Assessing Aircraft Noise Conditions Affecting Student Learning, Volume 1: Final Report (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER 2. LITERATURE REVIEW This literature review critically evaluates the field of noise effects on children’s cognitive performance and learning outcomes describing the evidence for effects from research conducted over the past few decades. In addition, the review evaluates the characterization of noise exposure in these studies; considers studies of classroom acoustics; describes current guidelines for the acoustical design of children’s classrooms; and identifies current gaps in knowledge and future research priorities. This review examines non-auditory effects of noise exposure: that is effects on human health which are not the direct result of sound energy. Non-auditory effects are less well established and accepted than the auditory effects of noise which are the direct result of sound energy, such as hearing loss. The review focuses on the non-auditory effect of noise on children’s cognition and learning, as well as briefly summarizing other non-auditory effects on children’s health. As part of the literature review a total of 106 key documents were identified. These have been reviewed and an annotated bibliography prepared with abstracts. The annotated bibliography is presented in Appendix A to this report. In addition, a document catalog has been developed in tabular form identifying key elements of each document, namely: Title, Date, Country, Location, Noise source, #Schools, #Classrooms, #Classes, #Students, School grades, Student performance measure, Noise measure, Research method, Data collection method, Analytical method, Citation, Finding, Suggested criteria, Notes. This document is presented in Appendix B. 2.1. Scope of the Review This literature review summarizes the field of noise effects on children’s cognitive performance and learning, describing the evidence for effects of noise on children’s learning from research conducted over the past few decades, as well as recent developments in the field. This is a narrative review, focusing on key studies in the field. The literature has been identified from searches of electronic databases including PubMed, IngentaConnect, Science Direct, Educational Resource Information Centre, Google Scholar, and the Acoustical Society of America Digital Library, as well as through searches of reference lists of papers, and searches of specific journals including ‘Noise and Health’ and the ‘Journal of the Acoustical Society of America’. This strategy has been supplemented by the research teams’ knowledge of existing reports and publications. The literature is predominantly drawn from Europe and the USA, with a focus on USA-relevant publications, where possible. The review also focuses on studies of aircraft noise exposure, where possible, but does draw on findings of other noise sources such as road traffic noise, where the evidence may be relevant for this project. For convenience, documents referred to in the review are listed at the end of this section. This review considers the characterization of noise exposure in these studies; examines the findings of epidemiological studies which focus on chronic rather than acute noise exposure; considers studies of classroom acoustics; reviews current guidelines for the acoustical design of children’s classrooms; and concludes by identifying current gaps in knowledge and future research priorities. 2-1

Article I. 2.2. Background to the Research Field The direct effect of sound energy on human hearing is well established and accepted (Babisch, 2005; Kryter, 1985). Auditory impairments are typically seen in certain industrial occupations, hence protective legislation requiring hearing protectors to be worn. In contrast, non-auditory effects of noise on human health are not the direct result of sound energy. Instead these effects are the result of noise as a general stressor: thus the use of the term noise not sound: noise is unwanted sound. The non-auditory effects of noise are less well established and accepted than auditory effects. This review focuses on the non-auditory effect of noise on children’s cognition and learning. The effect of environmental noise exposure on children’s cognitive performance and learning has been researched since the early 1970s. Typically, in more recent years, the effect of chronic noise exposure on children’s cognition has been examined by methodologically robust epidemiological studies, while the effect of acute noise exposure has been examined in experimental, laboratory studies. It has been suggested that children may be especially vulnerable to effects of environmental noise as they may have less cognitive capacity to understand and anticipate environmental stressors, as well as a lack of developed coping repertoires (Stansfeld, Haines and Brown, 2000). Exposure during critical periods of learning at school could potentially impair development and have a lifelong effect on educational attainment. Overall, evidence for the effects of noise on children’s cognition has strengthened in recent years. There is increasing synthesis between epidemiological studies, with over twenty studies having shown detrimental effects of noise on children’s memory and reading (Evans and Hygge, 2007). Most research has been carried out in primary school children aged 5 to 12 years. This is seen as a critical learning acquisition period for children in which future learning patterns are established. Studies have established that children exposed to noise at school experience some cognitive impairments, compared with children not exposed to noise: however, these effects are not uniform across all cognitive tasks (Cohen, Evans, Stokols, et al, 1986; Evans and Lepore, 1993; Evans, Kielwer and Martin, 1991). Tasks which involve central processing and language comprehension, such as reading, problem solving and memory appear to be most affected by exposure to noise (Cohen, Evans, Stokols, et al, 1986; Evans and Lepore, 1993; Evans, Hygge and Bullinger, 1995; Hygge, 1994). These are all tasks with high processing demands and the effect of environmental stressors on cognitive tasks with high processing demands is widely accepted in the environmental stress literature (Cohen, Evans, Stokols, et al, 1986; Smith, 1989). Recent years have seen several methodological advancements in the field including the use of larger epidemiological community samples and better characterization of noise measurement. Evidence from longitudinal studies is beginning to emerge, and studies have started to examine exposure-effect relationships, to identify thresholds for noise effects on health and cognition which can be used to inform guidelines for noise exposure. There has also been a better assessment of confounding factors: noise exposure and cognition are often confounded by socioeconomic position; individuals living in poorer social circumstances are more likely to have poorer cognitive abilities, as well as be exposed to noise. Therefore, measures of socioeconomic position need to be taken into account when examining associations between noise exposure and health. 2-2

