747398

research-article2018
PPSXXX10.1177/1745691617747398Vasilev et al.Auditory Distraction During Reading

Perspectives on Psychological Science

Auditory Distraction During Reading: 1­–31

© The Author(s) 2018

A Bayesian Meta-Analysis of a Reprints and permissions:

https://doi.org/10.1177/1745691617747398

Continuing Controversy sagepub.com/journalsPermissions.nav

DOI: 10.1177/1745691617747398
www.psychologicalscience.org/PPS

Martin R. Vasilev , Julie A. Kirkby, and Bernhard Angele

Department of Psychology, Bournemouth University

Abstract
Everyday reading occurs in different settings, such as on the train to work, in a busy cafeteria, or at home while
listening to music. In these situations, readers are exposed to external auditory stimulation from nearby noise,
speech, or music that may distract them from their task and reduce their comprehension. Although many studies have
investigated auditory-distraction effects during reading, the results have proved to be inconsistent and sometimes
even contradictory. In addition, the broader theoretical implications of the findings have not always been explicitly
considered. We report a Bayesian meta-analysis of 65 studies on auditory-distraction effects during reading and use
metaregression models to test predictions derived from existing theories. The results showed that background noise,
speech, and music all have a small but reliably detrimental effect on reading performance. The degree of disruption
in reading comprehension did not generally differ between adults and children. Intelligible speech and lyrical music
resulted in the biggest distraction. Although this last result is consistent with theories of semantic distraction, there was
also reliable distraction by noise. It is argued that new theoretical models are needed that can account for distraction
by both background speech and noise.

Keywords
reading, background noise, speech, music, meta-analysis

Reading is a critical skill that is indispensable in modern music affects the reading process. Although some inter-
society. Although reading performance is best in silence esting patterns of results have emerged, the research
when no distracting stimuli are present, such ideal con- literature has been undermined by a fair number of
ditions are rarely typical for daily life. Rather, much of inconsistent findings and the general lack of broader
everyday reading occurs in the presence of external theoretical frameworks that can explain how auditory
auditory stimulation, such as noise from nearby traffic, distraction during reading occurs. Although a number
music playing in the background, or a colleague talking of theoretical accounts have been developed for simpler
on the phone. The interest in how auditory stimuli tasks such as serial recall, it is currently not known how
affect human performance is almost as old as modern well they can account for all the findings from reading-
psychology itself (e.g., Cassel & Dallenbach, 1918; comprehension tasks that have accumulated over the
Morgan, 1917). From the widespread use of personal past several decades. In addition, because of the mixed
radios among students in the 1940s (Henderson, Crews, findings on some topics, the actual magnitude of
& Barlow, 1945; L. R. Miller, 1947) to the rise in popu- auditory-distraction effects—or even if they are reliably
larity of the TV (Armstrong, Boiarsky, & Mares, 1991; different from zero—is currently not well understood.
Cool, Yarbrough, Patton, Runde, & Keith, 1994) and
mobile devices (Kallinen, 2002), researchers and educa-
tors alike have been interested in whether background
sounds can distract students from reading and other Corresponding Author:
Martin R. Vasilev, Department of Psychology, Bournemouth
study-related tasks. University, P104, Poole House, Talbot Campus, Fern Barrow, Poole,
Over the past 8 decades, many studies have exam- Dorset, BH12 5BB, United Kingdom
ined how experimental exposure to speech, noise, and E-mail: mvasilev@bournemouth.ac.uk
2 Vasilev et al.

