The child hears but does not understand. How reverberation strips speech of clarity in the classroom

The teacher speaks from the board: open your textbook at page forty-two and read the first paragraph. The front two rows reach for their books at once. In the third row someone asks for the page number; in the back row two pupils glance at their neighbours. The teacher repeats it louder - and part of the class still is not sure whether it was page forty-two or forty-three. It raises a question rarely asked in the staff room: is the problem a voice that is too quiet, or what the room itself does to that voice on its way to the pupil?

Reverberation is not the same thing as noise

When a teacher speaks a sentence, what reaches the pupil first is the direct sound - the part of the wave that travels the shortest path from the speaker’s mouth to the listener’s ear. Immediately behind it come reflections from walls, ceiling, floor, board and furniture. In a typical room there are a great many of them, and they overlap to form reverberation - a gradually decaying “cloud” of sound that persists after the speaker has finished the syllable.

Reverberation is not an echo. We hear an echo when a single reflection returns late enough to be perceived as a separate repetition of the sound - which rarely happens in classrooms. Reverberation is continuous and usually unnoticed: we do not hear it as a “second voice”, but as a blurring of the first.

Two things that blur into one in discussions about schools are worth separating straight away. Noise level answers the question “how loud is it in this room”. Reverberation time answers the question “how long does sound persist in this room after the source stops”. These are two independent properties. An empty, high-ceilinged room with hard walls can be very quiet and at the same time have very poor speech intelligibility, because every word spoken “hangs” in the air for a moment and overlaps the next. The average sound level in decibels therefore does not describe a room’s acoustic quality - it describes only one dimension of it.

Take a short instruction: “underline the date”. In a room with well-controlled reverberation the successive syllables stay separate and arrive cleanly. In a room with long reverberation the tail of “un-” runs into “-der-”, and the ending of “date” is lost in the remains of the previous word. The pupil hears that the teacher is speaking - but not every element of the message arrives in a clear form.

This does not mean reflections are harmful in themselves. Reflections that arrive very soon after the direct sound are merged with it by the ear and reinforce audibility, especially towards the back of the room. The problem is energy arriving late, and reverberation that is too long and uncontrolled.

Why louder does not always mean clearer

Intelligibility is determined above all by the relationship between the speaker’s voice and the background - the signal-to-noise ratio (SNR). The signal here is the teacher’s voice; the noise is everything competing with it: pupils talking, ventilation and other building services, sounds from the corridor, traffic outside the window, equipment running in the room.

Direct sound weakens with distance. A pupil at the back receives noticeably less of it than one at the front - whereas reverberation and background sounds are distributed through the room far more evenly. That is why, in the same room, listening conditions can differ greatly depending on where a child sits.

Raising the voice improves this balance only partly. Louder speech also puts more energy into the room, and therefore produces stronger reverberation - audibility improves, clarity not necessarily. And if the class responds to a louder teacher by talking more loudly, the background level rises along with the signal and the voice’s advantage over the background does not change. A particularly difficult kind of background is several people speaking at once: it carries meaning, it fluctuates and it is hard to ignore - which is why it disrupts differently from steady noise at the same level.

The child hears the words but loses parts of them

Speech reception is not an all-or-nothing affair. You can hear that someone is speaking without recognising every word; you can recognise the words without assembling them into the sense of a sentence; you can understand the sentence and still not retain a whole three-part instruction. Reverberation and noise hit hardest at the elements of speech that are short and quiet: inflectional endings, fricatives and stops, the differences between similar-sounding phonemes. What suffers are short words, numbers, proper names and new terms - exactly the things a pupil cannot reconstruct from context, because he is hearing them for the first time.

The scale of this difficulty is not the same for everyone. It depends on age, on knowledge of the language and the topic, on hearing, and on whether the pupil can see the speaker’s face. Two children sitting next to each other can, in the same conditions, receive messages of differing completeness.

