, p < 0.05], telephone [F(3,21) = 15.40, p < 0.05], 4 kHz tone burst [F(3,21) = 1.13, p < 0.05] stimuli, but not for the 0.5 kHz tone burst [F(3,21) = 1.13, p = 0.36]. The mean response time showed that there were differences in the conditions for the broadband stimuli (firearm, [F(3,21) = 6.15, p < 0.05], telephone [F(3,21) = 5.21, p < 0.05].
|
The mean correct responses for localization to broadband stimuli in the unoccluded condition were 97.2% (firearm) and 98.6% (telephone). Mean scores for localization to these stimuli ranged from 67.4 to 54.2% correct in the HPD conditions when averaged across all locations. When averaged across all stimuli and locations, the mean response time for correct responses in the unoccluded condition (mean: 1.58 s, SD: 0.76) was not different from the mean response time to errors (mean: 1.64 s, SD: 0.81). In the HPD conditions, the mean response time to incorrect response was less (mean: 1.88 s, SD: 0.89) than to correct responses (mean: 1.90 s, SD: 0.81) but these differences were not significant.
|
Simpson et al. (2005)
|
Seven adults (age 18 to 39 years) with normal hearing.
|
Four ear conditions (i.e., no HPD (unoccluded), foam earplugs, earmuffs, combination of earplugs and earmuffs) and five auditory cue conditions (four HPDs condition and no cue condition) were used. Stimuli were visual targets with continuous broadband (70 Hz to 16 kHz) pink noise.
|
Repeated measures
|
Sound localization
|
Response accuracy (%) and search time (sec)
|
For response accuracy, repeated measures ANOVA revealed that neither of the main effects nor the interaction was found to be statistically significant (p > 0.05). However, for search time, repeated ANOVA showed significant main effects on the auditory condition [F(4,24) = 239.11, p < 0.05] and set size [F(2,12) = 328.75, p < 0.05]. The interaction between auditory condition and set size [F(8,48) = 93.84, p < 0.05] was significant.
|
The unoccluded condition showed lowest search time, regardless of the set size. Earmuffs and earplugs condition showed similar search time as a function of set size. The combination of earplugs and earmuffs showed higher search time than for other ear conditions.
|
Talcott et al. (2012)
|
Thirteen adults age 22 to 54 years with normal hearing.
|
Five ear conditions (i.e., unoccluded, earplug with 21 dB NRR and gain switch, ear tips with 21 dB NRR, earplugs with 7 dB NRR, earmuff with 21 dB NRR) and two noise conditions (i.e., 45–50 dBA ambient noise and 82 dBA diesel truck noise) were used.
|
Repeated measures
|
Sound localization
|
Percent correct response (%) and mean response time (sec)
|
The ANOVA showed a significant main effect of listening condition (F = 17.22, p < 0.0001). For the mean response time, the results of ANOVA showed that there was a significant main effect of the listening condition (F = 11.11, p < 0.0001).
|
The percent of correct response was significantly lower for the earmuff (21 ± 10%) than all listening conditions and significantly greater for the unoccluded ear (55 ± 17%) than all listening conditions. No other differences were significant between the HPD conditions. In addition, the mean response time was significantly higher for the earmuff condition (2.9 ± 1.4 s) than for all other HPDs and the unoccluded condition.
|
Zimpfer & Sarafian (2014)
|
Twenty listeners (age 24–51 years) with normal hearing.
|
Five HPDs (four earplugs and one earmuff) were used where two were passive HPD and three were active HPD .
|
Repeated measures
|
Sound localization
|
Number of correct responses and number of confusions
|
Two-way repeated-measure ANOVA revealed the significant main effect of the test condition factor [F(2.9,10.6) = 68.33, p < 0.001].
|
The unoccluded condition showed a significantly higher number of correct responses than any other HPDs conditions. Moreover, the active systems yielded lower scores (53 and 40%) than did the passive systems (63%).
|
Byrne & Palmer (2012)
|
Fifteen adults with normal hearing (age range 21 to 60 years).