In the present article, we address these issues in two and the environmental noise were combined. Using
ways. First, we present the first attempt to make a sta- similar sound stimuli, Ljung, Sorqvist, and Hygge (2009)
tistical synthesis of previous findings in a reading task found that road-traffic noise impaired the reading speed
to find out whether and to what extent auditory stimuli of 12- and 13-year-old children, but not their reading
can interfere with reading performance. To do this, we comprehension. However, a condition of children’s
adopted a Bayesian meta-analysis approach that makes babble intermixed with irrelevant speech affected nei-
it possible to quantify the degree of belief, given the ther measure.
data, that background sounds can disrupt reading. Sec- Studies of exposure to noise in adults have resulted
ond, we used Bayesian metaregression models to test in similarly mixed findings, sometimes even when the
the predictions derived from existing theories on audi- materials were identical (e.g., Martin, Wogalter, & Forlano,
tory distraction and to estimate how likely it is that they 1988, Experiments 4 and 5). Although most studies have
can explain the available data. The present article starts failed to find an effect of acoustical or environmental
with a brief overview of the literature that highlights noise on reading comprehension (Gawron, 1984; Jahncke,
the existing inconsistencies. Then, we consider theories Hygge, Halin, Green, & Dimberg, 2011; R. Johansson,
that can explain auditory-distraction effects during Holmqvist, Mossberg, & Lindgren, 2012; Veitch, 1990),
reading. Finally, the predictions from these theories are others have found such an effect after examining the
outlined and tested. mediating role of personality characteristics, such as intro-
version and extroversion (Furnham, Gunter, & Peterson,
1994; Ylias & Heaven, 2003). In summary, studies inves-
The Effect of Background Noise, tigating the effect of background noise on reading com-
Speech, and Music on Reading: prehension have yielded inconsistent results, although
An Overview some of them suggest that exposure to noise may be
detrimental.
Background noise
Background noise can be defined as any unwanted
Background speech
sounds that are not related to the reading task. Strictly
speaking, some degree of background noise is always Background speech is a specific kind of noise that often
present during reading; however, the intensity of the occurs in daily life. Compared with environmental and
background noise can vary enormously depending on acoustical noise, background speech has specific acous-
the environment. A number of epidemiological studies tic properties that make it salient to listeners. In addi-
have investigated the relationship between constant tion, if the background speech is intelligible, it also
exposure to noise and reading and have suggested that carries semantic meaning (completely unintelligible
long-term exposure to traffic noise is associated with background speech might also occur, but it is not very
lower reading ability in children (e.g., Haines, Stansfeld, frequently encountered unless one is in a foreign coun-
Job, Berglund, & Head, 2001b; Hygge, Evans, & try and does not understand the language). Perhaps
Bullinger, 2002; Papanikolaou, Skenteris, & Piperakis, owing to its semantic content, background speech is
2015; Stansfeld et al., 2005). Note, however, that only often rated as more distracting and more annoying than
very few studies have examined the effect of short-term acoustical noise (Haapakangas et al., 2011; Haka et al.,
experimental exposure to noise. 2009; Landström, Söderberg, Kjellberg, & Nordström,
In one early study, C. R. Johansson (1983) found that 2002). Consistent with this subjective perception, intel-
the reading comprehension and reading speed of ligible background speech has been found to disrupt
10-year-old children did not differ between quiet condi- reading comprehension in a number of experiments
tions and conditions of continuous or intermittent (Armstrong et al., 1991; Baker & Madell, 1965; Martin
acoustical noise. More recently, Dockrell and Shield et al., 1988; Sörqvist, Halin, & Hygge, 2010; however,
(2006) investigated the effect of typical classroom noise see Venetjoki, Kaarlela-Tuomaala, Keskinen, & Hongisto,
(which is quite different from acoustical white or pink 2006). In addition, some evidence suggests that this
noise) on reading comprehension in 8-year-old chil- disruption effect may be larger for participants who
dren. Participants completed the Suffolk Reading Scale have a poorer ability to immediately suppress the irrel-
in one of three conditions: silence, noise consisting of evant background speech (Sörqvist, Halin, & Hygge,
children’s babble, and the same babble combined with 2010; Sörqvist, Ljungberg, & Ljung, 2010).
intermittent environmental noise. The results showed A specific reading task that has been investigated in
that children performed better in the quiet condition more detail in connection with background speech is
than in the babble noise condition. Surprisingly, how- proofreading. Proofreading is an important part of
ever, reading performance was best when the babble many professions, especially those related to teaching
Auditory Distraction During Reading 3

and publishing. Proofreading is a more cognitively a reading/proofreading task was disrupted by back-
demanding task than reading alone because it also ground speech only when the text was formatted in a
requires allocating attention to look for mistakes in familiar font, but not when it was formatted in an unfa-
addition to reading the text. There are generally two miliar font (i.e., one that was more difficult to read).
types of mistakes that have been investigated in proof- Likewise, performance was disrupted only when the
reading studies: contextual mistakes, which require text was printed normally, but not when it was visually
understanding the meaning of the text to detect (e.g., degraded (i.e., harder to read). Therefore, these results
problems with pronoun agreement), and noncontextual suggest that increasing task engagement may decrease
mistakes, which require only processing of the current the detrimental effect of background speech on reading
word to detect (e.g., spelling mistakes). Because of the comprehension and proofreading accuracy (for a dis-
semantic content of intelligible speech, it can be cussion, see Sörqvist & Marsh, 2015).
hypothesized that background speech would disrupt Most studies that have been considered so far have
the detection of contextual errors more than the detec- investigated only the end product of reading and proof-
tion of noncontextual errors. reading (i.e., comprehension accuracy, proofreading
Some support for this prediction was found by an accuracy, or the overall time taken to read the text).
early study by Weinstein (1977), who reported that However, these studies do not tell us how the reading
background speech consisting of a radio news report process is influenced on a moment-to-moment basis.
significantly impaired the detection of contextual errors More recently, several eye-tracking studies have
but not the detection of noncontextual errors. However, addressed this question by showing that the effect of
Jones, Miles, and Page (1990) found exactly the oppo- background speech on reading can also be found at
site effect in another study. The authors manipulated the level of fixation durations and fixation probabilities
both the intelligibility of background speech (which (Cauchard, Cane, & Weger, 2012; Hyönä & Ekholm,
was played either normally or in reverse) and the inten- 2016; Vasilev, Liversedge, Rowan, Kirkby, & Angele,
sity of the sound (50 vs. 70 dBA). They found that the 2017; Yan, Meng, Liu, He, & Paterson, 2017). One key
intensity of the sound did not affect proofreading per- finding from these studies is that background speech
formance but that normal (i.e., intelligible) speech leads to an increased number of rereading fixations.
reduced the number of noncontextual errors that were Although these studies have been successful in explain-
detected. Critically, however, the intelligibility of speech ing when disruption by background speech occurs dur-
did not affect the detection of contextual errors ( Jones ing the reading process, one puzzling aspect is that
et al., 1990). More recently, Venetjoki et al. (2006) found none of the eye-tracking experiments have replicated
that background speech reduced the overall accuracy the disruption effect in comprehension accuracy found
on a similar proofreading task compared with continu- in behavioral studies. Why this inconsistency exists
ous noise. However, even though the task included remains unknown, but it raises questions about the
both contextual and noncontextual errors, there was reliability of the effect of background speech on read-
no significant effect of background speech on either ing comprehension.
error type in isolation. In a similar study, Landström In summary, background speech has been found to
et al. (2002) found that background speech, compared disrupt reading comprehension and proofreading accu-
with broadband noise (i.e., noise consisting of a wide racy in a number of experiments. In addition, the avail-
range of frequencies), did not affect proofreading per- able evidence suggests that this disruption is due to
formance for either contextual or noncontextual errors. processing of the semantic meaning of the speech
The auditory stimuli were presented at a sound inten- sound. These effects appear to be more reliable than
sity level comparable to that used in Venetjoki et al. the effect of nonspeech noise on reading, which has
(2006), although the speech consisted of random spo- not been consistently replicated. Nevertheless, several
ken statements. Finally, Smith-Jackson and Klein (2009) recent studies have found no effect of background
also found no effect of background speech (intermittent speech on reading comprehension, which casts doubt
or continuous) on overall proofreading accuracy. on its robustness and generalizability.
It is noteworthy that a few studies have also sug-
gested that the detrimental effect of background speech
Background music
on reading and proofreading can be diminished by
making the task harder and thus increasing participants’ Unlike noise and speech, which are usually a nuisance,
engagement with it (Halin, 2016; Halin, Marsh, Haga, playing music in the background is often done delib-
Holmgren, & Sörqvist, 2014; Halin, Marsh, Hellman, erately as a personal choice or a habit. Interest in the
Hellström, & Sörqvist, 2014). In a few experiments, potential effect of background music on reading started
Halin and his colleagues showed that performance on in the first half of the 20th century with the popularity
4 Vasilev et al.

