Noise Health Home 

[Download PDF]
Year : 2007  |  Volume : 9  |  Issue : 35  |  Page : 35--41

Hearing, communication and cognition in low-frequency noise from armoured vehicles

Ann Nakashima1, Sharon M Abel1, Matthew Duncan2, David Smith3,  
1 Individual Readiness Section, Defence Research and Development Canada - Toronto, Toronto, Canada
2 Collaborative Performance and Learning Section, Defence Research and Development Canada - Toronto, Toronto, Canada
3 Human Systems Integration Section, Defence Research and Development Canada - Toronto, Toronto, Canada

Correspondence Address:
Ann Nakashima
1133 Sheppard Ave West, Toronto, ON M3M 3B9


An experiment was performed to study auditory perception and cognitive function in the presence of low-frequency dominant armoured vehicle noise (LAV III). Thirty-six normal hearing subjects were assigned to one of three noise backgrounds: Quiet, pink noise and vehicle noise. The pink and vehicle noise were presented at 80 dBA. Each subject performed an auditory detection test, modified rhyme test (MRT) and cognitive test battery for three different ear conditions: Unoccluded and fitted with an active noise reduction (ANR) headset in passive and ANR modes. Auditory detection was measured at six 1/3 octave band frequencies from 0.25 to 8 kHz. The cognitive test battery consisted of two subjective questionnaires and five performance tasks. The earmuff, both in the conventional and ANR modes, did not significantly affect detection thresholds at any frequency in the pink and vehicle noise backgrounds. For the MRT, there were no significant differences between the speech levels required for 60% correct responses for three ear conditions in the pink and vehicle noise backgrounds. A small but significant (4 dB) increase in speech level was required in pink noise as compared to vehicle noise. For the serial reaction time task, the mean response time in the vehicle noise background (751 ms) was significantly higher than in pink noise and quiet (709 and 651 ms, respectively). The mean response time in the pink noise background was also significantly higher than in quiet. Thus, the presence of noise, especially low-frequency noise, had a negative effect on reaction time.

How to cite this article:
Nakashima A, Abel SM, Duncan M, Smith D. Hearing, communication and cognition in low-frequency noise from armoured vehicles.Noise Health 2007;9:35-41

How to cite this URL:
Nakashima A, Abel SM, Duncan M, Smith D. Hearing, communication and cognition in low-frequency noise from armoured vehicles. Noise Health [serial online] 2007 [cited 2022 Jul 2 ];9:35-41
Available from:

Full Text


Military and civilian workers may be exposed to high levels of low-frequency noise (LFN) in the range of 10-200 Hz. For example, the noise levels in Canadian Forces (CF) armoured vehicles (e.g., the Bison) driven at high speeds exceed 100 dB at frequencies below 63 Hz, while the levels between 63 and 200 Hz exceed 95 dB. [1] The communication headsets that are worn inside these vehicles provide passive noise attenuation and are generally poor at attenuating LFN. Alternative active noise reduction (ANR) technologies can increase the low-frequency attenuation of conventional earmuffs by as much as 20 dB. [2]

In the study cited above on noise levels in CF vehicles, [1] the subjects reported that they had difficulty in communicating in noisy environments. It is commonly believed that the use of hearing protective devices (HPDs) in noisy environments will interfere with both speech understanding and the perception of warning signals, thereby reducing situational awareness. Laboratory studies have shown that this is not the case for individuals with normal hearing. However, those who have pre-existing hearing loss may experience difficulty if the levels of target sounds are less than the sum of attenuation provided by the device and their raised hearing thresholds. [3],[4] Studies by Forshaw et al . [5] showed that the use of ANR hearing protectors improved signal detection during exposure to Sea King noise as compared to passive attenuation, but the effects of ANR on speech intelligibility were unclear. In later studies, ANR was shown to be beneficial for speech understanding in intermittent noise in hearing-impaired listeners. [6] Benefits for listening in LFN have not been documented.

