Luz, G, Valente, D. Ecological Psychoacoustics, Managing Response to Noise
The potential role of ecological psychoacoustics for managing community response to noise
Luz Social and Environmental Associates, Inc. 4910 Crowson Ave, Baltimore, Maryland 21212, USA
U.S. Army Corps of Engineers, Engineer Research and Development Center/Construction Engineering Research Laboratory (ERDC-CERL), 2902 Newmark Drive, Champaign, Illinois 61822–1076
The term ecological psychoacoustics is a relatively recent addition to interdisciplinary studies of sound perception. This paper begins with a discussion of how ecological psychoacoustics has emerged from a synthesis of soundscape analysis, acoustic ecology and physiological psychology. Building on S. Namba and S. Kuwano’s chapter on Environmental Acoustics: Psychological Assessment of Noise in R. Neuhoff’s landmark book, Ecological Psychoacoustics (2004), the authors discuss the potential of using concepts from ecological psychoacoustics towards understanding individual differences in community noise response. As an example, ecological psychoacoustics will be used to frame the problem of human response to wind farm noise. Two key concepts in this understanding are (1) Salt and Hullar’s (2010) demonstration that the human auditory system can respond to sound which is not consciously audible and (2) the demonstration by a number of researchers of the propensity of humans for auditory entrainment to rhythmic sound.
In an intellectual and academic sense, noise contour lines plotted on a city or county zoning map are a legacy of the father of psychophysics, Gustav Fechner. The winding path from Fechner’s psychophysics to psychoacoustics was documented in E.G. Boring’s (1942) Sensation and Perception in the History of Experimental Psychology.1 Boring’s history ended at a threshold of innovation when the demands of the telephone industry and World War II led to new uses of vacuum tube technology, initiating the golden age of psychoacoustics. As brilliant young students rushed to use this new technology, it is likely that many were unaware of two implicit assumptions inherited from earlier students of psychoacoustics: (1) The auditory system is a bottom-up system in which top-down effects can be ignored, and (2) with the exception of identifiable pathologies, human auditory systems are essentially the same and describable with the normal curve. A manifestation of the second assumption is the fact that research in psychoacoustics is often acceptable with relatively few subjects (e.g. less than 10). However research across numerous fields over the past half-century has uncovered knowledge about auditory neurophysiology, psychology, and behavior which challenge the validity of both of these assumptions, and is only beginning to affect the way that community response to noise is addressed.
2 THE LEGACY OF LOUDNESS
The historical record suggests that the first country to consider a methodology for regulating the noise of motor vehicles was the U.K., where the Ministry of Transport’s “Departmental Committee on Noise in the Operation of Mechanically Propelled Vehicles” commissioned the National Physical Laboratory (NPL) to measure the pass by noise of motor cars in 1934.2 Interrupted by World War II, NPL researchers did not return to this subject until 1959. In the meantime, the growth of the commercial airline business, particularly in the U.S., led to a new regulatory challenge.
The history of regulations for aircraft noise began with a single question asked by the New York Port Authority in 1956, “How loud is the Boeing 707 jet aircraft compared with the propeller-driven airplanes which have been using the New York International Airport at Idlewild, now JFK, Airport?” 3 The Boeing Company, which had compared the noise levels of both types of aircraft using the linear scale of the sound level meter, concluded that there was not a significant difference between aircraft. However, the acoustical engineering firm, Bolt, Beranek and Newman, which employed a psychologist, Karl Kryter, to compare the two types of aircraft, concluded that the jet aircraft would be perceived as significantly louder than the propeller-driven aircraft.
Kryter based his comparison on judgments of the “noisiness” of tape recordings from two aircraft: the Super-Constellation propeller aircraft and the Boeing 707 jet aircraft. To convert the spectrum of each sound into a single number representative of “noisiness,” he modified a model for the calculation of “loudness” developed by Dr. S.S. Stevens of the Harvard Psychoacoustic Laboratory. Whereas Stevens had based the calculation of loudness on the complex addition of “phons,” Kryter based the calculation of Perceived Noise Level (PNL or PNdB) on the complex addition of “noys.” 4
The idea of using the Stevens model for calculating the loudness of aircraft flyovers had broad appeal. However, when Kryter introduced this concept, digital technology was too rudimentary to allow his system to be incorporated into sound monitoring instrumentation. This task fell to Eberhard Zwicker. Zwicker’s digital loudness meters5 were not available in 1974 when the USEPA decided to base annoyance estimates on the simpler estimate of loudness, A-weighted equivalent level. Most recently, with ISO’s decision to replace both Stevens and Zwicker loudness with the more physiologically-based Moore model, these earlier models have become footnotes. Nevertheless, the heritage of loudness continues to permeate the regulation of noise annoyance.
