Sound Affects

Disruption of office-related tasks by speech and office noise
Banbury & Berry, British Journal of Psychology, 1998

“This series of experiments has shown that the irrelevant speech effect can be extended to more complex cognitive tasks. The experiments have consistently shown that performance on memory for prose and mental arithmetic tasks can be reduced by exposure to background irrelevant speech. It has also been shown that office noise, with or without speech, can disrupt performance on on these tasks. The magnitude of disruption is fairly impressive. In Expt 2, for example, when the background noise was played in both learning and recall phases, performance in the presence of noise (irrespective of the meaning) was reduced to about one-third of third level of that in quiet.”

BACK

Reducing noise distractions to increase worker productivity
By Elaine Lewis, Peter Lemieux, David Sykes, and Thomas Horrall

Independent research has documented productivity gains of 8 to 38 percent, job satisfaction increases of 125 to 174 percent, and reductions in stress up to 27 percent.

Bosses have not spotted the biggest productivity inhibitor ASID

A recent study completed for the American Society of Interior Designers by the Yankelovich Partners found "noise reduction" to be a major concern of office workers. In fact, 70 percent of the respondents indicated their productivity would increase if their offices were less noisy. A follow-up study of business executives, conducted for ASID by L.C. Williams and Associates, revealed that business executives were largely unaware of noise problems in the workplace and, in fact, 81 percent were unconcerned about office noises! Clearly, business decision makers in today's workplace need to better understand the importance of noise and its relationship to productivity. It's a problem that directly affects their "bottom line."

Stress and open-office noise.
by Evans, Gary W.; Johnson, Dana
from Journal of Applied Psychology. 2000 Oct Vol 85(5) 779-783

Forty female clerical workers were randomly assigned to a control condition or to 3-hr exposure to low-intensity noise designed to simulate typical open-office noise levels. The simulated open-office noise elevated workers' urinary epinephrine levels, but not their norepinephrine or cortisol levels, and it produced behavioral after-effects (fewer attempts at unsolvable puzzles) indicative of motivational deficits. Participants were also less likely to make ergonomic, postural adjustments in their computer work station while working under noisy, relative to quiet, conditions. Postural invariance is a risk factor for musculoskeletal disorder. Although participants in the noise condition perceived their work setting as significantly noisier than those working under quiet conditions did, the groups did not differ in perceived stress.

BACK

Noise and its effects (Administrative Conference of the United States, Alice Suter, 1991)

Extracts from section D: Effects on Performance and Behavior

Noise can cause adverse effects on task performance and behavior at work, and in nonoccupational and social settings. These effects are the subject of some controversy, however, since they do not always occur as predicted. Sometimes noise actually improves performance, and sometimes there are no measurable differences between performance in noisy and quiet conditions. The presence and degree of effects depends on a variety of intervening variables.

2. Noise variables

Sound level is one of the most important parameters when predicting performance effects. The level of noise necessary to produce adverse effects is greatly dependent upon the type of task. Simple tasks remain unaffected at noise levels as high as 115 dB or above, while more complex tasks are disrupted at much lower levels. Until fairly recently, the level of beginning effects was thought to be around 95 dB for most conditions, but a summary of recent research (Jones, 1990) points to effects at much lower levels. Effects on serial reaction tasks have been noted for continuous noise with C- weighted noise levels of 90 dB (Jones, 1983) and for intermittent noise with C-weighted levels of around 80 dB (Lahtela et al., 1986).

Frequency and temporal characteristics also play a part. High-frequency sound is more disruptive titan low-frequency sound, and intermittent noise can affect performance more adversely titan continuous noise of equivalent energy. Aperiodic intermittencies are more likely to produce adverse effects than regular ones, and impulse noise may be even more disruptive. Again the effects are variable, depending upon task complexity and other factors.

Much of the important research in the effects of noise on performance conducted over the last decade has focused on the effects of irrelevant speech. (17) The adverse effects of irrelevant speech appear to be fairly independent of sound level, at least in the 55-95 dB range, and therefore, are not mitigated simply by attenuating them by 10 dB or so (Jones, 1990). It also appears that irrelevant speech affects processes involving memory (e.g., reasoning, mental arithmetic, and problem solving) rather than attention. With respect to reading tasks, however, meaningful speech is more disruptive than meaningless speech (Jones, 1990). These findings have significance for many modern work and school environments, where information processing and exchange is so important, especially those of the "open plan" variety.