of personal radios and record players and their use by effect when they listened to preferred music. In addi-
students. However, these early studies did not paint a tion, they did not replicate the previous finding that
clear picture of the relationship between background participants’ studying habits modulated the results.
music and reading. Although some of them found that Adding further to the confusion, Perham and Currie
music can negatively affect reading comprehension in (2014) found that preferred and nonpreferred lyrical
children and university students (Fendrick, 1937; music (i.e., music with sung lyrics) is equally disruptive
Fogelson, 1973; Henderson et al., 1945), others found to reading comprehension, although they did not report
that background music either does not affect reading data on students’ studying habits.
at all (Freeburne & Fleischer, 1952; L. R. Miller, 1947; The influence of background music on reading may
Mitchell, 1949) or that it actually improves reading per- also be modulated by the acoustic properties of the
formance (Hall, 1952). Indeed, this controversy has music. Some factors that have been considered are its
persisted until the present day; even the only two eye- informational load (Kiger, 1989), loudness, and tempo
tracking studies to address this question (Cauchard (W. F. Thompson, Schellenberg, & Letnic, 2012); its
et al., 2012; R. Johansson et al., 2012) have failed to familiarity to participants (Hilliard & Tolin, 1979); and
find any effect of background music on fixation dura- its capability to induce a startle response (Ravaja &
tions or fixation probabilities during reading. Kallinen, 2004). These results are quite interesting in
To examine what conditions might give rise to dis- terms of understanding what types of music might
traction, some studies have investigated whether the cause distraction, although they would benefit from
effect of background music on reading comprehension further replication and extensions. In summary, previ-
is modulated by personality traits (Avila, Furnham, & ous studies suggest that certain types of music may be
McClelland, 2011; Furnham & Allass, 1999; Furnham & distracting, but a negative effect of background music
Bradley, 1997; Furnham & Stephenson, 2007; Furnham on reading performance has not been consistently
& Strbac, 2002; Furnham, Trew, & Sneade, 1999; Kou, observed.
McClelland, & Furnham, 2017). These studies have pre- The available evidence suggests that experimental
dicted, on the basis of Eysenck’s (1967) theory of per- exposure to background noise, speech, and music may
sonality, that individuals high in extraversion will be disrupt reading performance. The effect of background
distracted less by background music than individuals noise and music appears to be less consistent: Many
high in introversion because of the extroverts’ higher studies report nonsignificant effects on reading com-
cortical arousal threshold. However, the results from prehension. Although the effect of background speech
these studies have been mixed. Although some of them on reading appears to be more reliable, several experi-
have found such an interaction between personality traits ments have also failed to find an effect in reading-
and background music (Daoussis & McKelvie, 1986; comprehension and proofreading tasks. Therefore,
Furnham & Bradley, 1997; Furnham & Strbac, 2002), considerable uncertainty exists with respect to the mag-
others have not (Avila et al., 2011; Furnham & Allass, nitude of these distraction effects and what aspects of
1999; Furnham & Stephenson, 2007; Furnham et al., background sounds may be responsible for them. One
1999; Kou et al., 2017). A number of factors may have possibility is that only certain acoustical or linguistic
led to these inconsistencies, such as the way in which properties of background sounds may account for the
participants were classified as introverts and extroverts distraction. We now turn to this possibility by examin-
or the small sample size in some of the studies. ing existing theories of auditory distraction.
Another factor that has been considered is the genre
of the music (Kallinen, 2002; L. K. Miller & Schyb, 1989;
Theories of Auditory Distraction
Mullikin & Henk, 1985; Tucker & Bushman, 1991).
However, as the popularity of music genres changes One of the earliest theoretical accounts of auditory
with time, it is arguably better to investigate what distraction is the phonological-interference hypothesis.
aspects of the music may cause distraction. One factor This account is based on Baddeley and Hitch’s (1974,
that may play a role is participants’ preference for the 1994) model of working memory, in which the phono-
music. For example, Etaugh and colleagues (Etaugh & logical loop acts as an acoustic store in which memories
Michals, 1975; Etaugh & Ptasnik, 1982) reported that are registered and rehearsed through a process of sub-
preferred music decreased reading-comprehension vocalization. Salamé and Baddeley (1982, 1987, 1989)
scores, but only for students who rarely study while reported a series of experiments in which they showed
listening to music. In contrast, R. Johansson et al. (2012) that memory for visually presented digits is impaired
found that participants had lower comprehension accu- by unattended speech but not by unattended acoustical
racy when listening to nonpreferred music compared noise. In addition, a distraction effect was observed even
with a quiet control condition, but there was no such if the speech sound was in a language that participants
Auditory Distraction During Reading 5