3 EMERGENCE OF ECOLOGICAL PSYCHOACOUSTICS
The term, ecological psychoacoustics, was introduced as the title of a book edited by Neuhoff (2004).6 It appears that the first use of the term was in a 1990 paper7 where it was used as a synonym for the analysis of “soundscape”, a term coined earlier by R. Murray Schafer in connection with the World Soundscape Project.8 Neuhoff’s use of the term is more comparable to the term “ecological acoustics” in Warren and Verbrugge’s (1984) analysis of the human ability to discriminate between the sound of a breaking glass bottle from the sound of a bouncing glass bottle.9 Neuhoff’s definition envisions the subject as examining “divergent areas of hearing science, tying together the occurrence of acoustic events, physiological responses of the auditory system, the perceptual and cognitive experience of the listener, and how this experience influences behavior, action, and subsequent perception.”
In simplest terms, ecological psychoacoustics differs from traditional psychoacoustics in emphasizing (1) the analysis of response to real-life sounds (2) in a more realistic setting than the laboratory and (3) with reference to psychophysiological processes. Prior to the U.S. Park Service’s setting up the Natural Sounds Program at Fort Collins, CO, the U.S Federal agency most concerned with the first two elements was the Department of Defense (DoD). For example, Neuhoff’s work on auditory looming (reviewed in Chapter 4 of Ecological Psychoacoustics) was anticipated in a 1970 study from the Aerospace Medical Research Laboratories at Wright-Patterson Air Force Base, OH.10 Although this early work clearly suggested that an aircraft flying toward an observer would be more annoying than an aircraft of equal Sound Exposure Level flying past an observer, the Air Force could not use the information because (1) the technology to “look inside the brain” used by Neuhoff had not been developed and (2) signal processing technology for identifying an “auditory object” such as an aircraft in a waveform, had not yet been invented.
3.3 Operating at the Limits of the Auditory System
DoD’s interest in research into responses to real life sounds in real life settings derived from the fact that the some of the most annoying sounds propagating from military training areas into civilian residential areas are at the limits of the normal operating conditions for human hearing. Whereas psychologists can achieve correlations in excess of r = 0.95 between normal urban sounds and various subjective parameters when the levels are below 90 dBA11, the sounds of military training can be far more intense. For example, in an Air Force-funded study of the annoyance of low-level jet aircraft flights conducted inside a rented home, the maximum sound level of “realistic” flights ranged between 76.6 and 128 dBA.12
Whereas the sounds of low-level flights fall within the upper limits of intensity, the sounds of explosions often fall at the lower limits of the audiogram. The following example from a letter received from the USEPA in 1977 and forwarded to the U.S. Army Medical Department with names and address deleted is illustrative:
“…We have purchased a home down here and have felt ‘vibrations’ at various times of the day in this house. These ‘sound waves’ or whatever they are, rev up and peak and then diminish. All four of us react to them by nervousness, jitternyness (sic), and hyperactivity in the youngest. We sense the build-up and cessation simultaneously. ….The acoustics of our house are unreal. At times (we measure by a clap of the hands) sounds just reverberate off the walls. When the house is louder and reflects noise even more the sensation of ‘vibrations’ seems to build up and gets everyone uptight. We have contacted the Military about this back in November of ’77, and all they did was send us through a series of physcological (sic) sessions and shrugged their shoulders about any explanation to us.…”13
Thirty five years ago, when this letter was written, Army officials had no option but to “shrug their shoulders.” Today, there are more choices. Building on the pioneering work of Schomer, who studied the annoyance of real explosions experienced in realistic settings,14 the Acoustics Research Team at the U.S. Army Construction Engineering Research Laboratory has embarked on the study of weapons noise annoyance routinely experienced by people living near military ranges in their own homes. The first study in this series looked at sleep disturbance.15 Currently, there is an in situ study in which residents report their daytime annoyance to blasts from homes instrumented to record both sound and vibration. Of particular interest for the analysis of the in situ data will be any differences between subjects who are most perturbed and least perturbed by low frequency blast events.