What an acoustician actually measures

Reverberation time

Reverberation time describes how fast sound energy decays in a room - historically defined as the time in which the sound level falls by 60 decibels after the source is switched off (hence the label RT60). In practice a full 60 dB decay is rarely captured, so it is derived from shorter sections of the decay curve, labelled T20 and T30, and reported separately for frequency bands. The measurement follows a standardised procedure, with a controlled sound source and at many points in the room. A hand clap or a phone app is not a measurement method - at most it can suggest that the room is worth checking properly.

Speech intelligibility and the STI

The STI (speech transmission index) is a number from 0 to 1 describing how faithfully a room transmits those features of the speech signal that carry its legibility. It is not the percentage of words a particular person will understand: it is a technical parameter of the transmission path, derived from measurement, not the result of a test on people. A high STI means good conditions, but does not guarantee that every pupil will understand everything - and conversely, the STI alone is not enough to assess every type of room.

Sound insulation versus acoustic treatment

These are two different things, and in school practice the most frequently confused. Sound insulation concerns how much sound penetrates into the room from outside and from adjacent spaces - it depends on the building elements: walls, floors, doors, windows. Acoustic treatment concerns what happens to sound already inside the room - the control of reflections and reverberation using absorbing and diffusing surfaces. Everyday words like “soundproofing” blur the two, which is why they are best avoided in documentation.

Background noise is measured separately: it is the sound level in an unoccupied room, made up of ventilation and other services, road traffic, the buzz from the corridor, sound from neighbouring rooms and equipment running in the classroom. One rule is worth remembering: good sound insulation does not guarantee a good reverberation time, and a short reverberation time does not automatically mean good insulation from the corridor or the street. These are separate problems, separate measurements and separate solutions.

What the research shows

Systematic review (Murgia et al., 2023). A team from the University of Illinois asked how speech intelligibility in real primary-school classrooms relates to their acoustic parameters [1]. The review included 23 publications in which the participants were primary-school children, the room’s acoustic characteristics were described and intelligibility tests were carried out. The picture emerging from these studies is consistent: longer reverberation times and higher background noise levels were associated with worse conditions for receiving speech. The authors conclude that good classroom acoustics are needed to improve speech intelligibility and thereby children’s educational opportunities (Murgia et al., Systematic Review of Literature on Speech Intelligibility and Classroom Acoustics in Elementary Schools, 2023). Limitation: the included studies differed in methods, room types and pupil groups, which makes direct comparison - and transfer to a particular room - difficult.

An experiment in virtual classrooms (Klatte, Lachmann and Meis, 2010). The research question was how noise and reverberation affect speech perception and listening comprehension - and whether children are more susceptible than adults [2]. The experiment involved first- and third-grade pupils and adults. Speech perception was tested by matching a heard word to a picture, and listening comprehension by carrying out spoken instructions. The experiment was run in two virtual rooms with mean reverberation times of 0.47 s and 1.1 s, with and without background sounds. Children were disrupted more than adults by background sounds on both tasks. The key nuance: a longer reverberation time on its own did not impair speech perception in quiet, but it clearly amplified the disruption caused by background sounds - in every age group. Reverberation therefore does not act as an independent factor with a fixed effect; its significance emerges in combination with noise. Limitation: these were simulated conditions, not measurements in real classrooms during lessons.

An experiment in a simulated classroom (Valente et al., 2012). A team from Boys Town National Research Hospital examined how acoustic conditions affect not only the recognition of single sentences but also longer tasks resembling a real lesson [3]. School-age children listened to material in a simulated classroom environment in which two reverberation times (about 0.6 and 1.5 s) were crossed with two signal-to-noise ratios. The tasks covered sentence repetition (simple reception) and comprehension of a longer passage (complex reception). The results indicated that worsening acoustic conditions hit tasks requiring comprehension of longer content harder than sentence recognition alone. The implication for schools is direct: testing audibility with a single sentence may not reflect what happens during a twenty-minute discussion of a topic. Limitation: a simulated classroom offers experimental control, but is not an exact equivalent of any particular school room.