|
In the four testing conditions, two types of HPDs were used, an electronic hearing protector and conventional passive earmuff. The electronic hearing protector had three different settings (i.e., off, low, and high).
|
Repeated measures
|
Speech intelligibility
|
Correct percentage of HINT test score (%)
|
The results of repeated-measures ANOVA showed a significant main effect [F(2,28) = 1014.50, p < 0.0001] for the SNR condition. The highest scores were obtained at + 5 dB SNR condition, while the lowest scores were obtained at a -5 dB SNR condition. A significant main effect also was found for the earmuff condition [F(3,42) = 57.19, p < 0.0001]. Additionally, the interaction effect was significant [F(6,84) = 6.94, p < 0.0001].
|
For the types of HPDs, passive muff showed highest correct percent for the all SNR condition. While the off and low settings of electronic hearing protector showed similar correct percentage, and the high setting of hearing protector showed lowest correct percentage, regardless of the SNRs.
|
Dastpaak et al. (2019)
|
Thirty-two adults (mean age 26.12 years) with normal hearing.
|
Three ear conditions (unoccluded, HPD with 25 NRR, and HPD with 32 NRR) and three SNR condition (-10, 0, and + 10 dB) were used.
|
Repeated measures
|
Speech intelligibility
|
Averaged intelligibility score (%)
|
The results showed maximum speech intelligibility without HPDs and without noise and minimum average of speech intelligibility when the background noise was greater than the − 10 dB SNR.
|
For the comparisons of types of HPDs, 32 of NRR HPD showed lower averaged intelligibility scores in all testing SNR conditions. While the without HPD condition showed 8.13% (SD: 9.11), 35.44% (SD: 19.11), and 74.94% (SD: 7.67) for three SNR conditions, HPD with 32 NRR showed 11.38% (SD: 9.64), 43.25% (SD: 17.54), and 78.06% (SD: 9.57). The HPD with 25 NRR showed highest average intelligibility score for − 10 dB SNR (mean: 22.75%, SD: 12.04), 0 dB SNR (mean: 55.31%, SD: 14.41), and − 10 dB SNR (mean: 78.38%, SD: 13.39).
|
Giguère et al. (2015)
|
A total of forty-five adults (age of 23 to 81 years) participated. There were four groups based on t hearing loss: 1) normal hearing (n = 12), 2) slight-to-mild (n = 12), 3) mild-to-moderate (n = 12), and 4) moderate/severe (n = 9).
|
Three ear conditions were used (unoccluded and two HPDs). In the HPDs, one was the analogue device, and the other was a digital device. Two noises (95.3 dBA and 89.5 dBA) were used.
|
Repeated measures
|
Speech intelligibility
|
Speech recognition (% word recognition)
|
In a comparison of protected versus unprotected speech recognition scores by hearing loss, use of the higher level-dependent function almost fully restored speech recognition to unprotected values or provided significant improvements over an unprotected performance for all participants.
|
The passive HPD showed a mean decrease in performance of 27–29% across the noises compare to the unoccluded condition. However, when the level-dependent function was activated, it yielded an average overall benefit of 11–15% and 23–24% across the two noises, respectively. The active HPD showed a mean decrement in performance of 17–25% compared to unprotected listening across the two noises. Unlikely for the passive HPD, anactive HPD with activation of level-dependent function showed only a small effect on performance and a mean benefit of 6–7%.
|
Manning et al. (2016)
|
A total of 47 adults (19 with normal hearing, 15 with SNHL, and 13 with SNHL and tinnitus).
|
A total of 300 items consisting of six lists was used. The background noise was presented in three different SNR conditions (i.e., quiet, -6 and − 12 dB). Two types of TCAPS (i.e., in-ear headset for air-conduction, and headset for bone-conduction) and two types of talker (male and female) were used.
