Noise in Industry - Fundamentals and Assessment Tools for Occupational Ergonomics - page 613

31.3 Effects of Noise in Occupational Settings

31.3.1 Nonauditory Effects
Noise exposure in industry has been linked to several deleterious effects, some of which are nonauditory
and thus beyond the scope of this chapter. However, it is at least important to recognize that noise can
degrade operator task performance. Research studies concerning the effects of noise on performance are
primarily laboratory based and task noise specific; therefore, extrapolation of the results to actual indus-

trial settings is somewhat risky. 5 Nonetheless, on the negative side, noise is known to mask task-related
acoustic cues, as well as cause distraction and disruption of “inner speech,” while on the positive side,
noise may at least initially heighten operator arousal and thereby improve performance on tasks that
do not require substantial cognitive processing. 6 To obtain reliable effects of noise on performance,
except on tasks that rely heavily on short-term memory, the level of noise must be fairly high, usually
95 dBA or greater. Tasks that are simple and repetitive often show no deleterious performance effects
(and sometimes improvements) in the presence of noise, while difficult tasks that rely on perception
and information processing on the part of the operator will often exhibit performance degradation. 5
It is generally accepted that unexpected or aperiodic noise causes greater degradation than predictable,
periodic, or continuous noise, and that the startle response created by sudden noise can be disruptive.
Furthermore, noise has been linked to physiological problems other than those of the hearing organ,
including hypertension, heart irregularities, extreme fatigue, and digestive disorders. Most physiological
responses of this nature are symptomatic of stress-related disorders. Because the presence of high noise
levels often induces other stressful feelings (such as sleep disturbance and interference with conversing in
the home, and fear of missing oncoming vehicles or warning signals on the job), there are second-order
effects of noise on physiological functioning that are difficult to predict.

/

31.3.2 Signal Detection and Communications Effects
31.3.2.1 Interference and the Signal-to-Noise Ratio
One of the most noticeable effects of noise is its interference with speech communications and the
hearing of nonverbal signals. Workers often complain that they must shout to be heard and that they
cannot hear others trying to communicate with them. Likewise, noise interferes with the detection of
workplace signals, such as alarms for general area evacuation and warnings, annunciators, on-equipment
alarms, and machine-related sounds, which are relied upon for feedback. The ratio (actually the algebraic
difference) of the speech or signal level to the noise level, termed the signal-to-noise ratio (S

N) is the
most critical parameter in determining whether speech or signals will be heard in noise. An S

/

Nof

/

5 dB means that the signal is 5 dB greater than the noise, while a S 5 dB means that the signal

is 5 dB lower than the noise. Hearing protection is often blamed for exacerbating the effects of noise
on the audibility of speech and signals, although, at least for individuals with normal hearing, protectors
may actually facilitate hearing in some noisy situations, particularly those above about 90 dBA.

Nof

/ 2

31.3.2.2 Masking
Masking is technically defined as the tendency for the threshold of a desired signal or speech (the masked
sound) to be raised in the presence of an interfering sound (the masker). As an example, in a noisy airport
waiting area, a pay telephone's earphone volume must often be increased to enable the listener to hear the
party on the line, whereas a lower volume will be more comfortable while affording audibility when there
is no crowd or public address system noise present. The masked threshold is defined as the SPL required
for 75% correct detection of a signal when that signal is presented in a two-interval task wherein, on a
random basis, one of the two intervals of each task trial contains the signal and the noise and the other
contains only noise. In a controlled laboratory test scenario, a signal that is about 6 dB above the masked
threshold will result in near perfect detection performance. 7 Analytical prediction (as opposed to actual
experimentation with human subjects) of the interfering effects of noise on speech communications may