Although a number of studies have examined the effects of noise on performance, [7],[8] few have specifically investigated the effects of LFN. Laboratory studies on stress levels and cognitive performance during exposure to infrasound have generally failed to show any effect. [9] Perrson Waye et al. [10] studied cognitive performance and subjective reaction to moderate levels of low- and mid-frequency ventilation noise. The subjects rated the interference with performance on the cognitive tests to be greater with the low-frequency than the mid-frequency noise. The reporting of physical symptoms, often associated with LFN exposure (e.g., headache, nausea, fatigue), was generally low for both types of noise. However, significantly lower social orientation (feeling more disagreeable, irritated, ill-tempered and less co-operative) was reported by subjects working in LFN. The performance results on the cognitive tests indicated that the subjects had more difficulty coping with cognitive demands in LFN. The decrease in performance became more pronounced with time, suggesting a fatigue effect. The small number of subjects tested and the use of moderate noise levels support the need for further research in this area.

The current experiment was designed to determine whether a low-frequency background noise recorded from a LAV III armoured vehicle would interfere with the detection of warning signals, speech understanding and cognitive function. The effect on performance of the vehicle noise was compared with the effects of broadband noise and quiet, with the ears unoccluded and protected with an ANR HPD, with and without ANR operational.

 Materials and Methods


Thirty-six subjects (military or civilian; 18 males and 18 females), aged 18-55 years and fluent in English, participated in the study. Each was screened for a history of ear disease, hearing thresholds greater than 25 dB HL bilaterally between 0.5 and 4 kHz, the use of medications that might interfere with concentration, and the ability to complete the experimental protocol.

Experimental protocol

The experiment was conducted in the Noise Simulation Facility at Defence Research and Development Canada (DRDC), Toronto. [11] The facility is a semi-reverberant room (11 x 6 x 3 m 3 ) with an ambient of about 28 dB SPL. The noise was presented from an array of eight sub-low, two low, four mid and four high-frequency loudspeakers powered by 12 Bryston amplifiers. The array spanned the shorter wall of the room, facing the subject at a distance of 4.35 m. The subject was placed at a position where the standing waves of the room were at a minimum to reduce the amplification of the background noise at 16 Hz.

Twelve subjects (six males and six females) were randomly assigned to three groups, aged-matched by decade. The first group performed the battery of tests in quiet, the second in continuous broadband (pink) noise at 80 dBA, and the third group in continuous low-frequency noise at 80 dBA. The LFN (hereafter referred to as vehicle noise) was recorded near the ear of the crew commander while standing with his upper body through the hatch of a LAV III. The vehicle was traveling on a highway at approximately 80 km/h. The noise spectra are shown in [Figure 1].

The subjects were tested with the ears unoccluded and while wearing a David Clark headset (H10-76XL) in both passive and ANR modes (referred to as "ANR Off" and "ANR On," respectively). In each of the three ear conditions, the subjects performed three types of tests: auditory detection, speech understanding and cognitive function. The orders of presentation of the three ear conditions, test types and tasks within the cognitive battery were independently counterbalanced across subjects in each group to equalize the effects of fatigue and practice.

Auditory detection was measured once for each of six one-third octave noise bands, centered at frequencies ranging from 0.25 to 8 kHz for each of the three ear conditions, using a variation of Bιkιsy tracking. [12] For each threshold determination, the stimulus was pulsed continuously at a rate of 2.5/s. The pulse duration was 250 ms including a rise/decay of 50 ms. The subjects were instructed to depress an On/Off push-button switch whenever the pulses were audible and to release the switch when they could no longer be heard. The sound level of consecutive pulses was increased in steps of 1 dB until the switch was depressed and then decreased at the same rate of change until the switch was released. The tracking trial was terminated after a minimum of eight alternative intensity excursions with a range of 4-20 dB. The detection threshold was defined as the average sound level (dB SPL) of eight final peaks and valleys.