4 TEST CASE: ECOLOGICAL PSYCHOACOUSTICS AND THE WIND TURBINE SOUND DEBATE
4.1 The Health Effects Controversy
The symptoms reported by the writer of the 1977 letter are rare. Descriptions of fear, psychophysiological effects (e.g. sleep disturbance) or health effects (e.g. nervousness, migraines) from Army heavy weapons noise comprised only 4% in a 1983 analysis16 and 8% in a 2008 analysis.17 In this regard, complaints of health effects from low frequency blast noise are comparable to the findings in the scientific literature on wind turbine sound. As noted in the expert panel review on health effects of wind turbine noise prepared for the American Wind Energy Association and the Canadian Wind Energy Association in 2009, selective screening for wind turbine associated adverse health effects by an M.D., Dr. Nina Pierpont, yielded only10 families with a total of 37 persons suffering from a hypothetical “wind turbine syndrome.”18 The 2009 expert opinion noted that the symptoms reported by this rarified sample are not unique, overlapping with a list that includes: “distraction, dizziness, eye strain, fatigue, feeling vibration, headache, insomnia, muscle spasm, nausea, nose bleeds, palpitations, pressure in the ears or head, skin burns, stress and tension.” Although Pierpont’s recruiting procedure has been criticized, having requested subjects to come forward if they believed their health had been adversely affected by wind turbines, it is consistent with the recommendation made in 1977 by two pioneers in the study of low frequency environmental noise who recommended that
“a detailed survey could be made of those people throughout the country who complain of unidentifiable noise to try to determine the extent of correlation between them and to establish those problems which are imagined or self-generated and those which have their origin in real acoustic phenomena.” 19
Even the most neutral observer would have to acknowledge that the symptoms described by Pierpont are extreme,20 and it is tempting to discount them with other explanations. The 2009 expert opinion proposes two: (1) nocebo effects and (2) somatoform disorder. Joining this list as an alternative explanation for LFN complaints is tinnitus.21 In this case, the panel was signatory to the following statement.
“There is no evidence that the audible or sub-audible sounds emitted by wind turbines have any direct adverse physiological effects.”
However, one year after the report was issued, Salt and Hullar at Washington University in Saint Louis observed “abnormal states in which the ear becomes hypersensitive to infrasound.”22 Included in Salt and Hullar’s theory is the observation that “at very low frequencies the OHC are stimulated by sounds at levels below those that are heard.” This perhaps doesn’t change the panel’s conclusion, but suggests that low frequency noise, while not audible, may still be psychologically perceived by at least a small portion of a population.
4.2 Psychological Effects
The possible perception of inaudible low frequency noise from wind turbines, of course, leads to question of whether any of the reported physiological symptoms are consequences of the psychological distress that this may cause. In fact, many symptoms discussed above (stress, tension, distraction, jitteriness, insomnia, etc) are primarily psychological in nature. Clearly, the higher frequency “whoosh” of wind turbines has been reported to cause a significant amount of noise annoyance.23, 24, 25 There have been several studies reporting associations of psychological effects of wind turbine noise beyond those examining annoyance. For example, Bolin et al. (2011) reported a statistically-significant association between noise levels and self-reported sleep disturbance in two out of three cross-sectional studies of the annoyance from wind turbine noise. 26 Shepherd et al. (2011) found statistically-significant differences in a health-related quality of life (HRQOL) questionnaire between residents of the Makara Valley (New Zealand) living with 2 km vs. farther than 8 km from wind turbines.27
Although psychological effects were not considered in depth by the 2009 panel, a more recent panel assembled by the Massachusetts Department of Public Health did discuss the indirect health effects that sleep disturbance, annoyance, or stress may cause.28 Nevertheless, both panels have reached similar conclusions. To quote the 2009 panel:
“The sounds emitted by wind turbines are not unique. There is no reason to believe, based on the levels and frequencies of the sounds and the panel’s experience with sound exposures in occupational settings, that the sounds from wind turbines could plausibly have direct adverse health consequences.”