Article II. 2.3. The Assessment of Noise Exposure Studies of noise effects on children’s cognition typically use established metrics of external noise exposure, which indicate the average sound pressure for a specified period using dBA as the measurement unit (dBA is the unit of A-weighted sound pressure level where A- weighted means that the sound pressure levels in various frequency bands across the audible range have been weighted in accordance with differences in hearing sensitivity at different frequencies). Metrics typically employed are Leq16 and Lday which indicate noise exposure over a 16 hour daytime period usually 7am-11pm; Lnight which indicates noise exposure at night (11pm- 7am); and Ldn which combines the day and night measures to indicate average noise exposure over the 24 hour period, with a 10dB penalty added to the nighttime noise measure. In contemporary studies these metrics are usually modeled using standard airport noise modeling systems, using Geographical Information Systems to present the data, while older studies as well as some contemporary studies measure noise exposure in the community, which can be less reliable if measurements cover short time-periods. A few recent studies have also examined exposure to maximum noise levels (e.g. Lmax), as in pathophysiological terms it is not known whether the overall ‘dose’ of noise exposure is important in determining effects on children’s cognition or whether peak sound pressure events or the number of noise events might be important. This issue is of increasing importance given that the number of noise events for aircraft and road traffic noise are increasing, while noise emission levels per event are falling. In the community, people are often exposed to sounds from more than one source. However, to date, studies tend to focus upon only one type of exposure such as road traffic or aircraft noise exposure. Studies that have examined exposure to more than one source (Lercher, Evans and Meis, 2003) have not been able to attribute cognitive effects to specific noise sources within the environment, limiting their relevance for policy formation or noise abatement. Studies of the non-auditory effects of noise exposure typically use the term ‘noise’ to refer to the individual’s exposure to sound. The term noise is used, regardless of whether the exposure is high or low: lower levels in particular may strictly be better described using the term sound, and the term noise implies that the sound exposure is unwanted and that it is an environmental stressor. This tradition is maintained throughout this literature review. Article III. 2.4. The Effects of Noise Exposure on Children’s Learning 2.4.1. Background There is increasing synthesis between epidemiological studies, with many studies having shown detrimental effects of noise on children’s memory and reading (Evans and Hygge, 2007). While a recent study suggests that children may not be more susceptible to environmental noise effects on cognitive performance than adults (Boman, Enmarker and Hygge, 2005), studies have established that children exposed to noise at school experience some cognitive impairments, compared with children not exposed to noise. Overall, children exposed to chronic environmental noise have been found to have poorer auditory discrimination and speech perception (Cohen, Evans, Krantz, et al, 1980; Cohen, Evans, Stokols, et al, 1986; Cohen, Glass and Singer, 1973; Evans and Maxwell, 1997; Moch-Sibony, 1984): as well as poorer memory (Evans and Lepore, 1993; Hygge, 1994; Hygge, Evans and Bullinger, 2002; Stansfeld, Berglund, Clark, et al, 2005): deficits in sustained attention and visual attention (Hambrick-Dixon, 1986; Hambrick-Dixon, 1988; Moch-Sibony, 1984; Muller, 2-3

Pfeiffer, Jilg, et al, 1998; Sanz, Garcia and Garcia, 1993): and poorer reading ability and school performance on national standardized tests (Bronzaft, 1981; Bronzaft and McCarthy, 1975; Clark, Martin, van Kempen, et al, 2006; Cohen, Glass and Singer, 1973; Eagan, Anderson, Nicholas, et al, 2004; Evans, Hygge and Bullinger, 1995; Evans and Maxwell, 1997; Green, Pasternack and Shore, 1982; Haines, Stansfeld, Brentnall, et al, 2001; Haines, Stansfeld, Head, et al, 2002; Haines, Stansfeld, Job, et al, 2001a; Lukas, DuPree and Swing, 1981). The following section discusses the historical development of the research field, highlighting seminal studies in the field and synthesizing the knowledge base. 2.4.2. Epidemiological Studies (a) 2.4.2.1. The Early Studies. This field of research in children was first investigated by Cohen and colleagues who carried out a naturalistic field study of elementary school children living in four 32-floor apartment buildings located near an expressway (Cohen, Glass and Singer, 1973). The sample of 73 children were tested for auditory discrimination and reading achievement. Children living on lower floors of the 32-storey buildings (i.e. exposed to higher road traffic noise levels) showed greater impairment of auditory discrimination and reading achievement than children living in higher-floor apartments. Using a similar naturalistic paradigm Bronzaft and McCarthy (Bronzaft and McCarthy, 1975) compared reading scores of elementary school children who were taught in classrooms on the noisy side of a school near a railway line with the scores of the school children in classrooms on the quiet side of the same school. Children on the noisy side of the school building had poorer performance on the school achievement tests than those taught in classrooms on the quiet side of the school. The mean reading age of children in the classrooms on the noisy side of the school was 3 to 4 months behind the children in the low noise-exposed classrooms. The strength of this study is that the results cannot be attributed to self-selection of children into one school rather than another, a methodological problem found in many field studies. Neither could the results be explained by the selection of more able children into quieter classrooms, as children were not assigned in any systematic manner to classrooms on the noisy or quiet side of the school. The first study to combine both cross-sectional and longitudinal analysis was a naturalistic study around Los Angeles Airport (Cohen, Evans, Krantz, et al, 1980; Cohen, Evans, Krantz, et al, 1981). This study found impaired performance on difficult cognitive tasks in primary school children aged 8-9 years. There was also an effect on motivation: after experiencing success or failure on a task, children exposed to chronic aircraft noise were less likely to solve a difficult puzzle and were more likely to give up. At follow-up one year later (Cohen, Evans, Krantz, et al, 1981) the finding that noise-exposed children were less likely to solve a difficult puzzle was replicated but the finding that the same children were more likely to give up on a difficult puzzle was not. The first exposure-effect study was carried out by Green, et al. (Green, Pasternack and Shore, 1982) in New York, relating noise exposure scores based on noise exposure forecast contours for New York City Airports and relating that to the percentage of students reading below grade level between 1972 and 1976. The demonstration of exposure-effect relationships is an important element in confirming causal associations between noise and cognitive and health outcomes. The percent of students reading below grade level in each grade was regressed on noise exposure for each of the years 1972-76 and for all 5 years combined. Social disadvantage 2-4