3. Task variables

Task complexity has been identified in numerous experiments as a crucial determinant of the effects of noise on performance. Noise exposure usually leaves simple routine tasks unaffected, and can even improve performance of monotonous tasks, presumably by elevating one's level of arousal (Broadbent, 1971). Some tasks, such as tracking and jobs requiring intellectual function, can be momentarily disrupted without decrements in overall performance (Broadbent, 1979). But if the noise level is sufficiently high or if the task becomes more complex, noise will have an adverse effect. When two or more tasks must be performed simultaneously in a noisy environment, performance on the primary task usually remains unaffected, while performance on the subsidiary task deteriorates (Hockey and Hamilton, 1970; Davies and Jones, 1975; Finkleman and Glass, 1970).

4. After-effects

It seems that noise can have even greater effects after than during exposure. The most common after-effect appearing in the experimental literature is a reduced tolerance for frustration, manifested in a series of experiments as a reduction in willingness to persist in trying to solve insoluble puzzles (Glass and Singer, 1972; Percival and Loeb, 1980). This research also indicates that predictability of the noise signal greatly reduces its adverse after-effects (Glass and Singer, 1972). One study found that the type of noise also influenced the after-effect. Aircraft noise modified to produce sudden onsets and offsets resulted in a lower tolerance for frustration after the exposure than white noise that ad been similarly modified (Percival and Loeb, 1980).

5. Effects of noise on social behavior

There is an extensive literature concerning the effect of noise on social behavior, and just a few examples of this research will be discussed here. Singer et al. (1990) point out that noise has been used as a noxious stimulus in a variety of investigations because it produces the same biological and psychological effects as other stressors. In fact, they observe that the effects of noise combined with perceived control have been frequently demonstrated, and these investigations have also been extended to many other situations where the presence of control reliably moderates the effects of stress. (18)

In a frequently-cited laboratory study, Matthews and Cannon (1975) found that fewer subjects there willing to help someone who had "accidentally" dropped materials when background noise levels there 85 dB than when they there 65 dB. In a subsequent field study, the same results there demonstrated in a background of lawn mother noise, and this time the addition of a cast on the "victim's" arm enhanced helping behavior under quiet conditions, but failed to do so during the noise episodes (Matthews and Cannon, 1975). In another such experiment, Sauser et al. (1978) found that subjects recommended lower salaries for fictitious employees when exposed to A-weighted levels of office noise at 70 to 80 dB than in quiet. Broadbent (1979 and 1983) cites additional evidence suggesting that subjects will give each other increased amounts of shock and noise when they themselves are exposed to noise, and also cites evidence that noise increases anxiety levels (Broadbent, 1983).

As mentioned above, the presence of control, or even perceived control, is one of the most important predictors of adverse behavioral effects. Subjects who perceive that they have control over the noise show significantly greater tolerance for frustration than subjects without control, even if the control is never exercised (Glass and Singer, 1972). In a recent experiment, Singer and his colleagues found that subjects who there told that they had control of an A-weighted, 103-dB noise stimulus showed significantly greater persistence on a difficult task than subjects who had no control or subjects that had control for only part of the experiment (Singer et al., 1990). This finding occurred despite the fact that the subjects with only partial control reported feelings of control no different from those with full control. To the extent that these findings can be generalized to populations living in noisy areas, this kind of research may have significant sociological implications.

6. Summary: Effects on performance and behavior

Noise can adversely affect task performance in a variety of circumstances. In the past, research in this area has focused mainly on the occupational setting, where noise levels must be sufficiently high and the task sufficiently complex for performance decrements to occur. Recent research implicates more moderate noise levels, especially when speech is the disruptive noise stimulus. Some research indicates that noise can also produce disruptive after-effects, commonly manifested as a reduced tolerance for frustration, and it appears that the presence and timing of control over the noise are critical to the prediction of after-effects. Even moderate noise levels can increase anxiety, decrease the incidence of helping behavior, and increase the risk of hostile behavior in experimental subjects. These effects may, to some extent, help explain the "dehumanization" of today's urban environment.

BACK