could not understand (Salamé & Baddeley, 1987). The words in the current sentence. In addition, some mod-
authors argued that this occurs because speech sounds els of reading comprehension (e.g., Kintsch, 1998)
automatically gain access to the phonological loop and assume that word meanings are combined to form
thus interfere with the encoding and rehearsal of visu- propositions or “idea units” according to their syntactic
ally presented items. Although this hypothesis is derived relationships (Kintsch & Rawson, 2005). Forming these
from a memory task, Salamé and Baddeley (1989) units must also involve establishing and keeping track
argued that a similar disruption may also be observed of the order of words in the sentence, as well as their
in more complex cognitive tasks such as reading. syntactic relationships.
Martin et al. (1988) were the first to systematically A final account that is relevant in a reading task is
test the phonological-interference hypothesis in a the duplex theory of auditory distraction (Hughes, 2014;
reading-comprehension task. In a series of experiments, Hughes, Vachon, & Jones, 2005, 2007; Sörqvist, 2010b),
they found that the disruptive effect of unattended according to which auditory distraction can occur from
speech was due to the semantic properties (i.e., mean- two different processes: interference by process and
ing) of the speech, rather than its phonological features. attentional capture (Hughes, 2014). Interference by
More specifically, the authors found that English speech process (Marsh, Hughes, & Jones, 2008, 2009; Marsh &
(which was intelligible to participants) was more dis- Jones, 2010) occurs when the background sound inter-
tracting than Russian speech (which was unintelligible feres with a process that is important for the main task.
to participants). Likewise, a continuous speech stream For example, in a reading task, the semantic processing
of random words was found to disrupt comprehension of meaningful speech would interfere with the task
more than a continuous speech stream of nonwords. because reading also requires semantic processing to
To account for these results, Martin et al. (1988) argued extract the meaning of the text. Alternatively, auditory
that, unlike serial-recall tasks, reading comprehension distraction can also be caused by attentional capture
requires understanding the meaning of the text. There- (Hughes et al., 2005; Vachon, Hughes, & Jones, 2012)
fore, the semantic properties of the irrelevant speech where attention is temporally directed away from the
can interfere with building the semantic representations main task. For example, the sound “B” in the sequence
of the text that is being read. This prediction will be “AAAAAABA” would capture attention because another
referred to as the semantic-interference hypothesis. “A” is expected in the sequence (Hughes, 2014; for a
The changing-state hypothesis (Hughes & Jones, review of similar effects caused by deviant sounds, see
2001; Jones & Macken, 1993; Jones, Madden, & Miles, Parmentier, 2014).
1992) is another prediction derived from serial-recall In a reading task, the interference-by-process part
tasks. According to this hypothesis, interference is of the duplex theory makes the same prediction as the
caused by background sounds that exhibit considerable semantic-interference hypothesis by Martin et al. (1988)
acoustic variation but not by steady-state, aperiodic discussed earlier. The difference between the two
sounds that do not have such variation ( Jones et al., accounts is very subtle: According to Marsh et al. (2008,
1992). For example, a sound consisting of different 2009), distraction occurs because processing the mean-
consonants (e.g., “B, F, P, S, N”) should cause more ing of the background speech depends on the same
interference than a sound made up of the same conso- process used for extracting the meaning of the text that
nant (e.g., “M, M, M, M, M”) because it exhibits more is being read. In contrast, Martin et al. (1988) assume
acoustic variation. The hypothesized mechanism that the semantic properties of the speech cause the
through which interference occurs is that changing- interference. These two very similar views are difficult
state sounds contain information about the serial order to disentangle empirically, and because they make the
of their constituent sound elements (Hughes & Jones, same prediction for the purposes of the present analy-
2001). This information can then interfere with main- sis, we will consider them together. The second part of
taining the serial order of items in a memory task. the duplex theory—attentional capture—is a very inter-
Although reading is a more complex cognitive task esting concept. However, because tasks such as reading
than serial recall of items, it also involves maintaining typically involve longer exposure to sounds, it is more
the order of words in the sentence and their syntactic difficult to study and will not be considered further in
relations. For example, because models of parallel word this analysis.
processing such as SWIFT (Saccade generation With
Inhibition by Foveal Targets; Engbert, Nuthmann,
Present Study
Richter, & Kliegl, 2005) assume that readers can process
multiple words at the same time, they also have to The review of the literature showed that background
assume, at least implicitly, that readers are somehow noise, speech, and music may be detrimental to reading
able to maintain information about the order of these performance but that considerable uncertainty exists as
6 Vasilev et al.