A small change in reverberation, a measurable effect (Prodi and Visentin, 2022). Italian researchers asked whether a small change in reverberation time - from 0.57 to 0.69 s, that is within the range considered acceptable - is noticeable at all in task performance [4]. The experiment involved 302 pupils aged 11-13 with normal hearing. They performed three typical school tasks: speech perception, sentence comprehension and mental arithmetic, in two conditions: in quiet and with classroom noise. Besides accuracy, response times were measured as a behavioural indicator of listening effort. The effect of longer reverberation showed up mainly in the presence of noise, and depended on the type of task and the pupils’ age. The practical conclusion is cautious: a difference of 0.12 s looks like a detail in a measurement report, yet it can still be measurable in behaviour - which does not mean that every such difference will always produce a given drop in performance in every child.

Field measurements and tests with pupils (Rabelo et al., 2014). This Brazilian cross-sectional study combined acoustic measurements with tests taken by children [5]. Equivalent sound level (Leq), reverberation time T30 and the STI were measured in 18 classrooms in 9 public schools in Belo Horizonte, and a speech intelligibility test was administered to 273 pupils with a mean age of 9.4 years. Pupils performed better in rooms with lower noise levels, shorter reverberation times and higher STI. The authors conclude that acoustic parameters have a direct effect on pupils’ speech intelligibility (Rabelo et al., Effect of classroom acoustics on the speech intelligibility of students, 2014). Limitation: this is a cross-sectional study describing the co-occurrence of parameters and results at one moment; none of the three parameters explained the results on its own, because in practice they change together.

Pupils using hearing aids (Iglehart, 2020). This study looked at speech perception in conditions replicating classroom acoustics among children with hearing loss who wear hearing aids [6]. The reference point is the American acoustic standard for schools, which specifies a shorter reverberation time for rooms intended for pupils with hearing loss than for general classrooms. Findings in this line of research show why that stricter requirement makes sense: pupils using hearing aids need more favourable conditions to reach a level of comprehension comparable to their peers. Two caveats matter here. First, tests conducted in a booth with very little reverberation and with the signal delivered under control do not fully replicate a real room with movement, conversation and a varying distance from the speaker. Second, findings about children with hearing loss must not be transferred to the whole class - they describe the needs of a specific group.

Guidance, not a study: BB93. The UK Department for Education document Acoustic design of schools: performance standards (BB93) is a design standard, not the result of a scientific study [7]. It distinguishes room types and, for each, sets permissible background noise, sound insulation and a maximum mid-frequency reverberation time; for open-plan spaces it also sets requirements for speech intelligibility expressed as an STI. For newly built primary-school classrooms it specifies a reverberation time no longer than 0.6 s (room finished and furnished, but unoccupied); for rooms intended for pupils with particular hearing or communication needs - 0.4 s. For refurbishment of existing buildings the document allows more lenient values. These requirements apply in England and Wales; they are not Polish law and should not be transferred directly into Polish regulation - what they do show is how requirements are differentiated in practice according to the room’s purpose.

The Polish standard. In Poland, reverberation conditions and speech intelligibility in rooms are governed by the standard PN-B-02151-4:2015-06 [8], referenced in the technical building regulations. It sets requirements for many types of room found in schools, together with guidance on testing. The standard is a paid document distributed by the Polish Committee for Standardization - it is not a free public document, which is why we do not reproduce its tables or detailed values here. The standard’s current status (including any amendments) should be checked in the PKN catalogue each time.

Who finds reverberation hardest

The effects of poor conditions are unevenly distributed. In experimental studies younger children were more susceptible to disruption than adults under the same conditions [2]. Pupils with hearing loss, including those using hearing aids, need more favourable conditions to reach a comparable level of comprehension - which is reflected in the stricter requirements for rooms dedicated to them [6][7]. Pupils with less language experience are in a similar position: children with language difficulties and those learning in a language that is not their first have less scope to fill in unclear fragments of an utterance from context. It is also harder for pupils for whom listening in noise costs more effort and who tire faster in such conditions; research on listening effort shows that its indicators respond to acoustic conditions even when task accuracy has not yet changed [4].

These are not homogeneous groups and they do not respond identically. Nor are they diagnostic categories - observing behaviour in class allows nothing to be identified. What they do show is for whom the margin of error in a room’s acoustics is smallest.

What should not be promised

Good acoustics will not make every pupil attentive all the time - attention depends on many factors, which we covered in the previous article on attention and working memory. A short reverberation time will not remove noise coming from the corridor: that is a job for the building fabric, not for absorbing surfaces. Sound-absorbing materials will not substitute for the insulation of walls, floors and doors. One measurement does not answer every question - different parameters determine the quality of a classroom, a sports hall, a canteen or a corridor. And finally: meeting one parameter does not mean the room is good in every respect.

Summary

Let us return to the room we began with. The pupil in the back row did not hear the page number, not because the teacher spoke too quietly, and not necessarily because he was inattentive. Between her voice and his ear stood a room that changed that voice: it weakened the direct sound, added reverberation of its own and mixed it with the background. The research does not allow us to claim that every misunderstanding in class has such a cause. It does allow us to say that in rooms with long reverberation and a high background level part of the message simply does not arrive in a clear form - and that this affects different pupils unequally.

Effective communication in the classroom is therefore not solely a matter between teacher and pupil. The building takes part in it too. Good acoustics will not replace pedagogy, but they mean that an instruction given once has a better chance of reaching more of the class in a form that can be understood the first time.

References

1. Murgia S., Webster J., Cantor Cutiva L.C., Bottalico P., Systematic Review of Literature on Speech Intelligibility and Classroom Acoustics in Elementary Schools, „Language, Speech, and Hearing Services in Schools” 2023; 54(1): 322-335. DOI: 10.1044/2022_LSHSS-21-00181; PMID: 36260411.
2. Klatte M., Lachmann T., Meis M., Effects of noise and reverberation on speech perception and listening comprehension of children and adults in a classroom-like setting, „Noise & Health” 2010; 12(49): 270-282. DOI: 10.4103/1463-1741.70506; PMID: 20871182.
3. Valente D.L., Plevinsky H.M., Franco J.M., Heinrichs-Graham E.C., Lewis D.E., Experimental investigation of the effects of the acoustical conditions in a simulated classroom on speech recognition and learning in children, „The Journal of the Acoustical Society of America” 2012; 131(1): 232-246. DOI: 10.1121/1.3662059; PMID: 22280587.
4. Prodi N., Visentin C., A Slight Increase in Reverberation Time in the Classroom Affects Performance and Behavioral Listening Effort, „Ear and Hearing” 2022; 43(2): 460-476. DOI: 10.1097/AUD.0000000000001110; PMID: 34369418.
5. Rabelo A.T.V., Santos J.N., Oliveira R.C., Magalhães M.C., Effect of classroom acoustics on the speech intelligibility of students, „CoDAS” 2014; 26(5): 360-366. DOI: 10.1590/2317-1782/20142014026; PMID: 25388068.
6. Iglehart F., Speech Perception in Classroom Acoustics by Children With Hearing Loss and Wearing Hearing Aids, „American Journal of Audiology” 2020. DOI: 10.1044/2019_AJA-19-0010; PMID: 31835909.
7. Department for Education (United Kingdom), Acoustic design of schools: performance standards (Building Bulletin 93), February 2015. Dokument PDF (GOV.UK)
8. PN-B-02151-4:2015-06 Akustyka budowlana. Ochrona przed hałasem w budynkach. Część 4: Wymagania dotyczące warunków pogłosowych i zrozumiałości mowy w pomieszczeniach oraz wytyczne prowadzenia badań, Polski Komitet Normalizacyjny, 2015 (dokument odpłatny; status do sprawdzenia w katalogu PKN).

This article is for information only. It does not replace an acoustic measurement, an acoustic treatment design or an individual assessment of a room.