How much attention does noise cost? Concentration, memory and cognitive effort in a loud classroom

The instruction has three parts: take out your exercise books, copy the title, underline the date. At the same moment someone is finishing a conversation at the back, the corridor door opens and closes, and the previous sentence is still bouncing off the hard walls. Several pupils get to work, one asks for it to be repeated, two do only the first thing. It raises the question teachers ask themselves several times a day: was the pupil not paying attention - or did the information reach him in a form that cost too much to decipher? There is no single answer; a pupil’s behaviour can have many causes, and acoustic conditions are one of them. It is worth knowing what the research says about that one cause.

Attention is not an unlimited resource

Two concepts are enough to follow the rest of this text. The first is selective attention - the ability to pick out what matters at a given moment and push the rest aside. The second is working memory - the mechanism that lets us hold information briefly and do something with it at the same time. It is what allows a pupil to remember the first part of an instruction while listening to the second, and to combine them into a plan of action. Working memory is not the same as long-term memory: the question is not whether the child will remember the material for a test next week, but whether he can hold three pieces of information for twenty seconds.

Hearing the words and understanding the message are also two different things. Speech sounds can be received without any sense being built from them - especially when part of the signal was unclear and had to be reconstructed from context. Additional background sounds then compete with the information the task requires. Other people’s speech is particularly disruptive: it varies, it carries meaning and it is hard to ignore. There is no single working-memory “capacity” that could be given as a number of items - current models and measurement methods are more complex than the popular simplifications. What matters more is this: resources are limited, and whatever is spent at one stage is no longer available at the next. How large that cost is depends on the task the pupil happens to be doing.

What happens while listening in noise

Receiving an instruction in a loud room can be described as a sequence of steps. The pupil has to recognise the teacher’s voice among other sounds, separate it from the background, fill in the parts that were unclear, hold the earlier part of the utterance in memory, combine the successive pieces of information, understand the point of the task and finally start work. A difficulty at an early stage does not disappear - it carries forward. If deciphering the sound of the words consumed part of the resources, less is left for connecting the information and planning the work. The result need not be no response at all; more often it is a partial, delayed or guessed-at response.

This is a simplified, functional description, not a neurological model. It does show, though, why a child can “hear everything” and still carry out only the first of the instruction’s three steps.

Not all noise is the same

Research findings depend on which sound was studied. Pupils talking, unintelligible speech from the next room, the steady hum of ventilation, a single slammed door, a passing car, an aircraft overhead, the buzz from the corridor, the room’s reverberation and a teacher’s tired, hoarse voice are not one category. Two sounds with a similar average level in decibels can hinder work to very different degrees: what matters includes whether the sound is intelligible, whether it varies and whether it can be predicted. For that reason we note, with each finding in this article, what kind of noise it concerned. What the available data do not allow is a ranking of the “most harmful” sounds - studies usually compare selected conditions, not all of them at once.

What the research shows

Reviews and syntheses

In 2013 Maria Klatte, Kirstin Bergström and Thomas Lachmann published a short narrative review (not a meta-analysis) in Frontiers in Psychology [1]. The question was whether and how noise affects children’s learning. The authors organise the evidence by separating acute exposure - brief exposure while a task is being performed - from chronic exposure, meaning conditions at school or at home that last for years. In studies of acute exposure, noise lowered performance in speech perception and listening comprehension, but also in tasks that required no listening at all: recalling the order of visually presented items, and reading. On some tasks children proved more susceptible than adults. The authors themselves point out the limitations: the observed effects were sometimes small, some studies controlled confounding variables inadequately, and the stimuli and tasks differed between studies enough to make direct comparison difficult.

The 2022 systematic review with meta-analysis by Isabel Schiller and colleagues narrows the field to one question: speech processing [2]. It covered 31 studies of children aged 6-18 and analysed three areas - speech perception, listening comprehension and auditory working memory. Noise reduced the accuracy of responses on listening tasks in all three areas, and the size of the effect depended among other things on the signal-to-noise ratio, that is on how far the speaker’s voice rose above the background. The effect of reduced voice quality in the speaker was studied separately - it proved weaker than that of noise alone, but noticeable. These findings concern above all the processing of spoken language under controlled research conditions; they do not mean that every child in noise always makes more mistakes, or that every school task is disrupted to the same degree.