|
Repeated measures
|
Speech intelligibility
|
Mean rationalized arcsine units (corrected)
|
The mean rationalized arcsine units measured for each of the TCAPS under test were marginally, but significantly better for the bone-conduction headset [F(1,528) = 29.90, p < 0.01]. For the types of hearing loss, the effects of hearing loss were found to be statistically significant [F(2,528) = 3.74, p < 0.024]. There was no significant interaction of hearing loss with TCAPS [F(2,528) = 0.99, p = 0.372]. In the SNR conditions, performance increased, as the SNR increased [F(2,528) = 1881.47, p < 0.01].
|
The normal hearing listeners performed best,and SNHL listeners performed worst. The degrading effect of noise on speech recognition was approximately the same for the normal hearing and hearing impaired listeners; there was no measurable benefit for either TCAPS in noise.
|
Plyler & Klump
(2003)
|
Fourteen females (age 21 to 24 years) with normal hearing.
|
A total of three ear conditions including unoccluded were used. In the HPDs, no (?) HPDs which was custom acoustic HPD and custom electronic HPD were used. For the speech perception test, two levels of noise (75 and 90 dB SPL) were used. The HINT sentence lists were used as the stimuli.
|
Repeated measures
|
Speech intelligibility
|
Relative HINT score
|
The analysis revealed a significant main effect for HPD [F(1,13) = 10, p < 0.05]. However, main effects for sentence presentation level [F(1,13) = 1, p > 0.05] were significant. The interaction between sentence presentation level ⅹ HPD [F(1,13) = 2, p > 0.05] was not significant. These results indicated that communication during noise was significantly better when utilizing the acoustic HPD than when utilizing the electronic HPD at each sentence presentation level.
|
The averaged relative HINT scores indicated that the acoustic HPD showed 1.1 (SD: 1.9) relative HINT score which was higher than the electronic HPD (mean: -0.4, SD: 1.2) at 75 dB SPL. In the 90 dB SPL condition, acoustic HPD (mean: 0.1, SD: 1.1) also showed higher relative HINT score than that for the electronic HPD (mean: -0.6, SD: 1.3).
|
Smalt et al. (2019)
|
Thirteen adults average age of 31 years with normal hearing, except for one subject which showed 30 dB HL in one ear.
|
A total of four HPDs, two passive types and two active types, were used. For the speech intelligibility test, Modified Rhyme Test using three background noises (i.e., 60, 75, and 80 dBA) was used.
|
Repeated measures
|
Speech intelligibility
|
Percentage correct (%)
|
A two-way ANOVA demonstrated that there were significant main effects for HPD [F(4,48) = 3.716, p = 0.010] and noise level [F(2,24) = 1737, p < 2 ⅹ 10–16. However, the interaction between HPD and noise level was not significant. For the results of the post-hoc test, only the HPD pair showed a significant difference between active HPDs (p = 0.017).
|
Speech intelligibility was somewhat reduced for all the hearing protectors, but the differences induced by increasing the background noise level were much greater than the differences among the HPDs for a single noise level.
|
Tufts & Frank
(2003)
|
Thirty-two adults (mean: 25.4 years, SD: 4.9) with normal hearing.
|
Two types of HPDs (i.e., custom-molded foam ear plug and pre-molded earplug) were used. As stimuli, the 12 passages from the 20 passages of the Speech Intelligibility Rating test were chosen.
|
Repeated measures
|
Speech intelligibility
|
Signal-to-noise ratio (dB)
|
In the SNR analysis, the main effects of background noise [F(4,112) = 3041.646, p < 0.001] and ear condition [F(1,28) = 4965.188, p < 0.001] were significant. Further, the interactions of background noise ⅹ ear condition [F(4,112) = 60.568, p < 0.001] and background noise ⅹ ear condition ⅹ sex [F(4,112) = 2.776, p = 0.034] were significant.
|
For background noise at 100 dB SPL, the overall speech levels were 84.4 dB for an unoccluded condition, and 71.9 dB and 74.3 dB for the custom-molded and pre-molded ear plugs. The corresponding SNRs were − 15.6, -28.1, and − 25.7 dB.
|