to the reliability and the magnitude of such distraction previous meta-analyses have done, this does not tell us
effects. This uncertainty makes it difficult to draw firm what makes these sounds distracting. As discussed pre-
conclusions about the experimental effects and their viously, a few theories make specific predictions about
real-world significance. Are background sounds reliably what type of auditory stimuli should be distracting.
disruptive to reading, and is this disruption large Therefore, the second aim of the study was to test the
enough to be of any practical significance? In addition, predictions of these theories using Bayesian metare-
after 80 years of research on the topic, what theoretical gression models (Welton, Sutton, & Cooper, 2012). As
conclusions can be made about the types of back- some of the theories outlined above were not originally
ground sounds that are disruptive to reading? developed in reading-comprehension tasks, it is impor-
The present study addressed these questions by per- tant to keep in mind that the present study is not a strict
forming a Bayesian random-effects meta-analysis of test of these theories. Rather, it aims to find out whether
studies investigating experimental exposure to noise, they can accommodate the existing evidence in reading
speech, or music in the background. Both studies with tasks and, if not, to pave the way for the development
adults and children were considered. Bayesian infer- of future theories.
ence is especially suited to answer these questions
because it enables us to directly quantify the uncer-
Predictions
tainty of the estimate of auditory-distraction effects,
given the available evidence. This in turn makes it pos- All of the predictions in the present analyses are sum-
sible to derive the probability, given the data, that back- marized in Figure 1. The phonological-interference
ground noise, speech, and music can distract readers hypothesis (Salamé & Baddeley, 1982) makes the
from their task. Bayesian meta-analytical models have unique prediction that all types of speech sounds
traditionally been used in biology and medicine (e.g., should be equally distracting because they all gain
Sutton & Abrams, 2001; Sutton et al., 2000) but more access to the phonological store. Therefore, both intel-
recently have also been introduced to psychology and ligible speech (i.e., in participants’ native language) and
linguistics ( Jäger, Engelmann, & Vasishth, 2017; Marsman unintelligible speech (i.e., in a foreign language) should
et al., 2017; Vasishth, 2015; Vasishth, Chen, Li, & Guo, be equally distracting. In addition, the phonological-
2013; see also Kruschke & Liddell, 2018). They have interference hypothesis is not capable of explaining
been successfully used to address contentious research distraction by nonspeech background noise and non-
questions, such as the processing of relative clauses in lyrical music because neither sound gains access to the
Chinese (Vasishth et al., 2013) and the extent to which phonological store.
readers can preprocess words in parafoveal vision The semantic-interference (Martin et al., 1988) and
(Vasilev & Angele, 2017). interference-by-process (Marsh et al., 2008) accounts
The two available (non-Bayesian) meta-analyses both make the prediction that only intelligible speech
have addressed how background noise and music affect that can be processed semantically by participants
a wide range of behavioral and cognitive tasks (Kämpfe, would cause distraction. Therefore, intelligible speech
Sedlmeier, & Renkewitz, 2010; Szalma & Hancock, should be more distracting than unintelligible speech.
2011). Although the results from these meta-analyses In addition, they also predict that (a) lyrical music
are quite interesting, their more general focus on all should be more distracting than nonlyrical music
types of cognitive tasks does not make it possible to because the former contains lyrics that are intelligible
make firm conclusions about reading in particular. to participants, and (b) intelligible speech should be
Kämpfe (2010) reported a separate analysis of reading- more distracting than lyrical music because, on average,
only studies and estimated the general effect of music continuous speech has more semantic content than
to be r = −0.11 (d = −0.22). However, this estimate was lyrical music.1 However, because lyrical music that is
based on only eight studies and thus does not include intelligible to participants contains not only semantic
most of the currently available data. Therefore, one of information but also phonological information, it is not
the contributions of the present meta-analysis was to possible to rule out any involvement of phonology in
estimate the general effect of background noise, speech, this effect.
and music on reading and to calculate the probability, Finally, the changing-state hypothesis ( Jones et al.,
given all the available evidence, that these auditory 1992) predicts that sounds exhibiting considerable
stimuli are detrimental to reading performance. acoustic variation should be more distracting than
The second and more important goal of the present steady-state sounds that do not exhibit such variation.
analysis was to investigate which aspects of background This leads to two further predictions. First, nonlyrical
sounds give rise to distraction. Although it can be infor- music should be more distracting than acoustical noise
mative to estimate the overall size of the effects, as (e.g., white or pink noise) because the former exhibits
Auditory Distraction During Reading 7

Predictions of Theories on Auditory Distraction

Semantic-Interference Hypothesis
Phonological-Interference Hypothesis Interference-by-Process Hypothesis

Intelligible Speech Unintelligible Speech Intelligible Speech Unintelligible Speech

Background Noise
More Distracting Equally Distracting Not Distracting

Nonlyrical Music
Semantic-Interference Hypothesis
Interference-by-Process Hypothesis
Changing-State Hypothesis Phonological-Interference Hypothesis

Nonlyrical Music Acoustical Noise Lyrical Music Nonlyrical Music

Environmental Noise Acoustical Noise Intelligible Speech Lyrical Music

Fig. 1. A schematic summary of the predictions derived from theories on auditory distraction.

more acoustic variation than the latter. Nonlyrical music the data are for different values of the parameter (Lynch,
is the strongest test of this prediction because it avoids 2007). The result of Bayesian inference is a posterior
any potential confounds from spoken language that probability distribution, which is the researcher’s
would be present in lyrical music. Second, more com- updated belief about the parameter given the observed
plex environmental noise (e.g., traffic noise or office data.
noise containing phones ringing, indistinct chatter) The posterior probability distribution is derived from
should again be more distracting than steady-state Bayes’s theorem, which states that the posterior distribu-
acoustical noise because it also exhibits more acoustic tion is proportional to the product of the prior probabil-
variation. ity distribution and the likelihood (i.e., posterior ∝ prior ×
likelihood; for more details, see Lynch, 2007). In the
meta-analysis, the observed means are the empirical
Method effect sizes (i.e., the differences between conditions)
The goal of a meta-analysis is to pool evidence from reported in the original studies. In contrast, the poste-
multiple studies to estimate some parameter of interest rior mean of the effect sizes is simply the mean of the
(e.g., the true difference in comprehension accuracy posterior probability distribution that is derived from
between reading in silence and reading with music in the Bayesian meta-analysis. Therefore, the posterior
the background). A Bayesian meta-analysis differs from mean reflects our updated belief about the size of the
the classical (frequentist) meta-analysis in the sense that effect (i.e., the difference) in light of the observed data.
it uses Bayesian inference to estimate the parameter One important part of any meta-analysis is to assess
and the uncertainty surrounding this estimate. Before the data for publication bias and other reporting biases.
performing the analysis, the researcher needs to express One common way to do this is to use what is known
his or her prior belief about the parameter in terms of as a funnel plot (Egger, Smith, Schneider, & Minder,
a probability distribution. This is known as the prior 1997; Sterne et al., 2011). This is a scatter plot of all the
probability distribution, and it reflects the researcher’s effect sizes included in the meta-analysis against some
belief about the parameter before observing the data. measure of their precision, such as the standard error
After the data are collected, a likelihood function is or the inverse of the standard error. More precise stud-
constructed, which essentially conveys how probable ies (i.e., the ones with smaller standard errors) will
8 Vasilev et al.

appear more narrowly at the top of the plot, whereas screened records, thus showing that the search strategy
less precise studies (i.e., the ones with larger standard was effective in locating them (unpublished studies
errors) will scatter more widely at the bottom. When typically make up 8%–10% of all sources in systematic
there is no bias or heterogeneity between studies, the reviews and meta-analyses; Clarke & Clarke, 2000;
scatter of the plot will resemble a symmetrical inverted Lefebvre et al., 2008). These unpublished studies came
funnel (Sterne et al., 2011). Funnel plot asymmetry can from different sources, such as dissertations, conference
occur if studies are missing from one side of the plot, proceedings, reports, and unpublished manuscripts.
thus creating an asymmetrical funnel shape. For exam- The identified articles were evaluated against the
ple, this can happen if publication bias or other reporting inclusion criteria presented in Appendix A. In short,
biases are preventing the dissemination of studies with the studies had to (a) experimentally manipulate back-
negative findings (however, reporting biases are not the ground noise, speech, or music in a reading or a proof-
only possible source of asymmetry, and other factors reading task; (b) have a sound methodological design;
need to be explored as well; see Sterne et al., 2011). and (c) include reading in silence as a baseline condi-
tion. The inclusion criteria were developed before the
meta-analysis with the help of a smaller, qualitative
Literature search
review of the literature. Epidemiological studies of
The search of the literature was conducted by following extended exposure to traffic noise in children were not
the Preferred Reporting Items for Systematic Reviews included because they answer a qualitatively different
and Meta-Analyses (PRISMA) guidelines (Moher, question and are often confounded by other variables,
Liberati, Tetzlaff, Altman, & The PRISMA Group, 2009). such as social deprivation (Haines, Stansfeld, Head, &
A flowchart of the process is presented in Figure 2. Job, 2002). Overall, of the experiments for which eligi-
Google Scholar, Scopus, the Web of Science, and bility was assessed, 44% were included in the meta-
ProQuest Dissertations were searched with the follow- analysis. Although the inclusion rate may appear to be
ing keywords: “background noise AND reading,” “back- low, it was necessary to ensure that only studies that
ground speech AND reading,” and “background music were similar enough to be analyzed together were
AND reading.” Each search for one of the three back- included. Information about the included studies and
ground sounds was done separately. The literature their effect sizes is presented in Appendix B.
search covered articles published before June 25, 2017.
In addition, we examined the reference lists of all
screened articles as well as those of previous literature
Dependent measures
reviews and meta-analyses on similar topics (Beaman, The main dependent variable was reading-comprehen-
2005; Clark & Sörqvist, 2012; Dalton & Behm, 2007; sion accuracy, which was available for 54 of the studies
Kämpfe et al., 2010; Klatte, Bergström, & Lachmann, (83.1%). Therefore, most of the reported analyses are
2013; Shield & Dockrell, 2003; Szalma & Hancock, based on reading-comprehension accuracy. Moreover,
2011). effect sizes for reading speed were available for 13
When searching the literature, it is important to con- studies (20%), and these were analyzed separately.
sider relevant studies that have been conducted but Finally, experiments reporting proofreading accuracy
never published in a peer-reviewed journal or an edited (n = 7; 10.7%) were also analyzed for completeness,
book (i.e., the so-called file-drawer problem; Rosenthal, but this was again done separately from the analysis
1979). This issue was addressed through some of the on reading-comprehension accuracy.
databases that were searched. ProQuest Dissertations For the metaregression analyses, additional informa-
contains more than 2 million doctoral and masters’ dis- tion about the type of sound manipulation was also
sertations (Lefebvre, Manheimer, & Glanville, 2008), extracted (e.g., whether the noise was environmental or
which often contain unpublished research. In addition, acoustical, or whether the music was lyrical or nonlyri-
Google Scholar indexes a wide range of unpublished cal). If a study contained a background-music manipula-
sources, such as conference proceedings, dissertations, tion, M. R. Vasilev listened to the songs to determine
reports, and preprints. Furthermore, author searches whether they were lyrical or nonlyrical. Only studies that
were carried out for researchers who have done work could be unambiguously classified as either lyrical or
on this topic in the past 2 decades. These searches nonlyrical were added to this metaregression analysis.
included researcher networking web sites, such as
ResearchGate.net and Academia.edu, that also contain
Effect-size calculation
unpublished research (e.g., conference presentations
or unpublished manuscripts). In the present meta- Standardized effect sizes of the mean difference (g) and
analysis, unpublished studies accounted for 17% of all their variances were calculated from the reported
Auditory Distraction During Reading 9

423 Records Identified Through 118 Records Identified Through

Database Searching Other Sources

Identification 371 Records After Duplicates Were Removed

371 Records Screened

Screening

242 Records Excluded

• Not Reading (108)
• Not Relevant to Analysis (49)
• No Sound Manipulation (27)
• Not an Experiment (22)
• Other (36)

129 Full-Text Articles Assessed for Eligibility

(148 Experiments)

83 Experiments Excluded
Eligibility

• Chronic Exposure to Noise (22)

• No Silence Baseline (17)
• Methodological Issues (13)
• Intervention Study (6)
• Task Not Relevant to Analysis (9)
• Other (16)

65 Experiments Included in Meta-Analysis

Included

Fig. 2. A flowchart illustrating the stages of the literature-search process.

descriptive statistics. This was done by first calculating from the reported test statistics using existing formulas
Cohen’s d for the respective design of the study and (Borenstein, 2009; Lajeunesse, 2013). 2 In the analysis
then applying Hedges’s g correction for small sample of reading-comprehension accuracy and proofreading
bias (Hedges & Olkin, 1985). The effect sizes were accuracy, studies were coded so that negative effect
calculated with formulas 12.11 to 12.22 from Borenstein sizes indicate lower comprehension/proofreading
(2009). In all effect sizes, silence was the control condi- accuracy in an experiment’s sound condition. Like-
tion. Therefore, the effects represent the standardized wise, in the analysis of reading speed, negative effect
mean difference between reading in an experimental sizes also indicate slower reading speed in the experi-
sound condition and reading in silence in a control mental sound condition compared with the silent con-
condition. If descriptive statistics were unavailable or trol condition. One effect size was excluded as an
incomplete, the effect sizes were calculated by digitiz- outlier (see Fig. S1 in the Supplemental Material avail-
ing graphs (Rohatgi, 2015) or converted/approximated able online).
10 Vasilev et al.

Because 55.5% of the studies used a within-subjects comprehension, t(52) = −0.42, p = .67, or reading speed,
design, it was necessary to estimate the population t(11) = 0.08, p = .93. Proofreading accuracy was not
correlation (ρ) between the control and experimental considered here because funnel-plot tests of asymmetry
conditions (Borenstein, 2009; Szalma & Hancock, 2011). are not recommended when there are fewer than 10
Eight statistically independent estimates were obtained studies; Sterne et al., 2011). In addition, metaregression
from experiments for which the raw data were avail- analyses (Figs. 3e and 3f ) indicated that the size of
able, as well as one estimate from a study (L. R. Miller, auditory-distraction effects was not predicted by the
1947) that reported the required statistics. These rep- impact factor of the journal or the year of publication.
resented a wide range of experimental sound types and In summary, there was no evidence to suggest that
included measures of both reading comprehension and publication bias may have influenced the conclusions
reading speed. We followed Szalma and Hancock’s from the meta-analysis.
(2011) approach to meta-analyze the obtained correla-
tions and to obtain a weighted estimate of ρ. The result-
Data analysis
ing weighted value of 0.74 was used to calculate the
effect sizes for all studies that had a within-subjects Meta-analysis. The common choice in meta-analysis is
design. between a fixed-effects model and a random-effects
Effect sizes from within- and between-subjects stud- model. A fixed-effects model assumes that all effect sizes
ies are calculated with different standard deviation met- that are combined are estimating the same true underly-
rics and are thus not necessarily comparable (Morris & ing effect, which we will call θ. Therefore, the effect size
DeShon, 2002). Consistent with previous work (Kämpfe of the ith study, Ti, is assumed to come from a normal
et al., 2010; Szalma & Hancock, 2011), the effect sizes distribution with some mean θ and variance σi2:
from within-subjects studies were transformed to make
them comparable with the effect sizes of between-
subjects studies. This was done using Formula 11 from
( )
Ti ∼ Normal θ, σi2 i = 1, 2, 3,…, n. (1)

Morris and DeShon (2002). In addition, because some In this model, any variability in the estimate is due to
studies yielded more than one effect size, care was sampling error alone. On the other hand, a random-
taken to avoid statistical nonindependence in the analy- effects model relaxes this assumption by explicitly
ses (for a recent overview, see Noble, Lagisz, O’dea, & allowing for variability in the true effect size between
Nakagawa, 2017). If a study contributed multiple effect studies (Welton et al., 2012). In this case, Ti, the
sizes per analysis, these were averaged together to observed effect size of the ith study, is assumed to be
include only one effect size for that study (Lipsey & generated by a unique underlying true effect for that
Wilson, 2001).3 ith study, denoted here as θi. This unique underlying
effect θi is in turn assumed to come from a normal
Publication bias distribution with some (unknown) mean θ and a
between-studies variance of τ2:
In the present meta-analysis, 12.3% of all included stud-
ies were from the so-called gray literature (i.e., they
were not formally published in a peer-reviewed journal ( )
Ti ∼ Normal θi , σi2 i = 1, 2, 3,…, n, (2)
or in an edited book at the time of analysis). To assess
the data for publication bias and other related biases, we (
θi ∼ Normal θ, τ2 . )
performed a number of visual and statistical tests using
the meta (Schwarzer, 2007) and metafor (Viechtbauer, Therefore, the true effect sizes of individual studies in
2010) packages for the R software environment (R Core a random-effects meta-analysis can be informally
Team, 2016). The visualization of the results for reading thought of as random samples from a normal distribu-
comprehension is presented in Figure 3 (for the reading- tion of effect sizes (Welton et al., 2012).
speed results, see the Supplemental Material). The fun- In the present meta-analysis, a random-effects model
nel plots (Figs. 3a and 3b) indicate that there was some was chosen a priori because some between-studies het-
heterogeneity in the data, but there was no clear evi- erogeneity was expected as a result of differences in
dence of asymmetry that could indicate publication design, sound-intensity levels, participants, reading
bias. This was confirmed by a funnel-plot test of asym- materials, and so forth. A random-effects model can
metry performed on the basis of a weighted linear naturally account for such sources of variability between
regression of the effect estimates on their standard errors studies and is often the model of choice in studies on
(Sterne et al., 2011), which revealed no statistically sig- language processing (e.g., Jäger et al., 2017; Vasilev &
nificant evidence of asymmetry for either reading Angele, 2017; Vasishth et al., 2013). The full Bayesian
Auditory Distraction During Reading 11

a b
0.00
24.65

0.11

Precision (1/SE )
19.04
Standard Error

0.23
13.42

0.34
7.81

0.45
2.20
−2.00 −1.00 0.00 1.00 −2.00 −1.00 0.00 1.00
Effect Size (g ) Effect Size (g )
c d
300 Between-Subjects Design
Effect Size (g)/SE

Within-Subjects Design
0.50
2

Sample Size
0.00 200
0
−2
−0.50
−1.00 100
−1.50
0
0 1 2 3 4 5 −2.0 −1.0 0 1.0
Precision (1/SE ) Effect Size (g )

e f
1.0 1.0
Effect Size (g)

Effect Size (g)

0.0 0.0

−1.0 −1.0

−2.0 −2.0
1935 1955 1975 1995 2015 0.5 1.5 2.5 3.5 4.5 5.5
Publication Year Journal Impact Factor
Fig. 3. Visual assessment of publication bias and other related biases in the literature on reading-comprehension accuracy
(presentation format adapted from Nakagawa, Noble, