Page 31 - Fall 2005

“It does not appear...that classrooms are being built with the acoustic needs of
children in mind. It seemed clear to the working group that a standard could lead the way toward more favorable learning conditions for students.”
closest to the source are most favorable, but for those a few meters from the source, the SNR can approach zero. A number of research reports (including Soli and Sullivan, 1997) have shown that while adults can understand famil- iar speech under conditions of an SNR = 0 dB, children cannot.
It does not appear, then, that classrooms are being built with the acoustic needs of children in mind. It seemed clear to the working group that a standard could lead the way toward more favorable learning condi- tions for students.
How do teachers handle classroom noise?
Teachers seem aware of the noise
problem in schools, and adapt their teaching styles to com- pensate for poor acoustics. In a government survey, teachers reported that classroom noise was the most common build- ing facilities problem they faced (U.S. GAO report, 1995). Teachers in the Los Angeles Unified School District told working group members in 1999 that they were forced to alternate time between teaching and child comfort when noisy window air-conditioning units were installed in thou- sands of classrooms. Teachers report frequent voice prob- lems as they attempt to overcome background noise prob- lems in their classrooms. Smith et al. (1998) reported that 32% of teachers reported occasional voice problems, with 20% reporting that they had missed work occasionally because of vocal fatigue.
Some school districts have turned to electronic amplifi- cation systems to attempt to increase the teacher’s source voice level above that of the background noise. Although amplification systems can be helpful under specific circum- stances (especially lecture-style classrooms in nonreverber- ant spaces), they have obvious limitations for active, child- based learning strategies where multiple signal sources need to be heard. (Recall for a moment the last large-group meet- ing you attended in which certain members refused to wait for the microphone to be passed when speaking.) In addi- tion, it seems obvious that if walls dividing classrooms are insufficient for noise reduction, amplifying two rooms on opposite sides of this inadequate wall would not provide a satisfactory solution. Also, based on well-controlled studies of speech in noise at a range of overall levels, we firmly believe that quieter is better. Even if one could provide a strong, positive signal-to-noise ratio by elevating individual voice levels above background noise levels of 65 dBA, con- siderable research suggests that this is not optimal for speech recognition (Studebaker et al., 1999) nor for student and teacher fatigue. The best solution to all of these problems is to build quiet schools in the first place.
Members of the working group, then, were convinced of the need for improved acoustics in schools. We then looked for evidence that would guide the setting of performance
standards so that students could be assured full acoustic access to the spo- ken messages in classrooms.
Why do children need favorable acoustics?
The first, most over- whelming finding of the extensive liter- ature search was that all children need favorable acoustics for learning. It seems clear from extensive research that children do not detect, discrimi- nate, and understand signals in back- ground noise and reverberation in the same way as do adults. Children are significantly more negatively affected by noisy and reverberant rooms than are adults in those rooms. The younger the child, the more negative are the effects of background noise and reverberation. Adult per- formance levels, then, should not dictate the acoustic needs
of children in schools.
This article is not intended to be a detailed description of
the research supporting these claims. (For more detailed summaries and thorough bibliography for this material, please refer to Nelson, Soli, and Seltz, 2003). Briefly, though, several theories of developmental psychoacoustics have emerged that may explain the findings. Much of this good theoretical work in developmental psychoacoustics has been done by members of the ASA. Some of this excellent work will be referenced here.
A consistent finding throughout the studies of young children is that children start as inefficient listeners and con- tinue to develop as selective listeners well into adolescence. As young infants, children seem to be nonselec- tive–broadband listeners who are equally adversely affected by background noises of all types. They don’t yet know what signals to ignore and when to focus attention. Adults typi- cally experience masking from narrow bands of noise that are similar in frequency to the target sound (energetic masking). When two sounds (a masker and a target) that are close in frequency reach the cochlea (inner ear) the two stimulation patterns overlap. The presence of the masker sound, then, diminishes the ability to detect the target signal. For infants and young children, however, the stimulation patterns need not overlap in order for one sound to interfere with the detec- tion of another. Even when a noise is remote in frequency from the target sound and the excitation patterns are well separated from one another, the infant is less likely to detect the target sound (Werner and Bargones, 1991).
At times adults are confused and distracted by complex background noise, even when that noise does not cause ener- getic masking in the cochlea. Wightman et al. (2003) recent- ly showed that adults and children can experience significant amounts of masking by pure tone complexes, even when those tones are not close in frequency to the target signal tone. This has been called “informational” as opposed to “energetic” masking. In their investigation, younger children
Classroom Acoustics 29