4.3 Psychoacoustics of Loudness
There seems to be little evidence that wind turbines have adverse physiological consequences beyond those indirectly caused through noise annoyance and stress, be it to inaudible low frequency noise, home vibration, or the higher frequency “whoosh”. However, this conclusion must be qualified by the fact that nearly all of the studies to date have used only loudness-based metrics to assess any associations. Much of the research in ecological psychoacoustics points beyond loudness as being the primary acoustic quantity relevant for sound perception.
Anecdotally, for wind turbine noise, this is true as well. From a videotaped interview of one of Pierpont’s 37 subjects posted to the Internet, it is clear that at least one of these rarefied subjects discriminates between the annoyance of the audible components of the wind turbine acoustic signature and the symptoms attributed to inaudible sound.29 This interviewee stated that occupants of his home can use masking noise to block out the audible sound but that masking the sound does not stop the distress which he experiences from what has been described as infrasound. For this individual, it appears that the loudness of the acoustic event is not particularly relevant.
In the equal loudness curves of Fletcher and Munson, their practical application to the design of sound level meters (A, B and C weighting) and the Stevens loudness model, 1 kHz serves as the anchor against which the loudness of other pure tones is calibrated. The 1 kHz is ideally suited to be an anchor because it is the audiometric frequency at which (1) the standard deviation of the thresholds of young healthy ears is smallest30 and (2) the approximation of 10 dB increases corresponding to a subjective doubling of loudness extends over the largest dynamic range. The utility of the 1 kHz anchored models for assessing the annoyance of sounds in the mid-frequency range has been demonstrated numerous times. A recent example is Furihata’s demonstration that the A-weighted equivalent sound level can explain 65% of the variance in individual annoyance of 5 sec samples of urban noise among citizens of Japanese cities from which the sounds were recorded. With the inclusion of a single question on noise sensitivity, the explained variance increases to 72%.31
At the same time, it has been known for at least forty years that the loudness of sounds with spectra falling into the range of the spectra of wind turbine sound is underestimated by the Stevens model. In a paper given in 1972, Tempest reviewed a number of case studies from his own acoustical engineering practice and wrote, “It is the author’s final conclusion, at this stage in the work, that the problem is not in the scales, (dB(A) and PLdB) but arises because annoyance at low frequencies is not directly related to loudness, and the use of loudness scales, or scales based on loudness in this range is just not adequate.”32 Tempest also pointed out that “dB(A) was originally designed for use at fairly low intensities and makes no allowance for the rapid growth of loudness with intensity which occurs in this region.”
Rapid growth with intensity also characterizes the annoyance judgments of wind turbine sound. This point is illustrated by a comparison of the cumulative proportions of people experiencing different degrees of annoyance for different traffic noise exposures with the analogous growth for people experiencing different degrees of annoyance for different wind turbine noise exposures. For traffic noise experienced right outside the apartment, the data for 3,957 interviewees can be found in Figure 4 of a study of Norwegian citizens.33 For wind turbine noise experienced outside by Swedish citizens, the data can be found in Figure 1 of Pedersen and Waye (2008).24 Although it is not possible to plot these data on the same scale, the faster growth of annoyance with increases in level of wind turbine noise is apparent from the comparison.
4.4 Ecological Psychoacoustics of Entrainment
In addition to the more rapid growth in annoyance for wind turbine sounds compared with traffic noise, the threshold of noticeability appears to be lower for wind turbines. In Pedersen’s study of Swedish citizens,34 36% of the interviewees living on bottom land and 25% of those living in more hilly terrain reported noticing the source at levels below 32.5 dBA. Conceivably, this unusually low threshold of noticeability could be explained by the summation of critical bands, which are narrowest at the lowest frequencies.35 Another hypothesis derived from ecological psychoacoustics is that the sensitivity of some individuals to wind turbine sounds is enhanced by entrainment.
Entrainment is a psychoacoustic variable which cannot be captured by any loudness meter. An excellent discussion of this phenomenon can be found in M.R. Jones’ chapter in Neuhoff (2004).36 Through a series of experiments, Jones demonstrates that the human brain is not merely a passive timer but a system which actively anticipates the reoccurrence of an auditory event within an interval of roughly 2 Hz.