was adjusted for in terms of the percent eligible for free lunch programmes, along with adjustment for ethnic group, absentee admissions and departure rates, pupil-teacher ratio and teacher experience. The partial regression coefficients for the noise scale variables were all positive and were statistically significant at the 0.05 probability level in 15 of 18 regressions. A summary coefficient, with appropriate weighting, of 0.62 (95% Confidence Interval, CI, in 0.51-0.74) was estimated, suggesting that a one unit increase in noise score would be accompanied by an increase of 0.62% in the number of students reading one or more years below grade level in an average school. The authors described the data as largely compatible with a linear dose-response relationship between noise exposure and percent reading below grade level. The mean difference in the percent reading one or more years below grade level in the noisy schools, compared to the quietest schools, was 3.6% (95% CI 1.5-5.8). There are several limitations to this study, the crudeness of the noise exposure scale, the possibility that pupils transfer from schools across noise zones and therefore, have a varied history of noise exposure, the crudeness of the variables used to assess confounding and the aggregate nature of the percentage reading below grade level outcome which fails to take into account individual differences in reading ability. Nevertheless, these results are striking and it could be argued that were the methodological errors soluble, the size of the effect would be likely to be larger, rather than smaller. This study was very much ahead of its time and another exposure-effect study was not conducted until the RANCH study in the early 2000’s (Stansfeld, Berglund, Clark, et al, 2005). (b) 2.4.2.2. The Heathrow Studies. A repeated measures field study carried out around Heathrow airport in London, partially confirmed the findings of Cohen et al’s (1980; 1981) study of Los Angeles airport. This study compared the cognitive performance of children aged 9-10 years, attending four schools exposed to high levels of aircraft noise (exterior levels >66 dB LAeq 16hr) with children attending four matched control schools exposed to lower levels of aircraft noise (<57 dB LAeq16) (Haines, Stansfeld, Job, et al, 2001a). Children tested at baseline were re-tested a year later at follow-up (Haines, Stansfeld, Job, et al, 2001b). The results indicated that chronic exposure to aircraft noise was associated with impaired reading comprehension and sustained attention after adjustment for age, main language spoken at home and household deprivation (Haines, Stansfeld, Job, et al, 2001a). The within subjects analysis at follow-up indicated that children’s development in reading comprehension may be adversely affected by chronic aircraft noise exposure (Haines, Stansfeld, Job, et al, 2001b). This study was followed by a larger study - The West London Schools Study (Haines, Stansfeld, Brentnall, et al, 2001) which compared the cognitive performance and stress responses of children in ten high-noise schools with that of children in ten matched control schools. The results indicated that children in the noise-exposed schools experienced greater annoyance and had poorer reading performance on the difficult items of a national standardized reading test. A further multi-level modeling study of national standardized test scores (SATs) carried out around Heathrow airport (Haines, Stansfeld, Head, et al, 2002), examined test scores for 11,000 11-year-old children in relation to aircraft noise exposure contours for their school. The results showed that there was an exposure-effect relationship between noise exposure and performance on reading and math tests, but that this was influenced by socioeconomic factors. Overall, the evidence from the Heathrow studies is supportive of an effect of aircraft noise 2-5

exposure on children’s reading comprehension, suggesting that effects may not habituate over time and may be influenced by socioeconomic factors. While some previous studies have examined the role of socioeconomic factors and demonstrated an effect of noise on cognition over and above the influence of socioeconomic factors (Haines, Stansfeld, Job, et al, 2001a), other studies have not (Haines, Stansfeld, Brentnall, et al, 2001; Haines, Stansfeld, Job, et al, 2001b). Many studies have either failed to measure socioeconomic status adequately or have neglected to measure it at all, despite its potentially important role in the relationship between noise exposure and cognitive performance. (c) 2.4.2.3. The Tyrol Mountain Study. A recent cross-sectional study also found an effect of noise exposure on cognitive performance (Lercher, Evans and Meis, 2003). This study examined the effect of modest levels of ambient community noise (mainly train and road traffic noise) on the cognitive performance of 9-10-year-old children from rural Alpine areas. Half of the sample were from areas where ambient noise levels were below 50 dBA and half from areas where ambient noise levels were above 60 dBA; all children were tested in a sound-attenuated laboratory. Lercher and his colleagues found a significant effect of ambient noise exposure on memory, but no effect for attention; it is possible that the attention test was affected by a ceiling effect. The authors concluded that even relatively modest levels of exposure to community noise may be sufficient to have a detrimental effect on children’s cognitive functioning. (d) 2.4.2.4. The Munich Study. Stronger evidence to suggest the existence of noise effects on cognitive performance comes from intervention studies and natural experiments where changes in noise exposure have been accompanied by changes in cognition. One of the most interesting and compelling studies in this field is the naturally occurring longitudinal quasi-experiment reported by Evans and colleagues, examining the effect of the relocation of Munich airport on children’s health and cognition (Evans, Bullinger and Hygge, 1998; Evans, Hygge and Bullinger, 1995; Hygge, Evans and Bullinger, 2002). In 1992 the old Munich airport closed and was relocated. Prior to relocation, high noise exposure was associated with deficits in long-term memory and reading comprehension in children with a mean age of 10.8 years. Two years after the closure of the airport, these deficits disappeared, indicating that noise effects on cognition may be reversible if exposure to the noise ceases. Most convincing, was the finding that deficits in memory and reading comprehension developed over the 2 year follow-up for children who became newly noise exposed near the new airport: deficits were also observed in speech perception for the newly noise-exposed children. The Munich study is one of the few longitudinal studies in the field, providing important evidence for a cause- effect relationship between noise exposure and cognitive deficits. (e) 2.4.2.5. The RANCH Study. While by the end of the 1990s, studies had demonstrated effects of noise exposure on children’s cognition and learning, there were several significant limitations to the knowledge base. Firstly, little knowledge about exposure-effect relationships between noise and children’s cognition was available, as studies had tended to compare the performance of children in high noise exposure with children in low noise exposure: thus limiting the range of noise exposures examined. Secondly, it had not been possible to compare the effect size across countries, due to the differing methodologies employed in each country and also the different noise metrics examined. 2-6

The RANCH study (Road traffic and Aircraft Noise exposure and children’s Cognition and Health) (Clark, Martin, van Kempen, et al, 2006; Stansfeld, Berglund, Clark, et al, 2005), compared the effect of aircraft noise and road traffic noise on the cognition over 2000 9-10-year- old children attending 89 schools around three major airports in the Netherlands, Spain and the UK. This was the largest cross-sectional study of its type to date; was the first study to derive exposure-effect associations for a range of cognitive and health outcomes; and was the first to compare effect sizes across countries. Cognition (reading comprehension, recognition memory, conceptual recall, information recall, working memory and sustained attention) was measured using the same paper and pencil tests of cognition across the countries, administered in the classroom. Parents and children also completed questionnaires to obtain information about socioeconomic and demographic factors, and noise annoyance. The data were pooled and analyzed using multi-level modeling to take school-level variance into account. The study found a linear exposure-effect relationship between chronic aircraft noise exposure and impaired reading comprehension and recognition memory, after taking a range of socioeconomic and confounding factors into account including mother’s education, long- standing illness, the extent of classroom insulation against noise, and acute noise during testing (Stansfeld, Berglund, Clark, et al, 2005). No associations were observed between chronic road traffic noise exposure and cognition, with the exception of conceptual recall and information recall, which surprisingly showed better performance in high road traffic noise areas. Neither aircraft noise nor road traffic noise affected attention or working memory: this is the largest study to date to examine attention and this finding contrasts with older studies which had shown effects (Cohen, Glass and Singer, 1973; Moch-Sibony, 1984). In terms of the magnitude of the effect of aircraft noise on reading comprehension, a 5 dB Leq16 increase in aircraft noise exposure was associated with a 2 month delay in reading age in the UK and a 1 month delay in the Netherlands (Clark, Martin, van Kempen, et al, 2006): this association remained after adjustment for aircraft noise annoyance and cognitive abilities including episodic memory, working memory and attention. One further contribution of the RANCH study was that the exposure-effect associations identified between aircraft noise and reading comprehension and recognition memory, made it possible to start to quantify the magnitudes of noise induced impairments on children’s cognition. Figure 2-1 shows the exposure-effect association between aircraft noise exposure and reading comprehension in the RANCH study (Stansfeld, Berglund, Clark, et al, 2005), which can be used to guide decision making by stakeholders and policy makers, as well as to estimate the benefits of noise reduction. This figure indicates that reading falls below average (a z score of 0) at exposures greater than 55 dBA: however, as the relationship between aircraft noise and reading comprehension was linear, reducing exposure at any level should lead to improvements in reading comprehension. Another conclusion of the RANCH study was that while aircraft noise has only a small effect on reading comprehension, it was possible that children may be exposed to aircraft noise for many of their childhood years and the consequences of long-term noise exposure on reading comprehension and further cognitive development remain unknown. A follow-up of the UK RANCH sample is currently being analyzed, to examine the long-term effects of aircraft noise exposure at primary school on children’s reading comprehension (Clark, Stansfeld and Head, 2009). Preliminary analysis indicates a trend for reading comprehension to be poorer at 15-16 years of age for children who attended noise-exposed primary schools. There was also a trend for 2-7