Speech understanding was evaluated using the modified rhyme test (MRT) of consonant discrimination. [13] On each trial, the subjects were presented with an auditory carrier phrase ("Next word _______") and one of six possible words displayed on the computer monitor. The subject was instructed to depress one of six response keys whose spatial position on the response box corresponded to the spatial position on the screen of the word that he/she has heard. Across trials (word sets), the level of the carrier phrase remained fixed and clearly audible. However, the level of the target word changed in such a way to allow the estimation of the level at which 60% of the target words would be correctly recognized. The change in level over trials was controlled by a mathematic iteration procedure based on evidence that the function relating percent correct and level can be modeled by cumulative normal distribution function, with a standard deviation of 6 dB. In all, 50 sets of six words were available. Of these, half the sets contrasted the initial consonants ("raw, paw, law, jaw, thaw, saw") and half the final consonants ("teach, tear, tease, teal, team, teak"). For each trial, the program selected a set of six words at random. Pilot studies confirmed that a solution could be generated in less than 50 trials.

The cognitive test battery included a subset of tasks that have been used extensively in previous performance studies conducted at DRDC, Toronto. [8],[14],[15],[16],[17] The selected tasks investigated a diverse range of fundamental cognitive processes. They were presented to the subjects using a laptop computer. The battery comprised two subjective questionnaires that probed the subjects' level of mental and physical fatigue, motivation and mood. The subjects responded using a visual analog scale (VAS) from 1 to 10. In addition, there were five cognitive tests: short term memory (STM), four-choice serial reaction time (SRT), mental addition (MA), detection of repeated numbers (DRN) and a logical reasoning task (LRT). In the STM task, the subjects were presented with lists of five to nine numbers in sequence, with each number being presented for 500 ms, followed by a 500-ms pause. The subjects recalled the numbers in sequence immediately after the last number was presented. For the SRT task, one of four probe letters was presented on the screen (P, G, L and S), and the subject pressed the corresponding response button as quickly and accurately as possible. The MA task required the subjects to add a random sequence of eight numbers (between 1 and 16) that were presented one by one on the computer screen. In the DRN task, a series of three-digit numbers were flashed on the screen at a rate of 2/s. The subjects pressed a button when a three-digit number was repeated. For the LRT, a pair of letters was shown at the top of the screen (A B or B A) and a statement showing the spatial arrangement of letters was shown below, e.g., "B does not follow A." The subjects responded by indicating whether the statement was "true" or "false." A more detailed description of the tasks is presented in Abel et al . [8]


Analyses of variance (anova) [18] indicated that there were no significant differences in age or pure-tone hearing thresholds across the three groups and no significant differences between hearing thresholds for right and left ears within the subjects. The mean age across groups ranged from 25 to 28 years with standard deviations of 6-11 years. Averaged across groups and right and left ears, the mean hearing thresholds were -0.2, 0.5, -0.2 and 0.2 dB HL at 0.5, 1, 2 and 4 kHz, respectively, with standard deviations ranging from 4 to 8 dB.

The auditory detection results are shown in [Figure 2]. Detection thresholds (dB SPL) are plotted as a function of frequency with noise background (group) and ear condition as the parameters. In quiet, detection thresholds could not be obtained at 2 and 4 kHz in the unoccluded condition. Thresholds at these frequencies were below the range of the system. The anova applied to the data for 0.25, 0.5, 1 and 8 kHz indicated significant effects of background, ear condition, frequency, ear condition by background, frequency by background, ear condition by frequency, and ear condition by frequency by background ( P [18] showed that, averaged across ear conditions and frequencies, the mean detection threshold for quiet was significantly less by 34 dB ( P P P [19]

The levels required for 60% correct responses on the MRT are listed in [Table 1]. For this analysis, the 60% level rather than the 50% level (normally taken as the speech reception threshold) was selected since, in some conditions, the iterative process did not reach 50%. An anova applied to these data showed significant effects of group, ear condition and group by ear condition ( P P P P P P P P P P P P [20] Differences may be due to free-field presentation in a semi-reverberant listening environment. The increases in levels from 17 and 22 dB observed when listening with the muff in quiet with ANR Off and On, respectively, were within 6 dB of observed attenuation of the device in each of these modes at 1 kHz. In previous research, Smoorenburg [21] found that, in hearing-impaired subjects, speech reception thresholds in quiet were well predicted by hearing thresholds below 1 kHz. The use of hearing protection by normal hearing subjects may serve as an induced impairment. The results for MRT in pink noise showed that the speech level required for 60% correct understanding was not significantly different across the three ear conditions. The speech-to-noise ratio was roughly 8 dB. The same result was obtained in vehicle noise, although a small but significant (4 dB) increase in speech level was required for 60% correct understanding in pink noise as compared to vehicle noise. This indicated that vehicle noise had a weaker masking effect on speech and was expected since the energy at speech frequencies was comparatively lower.

With respect to the cognitive test battery, the data showed that the response to GVA question 4 ("How much of an effort is it to do anything?") for the ANR Off condition was significantly lower for the quiet background (3.7 2.0) than that for the vehicle background (5.8 2.5). However, it is not clear why this result was obtained only for the ANR Off condition. If the perceived effort level was increased by the presence of LFN, then a significant result should have also been obtained in the unoccluded condition. The results for the DRN task were also not consistent. If the presence of a noise background had a negative effect on the proportion of correct detections, then this result should have been evident for all the ear conditions, not just for ANR On. Thus, the significant outcome for this task is difficult to interpret and, indeed, may not be valid. This may indicate that DRN is not an appropriate task for this type of experiment. For the SRT task, the reaction time for the vehicle background was significantly higher than that for both quiet and pink noise backgrounds. However, given the overall pattern of results of cognitive tests, the test battery that was used may not be sufficient for detecting the effects of low frequency noise on cognitive performance.

This experiment was limited because the background noise had to be maintained not only at a safe level, but also at a level that would be considered tolerable to the test subjects. The actual recorded level of LAV noise was 91 dBA, which is allowed for unoccluded ears for an exposure time of up to 2 h, according to the Canada Labor code. [22] However, pilot testing indicated that, it would be difficult for the subjects to tolerate for the duration of the experiment. To resolve these issues, the level of vehicle and pink noise was thus lowered to 80 dBA. The results may have been different if it had been possible to use realistic background noise levels. Nonetheless, the results of this experiment have demonstrated that (1) the use of an HPD neither improved nor inhibited signal detection or speech understanding in a high-noise environment, regardless of the noise spectrum, and (2) the presence of LFN appears to increase reaction time and decrease vigilance. The use of an ANR HPD may help to improve vigilance in LFN.


The effects of low-frequency armoured vehicle noise (LAV III) and the use of different types of HPD's on hearing and cognitive performance have been investigated. While the noise levels used may not have been high enough to show the full effects of LFN on performance, the results indicated that LFN has a negative effect on reaction time. The use of an HPD neither helped nor hindered auditory detection or speech understanding in pink noise or vehicle noise.


The authors would like to thank Lauren Batho and Quan Lam for their assistance in running this study. This research was funded by the Military Operational Medicine Thrust (16ck02), Defence Research and Development Canada. The authors wish to thank those who graciously gave their time to participate as subjects[29].


1Nakashima AM, Borland MJ, Abel SM. Measurement of noise and vibration in Canadian forces armoured vehicles. Ind Health 2007;45:318-27.
2McKinley RL, Steuver JW, Nixon CW. Estimated reductions in noise-induced hearing loss by application of ANR headsets. In : Axelsson A, Borchgrevink HM, Hamernik RP, Hellsrom PA, Henderson D, Salvi RJ, editors. Scientific basis of noise-induced hearing loss. Thieme: New York; 1996. p. 347-60.
3Abel SM, Alberti PW, Haythornthwaite C, Riko K. Speech intelligibility in noise: Effects of fluency and hearing protector type. J Acoust Soc Am 1982;71:708-15.
4Abel SM, Kunov H, Pichora-Fuller MK, Alberti PW. Signal detection in industrial noise: Effects of noise exposure history, hearing loss and the use of ear protection. Scand Audiol 1985;14:161-73.
5Forshaw SE, Rylands JM, Crabtree RB. Evaluating the effectiveness of active noise reduction in flight helmets. DRDC Toronto Technical Report, 1988, DCIEM 88-RR-34.
6Abel SM, Spencer DL. Active noise reduction versus conventional hearing protection. Relative benefits for normal-hearing and impaired listeners. Scand Audiol 1997;26:155-67.
7Smith AP. A review of the effects on noise on human performance. Scand J Psychol 1989;30:185-206.
8Abel SM, Baranski JV, Crabtree RB, Smith D, Thompson M, Steeneken M, et al . The effects of habitual noise exposure on human performance in International Space Station operations. DRDC Toronto External Client Report, 2003, ECR 2003-004.
9Leventhall G. A review of published research on low frequency noise and its effects. Report for the Department for Environment, Food and Rural Affairs (Defra): UK; 2003.
10Persson Waye K, Rylander R, Benton S, Leventhall HG. Effects on performance and work quality due to low frequency ventilation noise. J Sound Vib 1997;205:467-74.
11Nakashima AM, Borland MJ. The noise simulation facility at DRDC Toronto. DRDC Toronto Technical Report, 2005, TR 2005-095.
12Yantis PA. Puretone air-conduction testing. In : Katz J, editor. Handbook of clinical audiology, 3 rd edn. Williams and Wilkins: Baltimore; 1985. p. 153-9.
13House AS, Williams CE, Hecker MH, Kryter KD. Articulation-testing methods: Consonantal differentiation with a closed-response set. J Acoust Soc Am 1965;37:158-66.
14Baranski JV, Gil V, McLellan T, Moroz D, Buguet A, Radomski M. Effects of modafinil on cognitive performance during 40 hours of sleep deprivation in a warm environment. Mil Psychol 2002;14:23-48.
15Angus RG, Heslegrave RJ. Effects of sleep loss on sustained cognitive performance during a command and control simulation. Behav Res Methods Instrum Comput 1985;17:55-67.
16Baranski JV, Pigeau R, Angus R. On the ability to self-monitor cognitive performance during sleep deprivation: A calibration study. J Sleep Res 1994;3:36-44.
17Pigeau R, Naitoh P, Buguet A, McCann C, Baranski J, Taylor M, et al . Modafinil, d-amphetamine and placebo during 64 hours of sustained mental work I: Effects on mood, fatigue, cognitive performance and body temperature. J Sleep Res 1995;4:212-28.
18Daniel WW. Biostatistics: A foundation for analysis in the health sciences. Wiley: New York; 1983.
19Abel SM, Armstrong NM, Giguθre C. Auditory perception with level-dependent hearing protectors. Scand Audiol 1993;22:71-85.
20Abel SM, Krever EM, Alberti PW. Auditory detection, discrimination and speech processing in ageing, noise-sensitive and hearing-impaired listeners. Scand Audiol 1990;19:43-54.
21Smoorenburg GF. Speech reception in quiet and in noisy conditions by individuals with noise-induced hearing loss in relation to their tone audiogram. J Acoust Soc Am 1992;91:421-37.
22MOL. Canada Labour Code, Part VII, Levels of sound. Canada Occupational and Safety Regulations, Ministry of Labour: Ottawa, ON; 1991.
23Abel SM, Sass-Kortsak A, Kielar A. The effect on earmuff attenuation of other safety gear worn in combination. Noise Health 2002;5:1-13.
24Bento Coelho JL, Ferreira A, Serrano J, Castelo Branco NA. Noise assessment during aircraft run-up procedures. Aviat Space Environ Med 1999;70:A22-6.
25Berglund B, Lindvall T. Community noise. Arch Cent Sens Res 1995;2:1-195.
26Casali JG, Robinson GS, Dabney EC, Gauger D. Effect of electronic ANR and conventional hearing protectors on vehicle backup alarm detection in noise. Hum Factors 2004;46:1-10.
27CSA. Hearing protection devices - performance, selection, care and use. Canadian Standards Association: Rexdale, ON; 2002. CSA Z94.2-02.
28Mills JH, Osguthorpe JD, Burdick CK, Patterson JH, Mozo B. Temporary threshold shifts produced by exposure to low-frequency noises. J Acoust Soc Am 1983;73:918-23.
29Tempest W. The noise handbook. Academic: London; 1985.