The intervals explored by Jones overlap with the intervals studied by Guttman and Julesz (1963) in their study of “frozen noise segments” (RFNs).37 They reported that repetitions of the acoustic signature of a segment of noise could be easily detected with a repetition rate of 1 sec. For RFNs from 1 sec to 250 msec, the experience was “whooshing” and for repetition rates of 250 msec to 50 msec (20 Hz), the experience was described as “motorboating.”
In the natural world, the ability of both predator and prey to capture the acoustic signatures of faint sounds and detect their repetition is essential for survival. At the same time, retention of the acoustic memory trace over long periods would be maladaptive. With training, some listeners can detect RFNs as long as 10 sec,38 but 1 sec is more normative. In short, there is reason to believe that the human auditory system is designed to respond to sounds with repetition rates comparable to wind turbine sound. Quite possibly, the important question is not “how can people be annoyed by such faint sounds” but “what characterizes a person who has difficulty in turning off their attention (i.e. habituating) to such faint sounds?”
At this point, the authors of this paper are only aware of anecdotal evidence for individual differences in habituating to the repetitive sounds of machinery. One anecdote concerns the author, Theodor Geisel (Dr. Seuss). As reported on U.S. National Public Radio, Geisel traveled from Europe to the U.S. in 1937.39 For eight days, he listened to the repetitive throb on the ship’s engine. “The sound got stuck in his head, and he started writing to the rhythm. Eventually, those rhythmic lines in his head turned into his first children’s book.” This association between rhythm and language is not coincidental. Entrainment in terms of music and movement appears to be a characteristic of species capable of vocal mimicking.40
If, as hypothesized here, a component of the annoyance of wind turbine noise is a tendency toward entrainment, then the discipline of neuroscience promises to allow objective evaluations of individual differences. As noted in an earlier section of this paper, a major difference between the US Air Force’s investigation into auditory looming in 1970 and Neuhoff’s study of auditory looming in the 21st Century is the opportunity for Neuhoff to look at neurophysiological correlates in the brain. The feasibility of studying the response of “very annoyed” by wind turbine noise and the unperturbed is demonstrated by a study of brain wave synchronization published by Will and Berg.41 As noted by these authors, “An area nearly absent from previous research is the synchronization of brain waves to auditory stimuli with repetition rates below 10 Hz.” In their study, they compared three categories of acoustic stimulation: (1) drum sounds and clicks with repetition rates of 1–8 Hz), (2) silence and (3) pink noise. Their findings suggest that there is an opportunity to measure three distinct processes in response to repetitive stimuli.
The whoosh of a wind turbine is an auditory object which is an uncomfortable distracter to the brain’s ongoing activity. The awareness of this auditory object is enhanced by the visual image as well as by the rhythmic entrainment of attention. To minimize annoyance, individuals need effective neurological mechanism to block this auditory object from consciousness. At the same time, the brain’s tendency toward neural resonance42 becomes an impediment to “non-attending.”
5 SUMMARY AND CONCLUSIONS
For the acoustical engineer engaged in regulating residential land use through the normally powerful tool of noise mapping, wind farm noise presents four unique challenges which are not usually found with noise mapping of traffic noise: (1) Day-to-day variability in sound level is higher than for traffic noise, (2) Ambient background is important, (3) The standard practice of plotting 5 decibel contour lines is less informative because of the accelerated transition between the threshold of noticeability and significant annoyance, (4) Individual differences in annoyance appear to be larger. Given the uncertainty about individual differences, it is difficult to predict whether a particular person will be annoyed by wind farm noise. For this reason, we believe that the practice of paying people in advance to not complain about wind farm runs counter to environmental justice.43 Instead, we recommend two areas of future research:
- Identification of neurological mechanisms to account for differences between people who report unawareness of sound at levels where a small minority report “high annoyance”
- Research into acoustic instrumentation which can go beyond simple loudness calculations to mimic the brain’s ability to form auditory objects.
These suggestions are not merely academic. If individual differences could be predicted in advance, then it might be possible to accommodate the hyper-sensitive in a fair and equitable manner.
The collaboration in this paper was independent of the junior author’s employment by the Department of Defense. All references to research supported by DoD funding are to publications cleared for public distribution, and no new information about DoD funding of research into noise annoyance has been discussed. In no case should the opinions expressed in this paper be construed as representing present or future policy positions by DoD.
Download the Luz — Valente powerpoint presentation entitled: The Potential Role of Ecological Psychoacoustics for Managing Community Response to Noise