reading comprehension to be poorer in aircraft noise-exposed secondary schools. However, further analyses adjusting for confounding factors are ongoing and are required to confirm these initial conclusions. Reprinted from The Lancet, Vol. 365, Stansfeld, S. A., Berglund, B., Clark, C., pp. 1942-1949, 2005 with permission from Elsevier. Figure 2-1. RANCH study adjusted mean reading Z score (95% Confidence Intervals) for 5 dBA bands of annual aircraft noise exposure at school (adjusted for age, sex and country) (Clark, Martin, van Kempen, et al, 2006; Stansfeld, Berglund, Clark, et al, 2005). [NB: reading comprehension was measured using a z score which has a mean score of 0 and a standard deviation of 1]. (c) 2.4.2.6. The FICAN Pilot Study. The Federal Interagency Committee on Aviation Noise (FICAN) recently funded a novel pilot study which assessed the relationship between aircraft noise reduction and standardized test scores (Eagan, Anderson, Nicholas, et al, 2004; FICAN, 2007). The study evaluated whether abrupt aircraft noise reduction within classrooms, caused either by airport closure or newly implemented sound insulation, was associated with improvements in test scores, in 35 public schools near three US airports in Illinois and Texas. This study is one of the only recent studies to examine the effectiveness of school sound insulation programmes as few intervention studies have been carried out (Bronzaft, 1981; Cohen, Evans, Krantz, et al, 1981). This study is novel as it assesses noise exposure using a number of different metrics to compare the associations between different metrics and learning outcomes. This study examined Leq [the indoor equivalent sound level averaged over the 9-hour school day]; a Speech Intelligibility Index (SII) [the number of events disrupting indoor speech ANEv<0.98SII – disruption of 1% of words]; and Speech Interference Level (SIL), which was assessed by the number of events above 40dBA disrupting indoor speech – ANEv>40SIL, and by the fraction of time speech is disrupted above 40dBA. However, it is important to note for comparison purposes that this study relies on computed noise exposure metrics, which were converted to indoor values; this makes comparison with other studies difficult, as most studies in this field use outdoor exposure values or computed metrics to assess effects on children’s learning. Children’s -.2 0 .2 .4 R ea di ng Z -s co re 30 35 40 45 50 55 60 65 70 aircraft noise dB(A) 2-8

learning outcomes assessed from standardized tests scores included the percentage of students with the worst test grade; the average numerical score; and the percentage of students with the best test grade. Class averaged scores on these tests after one year of schooling following noise reduction were compared with scores in the years prior to noise reduction. After adjusting for demographic and school construction factors, the study found a significant association between noise reduction and a decrease in failure rates but only for high school students and not middle or elementary school students: conversely, there were some weaker associations between noise reduction and an increase in failure rates for middle and elementary schools. For high school students, when the time exposed to over 40 dBA was reduced by 5%, the failure rate decreased by 20%. The study also found a significant association between noise reduction and average test scores for high school, middle, and elementary school students: average test scores increased by 7-9% when the average time exposed over 40dBA was decreased by 5%. When the number of events with LAmax greater than 40dB was examined, middle and elementary school students showed modest improvements in average score, between 4-5% when the number of events decreased by 20, while for high school students a reduction in the number of events was associated with poorer average scores, between 17-19%. Overall the study found that the associations observed were similar for children with or without learning difficulties, and between verbal and math/science tests. Overall, this small scale study does find some evidence for effects of aircraft noise reduction and improved test results, although it must be acknowledged that some associations were null and some associations were not in the direction hypothesized. This was a pilot study and the authors stress that the airports and schools selected for the study may not be representative; that further, larger studies are required; that future studies should utilise airport data for noise exposure assessment; and that outdoor-to-indoor noise measurements at each school should be considered. 2.4.3. Mechanisms The findings of studies of noise effects on children’s cognitive performance suggest that noise may directly affect reading comprehension or that noise effects could be accounted for by other mechanisms including teacher and pupil frustration (Evans and Lepore, 1993), learned helplessness (Evans and Stecker, 2004) and impaired attention (Cohen, Glass and Singer, 1973; Evans and Lepore, 1993). Noise exposure can cause arousal, which improves performance on simple tasks but impairs performance on complex tasks (Yerkes and Dodson, 1908). Further, noise could restrict attention during complex tasks. Typically, it has been assumed that children may adapt to noise interference during activities by filtering out unwanted noise stimuli (Cohen, Evans, Stokols, et al, 1986). This ‘tuning out’ strategy may over generalize to situations where noise is not present such that children tune out indiscriminately. Thus, they may tune out when they should be listening to important information. The presence of this tuning out response is supported by the finding of some older studies that children exposed to noise have deficits in attention, auditory discrimination and speech perception (Cohen, Glass and Singer, 1973; Evans, Hygge and Bullinger, 1995; Moch-Sibony, 1984). However, there is evidence from more recent studies that sustained attention is not impaired by aircraft noise (Haines, Stansfeld, Job, et al, 2001b; Stansfeld, Berglund, Clark, et al, 2005) and that noise effects on cognition are not mediated by impairment of attention (Hygge, Boman and Enmarker, 2003). 2-9