The most recent synthesis was published in 2025 by Gabriela Fretes and Ramon Palau in Applied Sciences [6]. This meta-analysis covered 21 studies and 152 effect sizes, concerning environmental and classroom noise and a range of functions: attention, memory, comprehension and overall school performance. The average effect was −0.46 (95% confidence interval −0.54 to −0.38), which the authors interpret as a moderate negative effect of noise on task performance. Such a figure has to be read with care: it averages very different kinds of noise, different age groups and different tasks, so it describes a general tendency in the literature rather than the effect to expect in a particular classroom. A synthesis of this kind supplements, rather than replaces, reading the primary studies.

Observational studies in schools

The most extensive study of cognitive development is the work of Maria Foraster and colleagues published in 2022 in PLOS Medicine [3]. It was a cohort study on a population-based sample: 2,680 children aged 7-10 from 38 schools in Barcelona, followed for about a year, from January 2012 to March 2013. During that time the children completed computerised tests of working memory, complex working memory and inattentiveness four times. The researchers measured road traffic noise both outside the schools and inside the classrooms, taking account not only of the average sound level but also of its variability, and estimated exposure at the children’s homes.

Higher exposure to road traffic noise at school was associated with slower development of working memory and greater inattentiveness over the year of observation; exposure at home showed no such association. Inside the classrooms it was above all the variability of the sound level, rather than the average itself, that was consistently associated with the results. The authors conclude that higher exposure to road traffic noise at school is associated with slower development of attention and working memory (Foraster et al., PLOS Medicine, 2022). This is an observational study - it shows an association; it does not prove that noise was the sole cause of the differences observed. It concerns road traffic noise reaching schools, not pupils talking during lessons, and the findings cannot automatically be transferred from one to the other. The “inattentiveness” measured in the tests is an indicator of task performance, not a diagnosis of any attention disorder. The authors themselves also note the limited generalisability of the results to other populations.

Earlier, in 2008, Bridget Shield and Julie Dockrell published in The Journal of the Acoustical Society of America an analysis linking the acoustic conditions of London primary schools with pupils’ results in standardised tests at the end of two key stages - at ages 7 and 11 [4]. Noise levels were measured outside schools in three London boroughs (142 schools in total), and internal measurements were also taken in eight selected schools; external L_Aeq levels ranged from 49 to 75 dB(A). The key move was to distinguish two sources: external noise and background noise in occupied classrooms. Higher external noise levels were associated with poorer test results, and the association was more pronounced among the older children in the analysis. This is an observational analysis of chronic exposure: schools in noisier locations may differ from others in other respects too - social composition, resources, the state of the building - and although some of these factors were controlled for statistically, their influence cannot be ruled out entirely.

An experiment: reading in simulated classroom noise

In 2019 Daniel Connolly and colleagues tested under experimental conditions how the level of classroom noise affects reading and vocabulary learning [5]. The study involved 976 pupils from English secondary schools (564 aged 11-13 and 412 aged 14-16), who read popular-science texts on laptops and answered test questions while listening to a recording of classroom noise through headphones. Two comparisons were made: 50 versus 70 dB L_Aeq, and 50 versus 64 dB L_Aeq. In the 70 dB condition all pupils performed significantly worse - both in the number of items attempted and in the accuracy of answers about facts and about the meaning of new words. At 64 dB the picture was less clear-cut: the decline affected only the older group. The authors note that disruption was evident at 70 dB, and that it is concerning to find negative effects at 64 dB in older pupils as well (Connolly et al., 2019) - adding that this is a level typical of an average secondary-school classroom. This does not mean that 64 or 70 dB is a universal threshold beyond which a child stops learning. These were particular tasks, a particular recording and a brief exposure delivered through headphones, not a reproduction of a whole day at school.

Why not every task is disrupted equally

Taken together, these studies reveal a regularity more important than any single number: the effect of noise depends on what the pupil is doing. What matters is the difficulty of the task, the pupil’s age, the type and predictability of the sound, whether the task requires listening at all, how heavily it loads working memory, whether an order has to be remembered, how long the exposure lasts, whether the instruction can be read again - and individual differences between pupils.

A simple comparison makes this clear. Copying a sentence from the board requires neither listening nor holding several pieces of information at once. Following a multi-step instruction requires both. Reading comprehension requires no listening, but it does require holding the thread - and it is disrupted nonetheless, as both the studies discussed in the 2013 review [1] and the 2019 experiment [5] showed. Remembering a new concept introduced orally combines the difficulty of listening with the difficulty of remembering. The most vulnerable tasks are those in which several pieces of information must be held and operated on at the same time. This does not mean that simple tasks are always immune to noise - research on the so-called irrelevant speech effect shows disruption in tasks that appear to have nothing to do with hearing [1].