Another possibility is that noise interferes in the interactions between teachers and pupils. Teacher frustration and interruptions in communication between teachers and children could also be a mechanism for cognitive effects (Evans, Kielwer and Martin, 1991). In the noisiest schools teachers may have to stop teaching while aircraft fly over and if this is frequent it may contribute to interruptions in communication and fatigue in teachers and children and to a reduction of morale and motivation in teachers. A further mechanism which has been proposed is that of learned helplessness, which is when children do not perceive themselves to be in control of their environment. This could be a mechanism that accounts for deficits in motivation in children exposed to noise (Cohen, Evans, Krantz, et al, 1980; Evans, Hygge and Bullinger, 1995). Certainly in both the earlier Los Angeles studies and in the Munich study children exposed to high levels of aircraft noise, did not persist as long as children not exposed to noise, at difficult standard puzzles (Cohen, Evans, Krantz, et al, 1980; Evans, Hygge and Bullinger, 1995), which is suggestive of learned helplessness. Noise also causes annoyance, especially if an individual feels their activities are being disturbed or if it causes difficulties with communication. In some individuals, this annoyance may lead to stress responses. Although the recent FICAN pilot study speculated that impaired learning may result from noise stress (Eagan, Anderson, Nicholas, et al, 2004) at present there is little evidence to directly support the annoyance pathway as a mechanism for effects on cognition. Another pathway for the effect of noise on cognition which has been considered is that of sleep disturbance caused by noise exposure at home. Where catchment areas for schools are fairly small, there is a strong correlation between home and school aircraft noise exposure (Clark, Martin, van Kempen, et al, 2006). Sleep disturbance can impact on well-being causing annoyance, irritation, low mood, fatigue, and impaired task performance (HCN, 2004). Overall few studies have examined sleep disturbance as a mediator of noise effects on cognitive performance. A recent analysis of the cross-sectional Munich and RANCH datasets found that self-reported sleep disturbance did not mediate the association of aircraft noise exposure and cognitive impairment in children (Stansfeld, Hygge, Clark, et al, In press). Future studies should examine this pathway further and include both subjective and objective assessments of sleep disturbance. Experimental studies have tried to develop a greater understanding of how cognitive processes are affected by noise. A series of recent papers (Boman, Enmarker and Hygge, 2005; Enmarker, Boman and Hygge, 2006; Hygge, Boman and Enmarker, 2003) all exploit the same experimental data where participants from four age groups (13-14y, 18-20y, 35-45y, and 55-65y) completed 18 memory tests, covering episodic and semantic systems in declarative memory, while exposed to one of three noise conditions: meaningful but irrelevant speech (speech that has meaning but is not relevant to the subject under review), road traffic noise or quiet. In terms of noise effects on performance, both road traffic noise and meaningful irrelevant speech had a similar effect on task performance and noise effects were strongest for memory of texts, followed by episodic and semantic memory tasks (Boman, Enmarker and Hygge, 2005; Hygge, Boman and Enmarker, 2003). These findings suggested that noise may affect memory by impairing the quality with which information is rehearsed or stored in memory (Hygge, Boman and Enmarker, 2003). However, a later paper, (Enmarker, Boman and Hygge, 2006) found that noise exposure did not alter the structure of different aspects of declarative memory, suggesting that noise was not influencing performance via a change in resource allocation or strategy. 2-10

Overall, several plausible pathways and mechanisms for the effects of noise on children’s cognition have been put forward, but in general evidence for these mechanisms is fairly sparse. 2.4.4. Classroom Acoustics and Children’s Performance (g) 2.4.4.1. Background. Only 6 years ago, Lubman and Sutherland (Lubman and Sutherland, 2004) suggested that there was an unawareness of the educational impact of poor school acoustics, and the joint assessment of classroom acoustics and children’s learning outcomes is a fairly recent development in the field. The past few years have seen the publication of several papers examining the role of classroom acoustics in noise effects on cognition (Astolfi and Pellerey, 2008; Bradley and Sato, 2008; de Oliveira Nunes and Sattler, 2006; Dockrell and Shield, 2004; Dockrell and Shield, 2006; Sato and Bradley, 2008; Shield and Dockrell, 2004; Shield and Dockrell, 2008). These studies typically focus upon noise interference with verbal communication as the mechanism for the effect: some studies simply describe the acoustic characteristics of classrooms, some specifically assess speech intelligibility, and a few relate acoustic conditions to performance outcomes. 2.4.4.2. Acoustic Characteristics of Classrooms. Several studies have attempted to identify which acoustical factors may contribute to poor room acoustics in educational facilities (Bradley, 1986; Bradley and Sato, 2008; Houtgast, 1981; Picard and Bradley, 2001; Sato and Bradley, 2008; Yang and Bradley, 2009). Research has typically focused on the effects of noise levels, reverberation times, and the speech-to-noise ratio on speech intelligibility within classrooms. A review of the field, suggests that younger children require quieter conditions for optimum speech intelligibility, leading the authors to suggest that students aged 6-7 years would require maximum ambient sound levels in occupied classrooms of 28.5 dBA, rising to 34.5 dBA for 8-9 year olds, to 39 dBA for 10-11 year olds, and to 40 dBA for students aged 12 years or older (Picard and Bradley, 2001). Picard and Bradley also concluded that reverberation time was less important than sound levels, with RT 1 kHz around 0.5 seconds being optimum in occupied classrooms: along with speech-to-noise ratios of 15dBA. Another important issue to take into account when considering noise effects on children’s learning and classroom acoustics is that speech intelligibility may vary with age. A recent study, proposes that the suggested 15 dBA speech-to-noise ratio may not be adequate for younger children (Bradley and Sato, 2008) as children’s ability to recognize speech in noise improves with age, leading to the suggestion that 6-year-old children require 7 dBA higher speech-to-noise values to achieve the same speech intelligibility scores as 11 year olds. Younger children are also less likely to have the benefit of increased intelligibility in situations where reverberation times are decreased (Picard and Bradley, 2001; Yang and Bradley, 2009). However, as argued by Picard and Bradley, noise levels in classrooms are often in excess of these optimum conditions leading to problems with speech perception. A recent US study found that ambient noise levels in unoccupied classrooms ranged between 38 and 55 dBA and that occupied reverberation times ranged between 0.3 to 1.1 seconds (Bowden, Wang and Bradley, 2004). In terms of occupied classrooms, a study of Canadian elementary school classrooms found that on average students experienced sound levels of 49.1 dBA, teacher speech levels of 60.4 dBA, and a mean speech-to-noise ratio of 11 dBA during teaching activities (Sato and Bradley, 2008). 2-11

(h) 2.4.4.3. Associations between Classroom Acoustics and Performance. Following on from their studies of classroom acoustics in the UK (Dockrell and Shield, 2004; Shield and Dockrell, 2004), examined the associations between classroom acoustics and the performance of primary school children on a series of verbal literacy and non-verbal speed tasks (Dockrell and Shield, 2006). Children completed the tasks under one of three experimental conditions; quiet, babble (noise of children at 65dB Leq), or babble and environmental noise (65dB Leq). Noise affected verbal and non-verbal performance in different ways: non-verbal processing tasks were performed significantly more poorly by those exposed to babble and environmental noise, while the verbal literacy tasks were performed most poorly by those exposed to the babble noise. Children with special education needs performed differently in noise compared with the rest of the sample: they had poorer performance on the verbal tasks in the babble condition, but better performance on the non-verbal tasks in babble. A recent further study has confirmed associations between external and internal noise exposure at school and the results of national tests for children aged 7-11 years attending London primary schools (Shield and Dockrell, 2008). External noise showed a larger effect on the performance of older children and Lmax showed the strongest association with test scores, suggesting that individual noise events may play an important role on performance effects. The latter finding is also supported by another study which found that pupil’s subjective assessments of noise disturbance and noise intensity showed a stronger relationship with Lmax than with Leq or L90 noise measurements (Astolfi and Pellerey, 2008). Astolfi and Pellerey concluded that pupils seem to be disturbed more by intermittent loud noises than by constant noise. Article IV. 2.5. Current Guidelines for Acoustics in Children’s Classrooms Several guidelines for both internal and external noise exposure at children’s schools have been proposed in recent years in the US and across Europe. The majority of these are guidelines which are not statutory and most specify noise limits for ambient environmental noise, rather than specifying limits for specific noise sources such as aircraft or road traffic noise. This section summarizes relevant international, American, and European guidelines. One of the most influential and foremost guidelines tackling the effects of noise exposure on human health are the World Health Organization (WHO) Community Noise Guidelines (WHO, 2000). These international guidelines specify that for pre-school and school classrooms, internal sound levels should not exceed 35 dB LAeq during class time and that outdoor levels in school playgrounds should not exceed 55 dB LAeq during play. These guidelines are often cited, but are felt by many acousticians to be unachievable, given the extremely low level of sound specified for occupied classrooms. The American National Standards Institute (ANSI) specified a standard for school acoustics in 2002 (ANSI, 2002), which is voluntary not mandatory. The standard specifies a 35 dBA internal background noise limit for unoccupied classrooms: this level was chosen to achieve a minimum 15dB speech-to-noise ratio at the back of the classroom. The standard also specifies a 0.6 s reverberation time for classrooms < 283m3, rising to 0.7 s for classrooms > 283m3 to 566m3 and that intermittent noise should not exceed 40 dBA. The ANSI standard is supported by the Acoustical Society of America and INCE-USA (Lubman and Sutherland, 2004) and it is estimated that two-thirds of American classrooms fail to meet the specified 35 dBA background noise limit (Lubman and Sutherland, 2004). While the WHO and the ANSI guidelines both 2-12

specify a maximum sound level of 35 dBA, it should be noted that for ANSI guidelines this is for unoccupied classrooms, while for the WHO guidelines this is for occupied classrooms. This difference reflects a more pragmatic approach by the ANSI standard. The 35 dBA ANSI standard is also in agreement with the work of Bradley and colleagues on optimum classroom acoustics (Bradley and Sato, 2008). The ANSI guideline was reviewed in 2010 and the requirement that intermittent noise should not exceed 40 dBA was removed from the revised standard (ANSI 2010). There are no Europe-wide guidelines for children’s noise exposure. However, under the EU Noise Directive (2002/49/EC), European countries can prepare noise action plans or include noise reduction in national environment and health plans (Bistrup, Haines, Hygge, et al, 2002). Many European countries, such as Sweden, the UK, Germany, and the Netherlands, have issued guidance on safe levels for noise exposure outside and inside schools but all recommended levels for external noise exposure are guidelines and cannot be mandated. Within Europe, the recommended external environmental noise level for schools ranges from 50 – 60 Leq dB. For example, in the Netherlands the Noise Nuisance Act suggests that dwellings, schools, and hospitals should be exposed to indoor levels no greater than 35 dBA Lcorrected, 24h and outdoor levels no greater than 50 dBA Lcorrected, 24h. The Environmental Management Act 1998 also specifies that between 7am and 7pm noise levels outside of schools should not exceed 50 dBA and inside should not exceed 35 dBA. Building regulations further stipulate that the average reverberation time in a classroom should not exceed 1.0 second. Standards for classroom acoustics are also available in some European countries. For example, in the UK Building Bulletin 93 (BB93) was introduced to govern the acoustical design on newly built schools and school building extensions (DfES, 2004). BB93 stipulates that in unoccupied primary and secondary school classrooms internal ambient noise levels should not exceed 35 dB Leq, 30 min. Reverberation times should not exceed 0.6 seconds and 0.8seconds for primary and secondary school classrooms, respectively. These regulations are compulsory for new school buildings. Article V. 2.6. Non-Auditory Effects on Children’s Health 2.6.1. Background While the focus of this review is on noise effects on children’s cognition and learning, other non-auditory effects on children’s health have been researched, including noise annoyance, psychological health, physiological function, and sleep disturbance. Noise could indirectly influence health in several ways. Acute noise exposure directly causes a number of predictable short-term physiological responses including increased blood pressure and endocrine outputs and it is proposed that chronic noise exposure may cause a longer-term activation of these responses resulting in subsequent symptoms and illness. Alternatively, if the noise levels are sufficiently high to cause significant annoyance, then it may be annoyance itself that activates these physiological responses. In some individuals noise annoyance may lead to stress responses and subsequent symptoms and illness. This section presents a brief overview of research in this field, summarizing the current strength of the evidence for the effects of aircraft noise on child health outcomes. 2-13

2.6.2. Noise Effects on Children’s Health (a) 2.6.2.1. Noise Annoyance. Noise annoyance is a multifaceted psychological concept including both evaluative and behavioural components used to describe negative reactions to noise (Guski, Schuemer and Felscher-Shur, 1999). Annoyance is an important health effect of noise (WHO, 2000) and is often the primary outcome used to evaluate the effect of noise on communities. Noise annoyance is a widely observed response to noise and this has also been examined with particular reference to children’s noise annoyance. Studies have consistently found evidence that exposure to chronic environmental noise causes annoyance in children (Bronzaft and McCarthy, 1975; Evans, Hygge and Bullinger, 1995). In the Munich Airport study noise- exposed children were found to be more annoyed by noise as indexed by a calibrated community measure (Evans, Hygge and Bullinger, 1995). In London, child-adapted standard self-report questions (Fields, de Jong and Brown, 1997) were used to assess annoyance showing higher annoyance levels in noise-exposed children (Haines, Stansfeld, Brentnall, et al, 2001). In a follow-up one year later the same result was found suggesting that annoyance effects were not subject to habituation (Haines, Stansfeld, Job, et al, 2001b). Studies have derived exposure-effect associations for the effects of aircraft and road traffic noise for adults (Midema and Oudshoorn, 2001) and the recent RANCH study derived exposure-effect relationships for children examining aircraft noise and road traffic noise (van Kempen, Lopez Barrio, Haines, et al, 2009). This study predicted for aircraft noise an increase in the percentage of severely annoyed children of around 5.1% at 50 dB to around 12.1% at 60 dB. The extent to which noise is found to be annoying appears to depend upon the extent to which the noise interferes with the activity in progress. Thus, activities involving speech communication such as conversation, watching television, or listening to the radio are the activities most disturbed by aircraft noise (Hall, Taylor and Birnie, 1985), thus annoyance could be an important end-point to consider in effects of aircraft noise on children’s learning. The RANCH study found that aircraft noise annoyance reduced the effect of aircraft noise exposure on reading comprehension to some extent, suggesting that noise annoyance may play a role in the mechanism for noise effects on reading comprehension (Clark, Martin, van Kempen, et al, 2006). (b) 2.6.2.2. Psychological Health. Given the effect of chronic noise exposure on annoyance responses, it has been proposed that chronic noise exposure could have an effect on psychological health, as prolonged annoyance could lead to poor psychological health (McLean and Tarnopolsky, 1977). Several studies have examined associations between noise exposure and children’s psychological health. The Tyrol Mountain study compared child and teacher ratings of psychological health for children exposed to ambient noise either <50 dBA Ldn or > 60 dBA Ldn (Lercher, Evans, Meis, et al, 2002). Ambient noise was associated with teacher ratings of psychological health but was only associated with child rated psychological health for children who also had early biological risk which was defined as low birth weight and/or premature birth. The West London School study of children attending schools near Heathrow airport also found that children who were exposed to aircraft noise had higher levels of psychological distress, as well as a higher prevalence of hyperactivity (Haines, Stansfeld, Brentnall, et al, 2001). However, 2-14

the RANCH study only partially replicated the West London School study findings in a larger sample drawn from the United Kingdom, the Netherlands, and Spain. While no associations were observed between either aircraft noise or road traffic noise and psychological distress, the association between aircraft noise exposure and hyperactivity was replicated (Stansfeld, Clark, Cameron, et al, 2009). As with studies of adults, overall the evidence suggests that chronic noise exposure is probably not associated with serious psychological illness and disorder but there may be effects on well-being and quality of life. This conclusion is however limited by the lack of longitudinal research in the field. Also few studies examine psychiatric diagnoses. Other research needs include the further exploration of whether hyperactive children are more susceptible to stimulating environmental stressors such as aircraft noise. (c) 2.6.2.3. Coronary Risk Factors. Recent years have seen a strengthening of evidence for effects of chronic noise exposure on adult cardiovascular outcomes such as hypertension and coronary heart disease (Babisch, 2006; Jarup, Babisch, Houthuijs, et al, 2008). A meta-analysis of adult studies suggests that a 5 dBA increase in noise is associated with a 25% increase in risk of hypertension compared with those not exposed to noise (Van Kempen, Kruize, Boshuizen, et al, 2002). However, epidemiological evidence for effects of noise on coronary risk factors for children is mixed. There is some evidence that chronic noise exposure may give rise to physiological effects in children in terms of raised blood pressure. In the Los Angeles Airport Study (Cohen, Evans, Krantz, et al, 1980), chronic exposure to aircraft noise was found to be associated with raised systolic and diastolic blood pressure. However, these increases although significant were within the normal range and were not indicative of hypertension. At follow-up a year later (Cohen, Evans, Krantz, et al, 1981) the findings were the same showing that these effects had not habituated. In the Munich study chronic noise exposure was found to be associated with both baseline systolic blood pressure and lower reactivity of systolic blood pressure to a cognitive task presented under acute noise (Evans, Hygge and Bullinger, 1995; Hygge, Evans and Bullinger, 2002). After the new airport opened a significant increase in systolic blood pressure was observed providing evidence for a causal link between chronic noise exposure and raised blood pressure (Hygge, Evans and Bullinger, 2002). No association was found between noise and diastolic blood pressure or reactivity. The RANCH study of 9-10-year- old children’s blood pressure around Schiphol and Heathrow airports found an effect of aircraft noise at home, as well as nighttime aircraft noise exposure on systolic and diastolic blood pressure but no effect for aircraft noise at school (van Kempen, Van Kamp, Fischer, et al, 2006). These findings suggest that it may be aircraft noise exposure in the evening and at night that are important for effects on children’s blood pressure. Overall, the evidence for noise effects on blood pressure is mixed and further research is required before conclusions can be drawn about noise effects on children’s blood pressure. (d) 2.6.2.4. Stress Hormones. Studies have examined the effect of chronic noise exposure on adrenaline, noradrenaline, and cortisol, all of which are released by the adrenal glands in situations of stress. Studies of noise effects on stress hormones typically have fairly small sample sizes. A further methodological limitation to these studies is that these stress hormones are notoriously difficult to assess, as urinary and salivary measures of these hormones are easily influenced by other unmeasured factors. For example, 2-15

cortisol has a diurnal variation and is usually high in the morning and low in the evening, making it difficult to measure effectively. Several airport noise studies have assessed the effects of noise on children’s endocrine disturbance. The Munich Airport study (Evans, Bullinger and Hygge, 1998; Evans, Hygge and Bullinger, 1995) examined overnight, resting levels of urinary adrenaline and noradrenaline. In the cross-sectional study at the old airport endocrine levels were significantly higher in the noise- exposed children, indicating raised stress levels. The longitudinal data also revealed a sharp increase in adrenaline and noradrenaline levels in noise-exposed children following the opening of the new airport. Cortisol levels were also examined but no significant differences were observed in either the cross-sectional or longitudinal data. This latter finding is consistent with the findings of the Heathrow Studies (Haines, Stansfeld, Brentnall, et al, 2001; Haines, Stansfeld, Job, et al, 2001a) which found no association between aircraft noise exposure above 66 dB LAeq16 and morning salivary cortisol measures (Haines, Stansfeld, Job, et al, 2001a), nor between aircraft noise exposure above 62 dB LAeq16 and twelve hour urinary cortisol, adrenaline and noradrenaline measures (Haines, Stansfeld, Brentnall, et al, 2001). Overall further studies of this field are required to clarify where effects may be observed between chronic aircraft noise exposure and stress hormones. The pathway for such effects also needs further clarification: there is a lack of understanding about whether raised endocrine responses represent normal short-term responses to environmental stress or whether they represent longer-term activation of the endocrine system. If they represent the latter, there is also a lack of understanding about how long-term activation of the endocrine system links to health impairment and whether endocrine responses can habituate to noise exposure. (e) 2.6.2.5. Sleep Disturbance. Exposure to nighttime noise can potentially interfere with the ability to fall asleep, shorten sleep duration, cause awakenings and reduced the perceived quality of sleep (Michaud, Fidell, Pearsons, et al, 2007), which could influence well-being causing annoyance, irritation, fatigue, low mood and impaired task performance (HCN, 2004). Recent reviews of evidence for noise effects on sleep disturbance in adults are available (Michaud, Fidell, Pearsons, et al, 2007; Midema and Vos, 2007) yet few studies of the effects of noise on sleep disturbance have included children. A sub-study of the RANCH study in a Swedish sample used sleep logs and actigraphy to compare the effect of road traffic noise on child and parent sleep, finding an exposure-effect relationship between noise exposure and sleep quality and daytime sleepiness for children and an exposure-effect relationship between noise and sleep quality, awakenings and perceived interference from noise for parents (Ohrstrom, Hadzibajramovic, Holmes, et al, 2006). While the findings of this study are suggestive of effects of noise exposure on children’s sleep disturbance, it is difficult to generalize from one Swedish study, where road traffic noise may not be as high as that found in other cities. In addition, longitudinal data is needed to confirm causal associations between noise exposure and sleep disturbance, as well as between noise exposure, sleep disturbance, and well-being related outcomes. Specific studies focusing on aircraft noise and children’s sleep disturbance are required. While the Munich and RANCH studies found that self-reported sleep disturbance did not mediate the association of aircraft noise exposure and cognitive impairment in children (Stansfeld, Hygge, Clark, et al, In press): whether the same results would be found with objective measures of sleep disturbances remains to be answered. 2-16

Sleep disturbance could be an important issue to consider further in future studies of the effects of aircraft noise on children’s learning. 2.7. Summary of the Noise Effects on Children’s Cognition and Learning The effect of environmental noise exposure on children’s cognitive performance and learning has been researched since the early 1970s. Studies are predominantly cross-sectional, examining associations between noise exposure at school and children’s cognitive performance or learning outcomes. Studies tend to focus on one type of noise exposure such as aircraft noise or road traffic noise. There are only a few longitudinal studies, some of which are before-and- after studies assessing naturalistic changes in noise exposure via insulation programmes or airport closures. To understand the causal pathways between noise exposure and learning, and design preventive interventions, there is a need to study associations longitudinally. Recent advancements in the field include the use of larger-scale epidemiological community samples and better characterization of noise measurement. Studies of noise effects on children’s cognition typically use established metrics of external noise exposure, such as Leq16 which indicates average noise exposure in dBA over a 16 hour daytime period. Studies of noise effects on children’s learning typically measure noise exposure but some recent studies model exposure using Geographical Information Systems. Overall, evidence for the effects of noise, and in particular aircraft noise, has strengthened and synthesized in recent years, with many studies demonstrating that children exposed to chronic aircraft noise exposure at school have poorer reading comprehension and memory than children who are not exposed. This evidence is predominantly cross-sectional but has also been confirmed with longitudinal data. Similarly, studies have shown that children who are exposed to chronic aircraft noise at school perform more poorly than their non-noise-exposed counterparts on nationally standardized tests. Demonstrating exposure-effect relationships between aircraft noise exposure and children’s cognition and learning is important for confirming causal associations between noise and cognition as well as for identifying thresholds for the effects. To date, only two studies have examined exposure-effect relationships: both suggest a linear relationship between aircraft noise exposure and children’s cognition. The most recent study suggests that reading comprehension begins to fall below average at aircraft noise exposure greater than 55 dB Leq16 but as the association is linear, any reduction in aircraft noise exposure should improve reading comprehension. Few studies have examined the influence of noise abatement, via insulation schemes or airport relocation, on children’s learning and cognition. Overall, these studies suggest that a reduction of noise exposure can eliminate previously observed cognitive deficits associated with noise but further studies in this area remain a priority. 2.7.1. Mechanisms Aircraft noise could directly influence cognitive skills such as reading comprehension, or noise effects could be accounted for by other mechanisms including teacher and pupil frustration and communication difficulties, learned helplessness, impaired attention, impaired memory, and noise annoyance. In general, research evidence for most of these mechanisms is quite limited. 2-17

2.7.2. Classroom Acoustics and Guidelines Recent years have seen research begin to explore the link between classroom acoustics and children’s learning outcomes. Research has focused on internal noise levels, reverberation times, and speech-to-noise ratios as important determinants of speech intelligibility within classrooms. Interestingly, younger children may require quieter conditions for optimum speech intelligibility compared with older children. Some recent studies have shown associations between classroom acoustics and performance on individually completed cognitive tests, as well as on nationally standardized tests. Guidelines for internal and external noise exposure at children’s schools exist in the US and in Europe. The majority of these guidelines are not statutory and suggest noise levels that apply to ambient environmental noise, rather than specific noise sources per se. Of particular relevance to the current project, is the ANSI standard for school acoustics (ANSI 2010), a voluntary code which specifies 35 dBA as a background noise limit for unoccupied classrooms and reverberation times between 0.6 seconds to 0.7 seconds depending upon the size of the classroom. 2.7.3. Noise Effects on Children’s Health There is also evidence for non-auditory effects of noise on children’s health. Evidence for the effect of aircraft noise on children’s annoyance responses is convincing, while evidence for effects on children’s psychological symptoms, blood pressure, stress hormones, and sleep disturbance is more moderate and would benefit from further and longitudinal study in child samples. 2-18