Younger children may pay a higher price

Several things make younger pupils more susceptible to disruption: their ability to select information is still developing, their language experience is smaller, they find it harder to fill in missing parts of an utterance from context, their working memory is still maturing, and their knowledge of the lesson topic less often lets them guess what they did not hear. The 2013 review points to greater susceptibility in children than in adults on some tasks [1].

The picture is not simple, however. Development does not follow the same course or pace in everyone, and in some studies - as in the reading experiment [5] - it was the older pupils who proved more sensitive to the lower noise level, probably because they were doing harder tasks and attempting more of them. Age differences therefore depend on the task used and the conditions of the study, not on age alone.

A pupil will not always say he did not hear

In school practice, difficulty in receiving a message is rarely reported directly. Sometimes it shows in behaviour: frequently asking for repetition, checking what others are doing, carrying out only the first part of an instruction, starting work later, guessing, withdrawing from listening-based tasks, apparently “switching off”, or marked tiredness after lessons that demand intensive listening.

This has to be said unambiguously: none of these behaviours on its own proves an acoustic problem. Each can have many other causes, and none is grounds for identifying hearing loss, ADHD, an auditory processing disorder or any other difficulty - that is a job for specialists. The signal is different: if such behaviours recur in a particular room, with a particular type of task, or in several pupils at once, acoustic conditions are one of the things worth checking.

Noise does not explain everything

A pupil’s concentration and results are shaped by many factors at once: sleep, stress, mood, relationships in the class, how the lesson is run, the difficulty of the material, individual educational needs, hearing loss, language difficulties, the situation at home, the temperature and air quality in the room, and other stimuli in the surroundings. None of the studies discussed claims that acoustics explain all the differences between pupils - on the contrary, the authors consistently control for some of these variables and describe the uncertainty that remains.

This is not an argument for playing acoustics down. It is an argument for treating them as one of the few factors a school can actually influence - and one which, unlike many others, can be measured.

Summary

Let us return to the three-part instruction from the beginning. The question “why didn’t the pupil do it?” is often narrowed to motivation and behaviour, and that is only one of the possible answers. The research does not allow us to claim that noise causes difficulties with concentration or damages memory. It does allow something more cautious and more useful: noise does not have to make a task impossible in order to make it harder. It can mean that a larger share of limited attentional resources goes on selecting, ordering and deciphering information - and less is left for what the pupil came to the lesson for.

Good acoustics do not guarantee concentration. They do reduce the cognitive cost of taking in basic information - and that is the condition everything else starts from.

References

1. Klatte M., Bergström K., Lachmann T., Does noise affect learning? A short review on noise effects on cognitive performance in children, „Frontiers in Psychology” 2013; 4: 578. DOI: 10.3389/fpsyg.2013.00578; PMID: 24009598.
2. Schiller I.S., Remacle A., Durieux N., Morsomme D., Effects of Noise and a Speaker's Impaired Voice Quality on Spoken Language Processing in School-Aged Children: A Systematic Review and Meta-Analysis, „Journal of Speech, Language, and Hearing Research” 2022; 65(1): 169-199. DOI: 10.1044/2021_JSLHR-21-00183; PMID: 34902257.
3. Foraster M., Esnaola M., López-Vicente M. i wsp., Exposure to road traffic noise and cognitive development in schoolchildren in Barcelona, Spain: A population-based cohort study, „PLOS Medicine” 2022; 19(6): e1004001. DOI: 10.1371/journal.pmed.1004001; PMID: 35653430.
4. Shield B.M., Dockrell J.E., The effects of environmental and classroom noise on the academic attainments of primary school children, „The Journal of the Acoustical Society of America” 2008; 123(1): 133-144. DOI: 10.1121/1.2812596; PMID: 18177145.
5. Connolly D., Dockrell J., Shield B., Conetta R., Mydlarz C., Cox T., The effects of classroom noise on the reading comprehension of adolescents, „The Journal of the Acoustical Society of America” 2019; 145(1): 372-381. DOI: 10.1121/1.5087126; PMID: 30710912.
6. Fretes G., Palau R., The Impact of Noise on Learning in Children and Adolescents: A Meta-Analysis, „Applied Sciences” 2025; 15(8): 4128. DOI: 10.3390/app15084128.

This article is for information only. It does not replace an acoustic measurement, a medical diagnosis or a consultation with a psychologist or educational specialist.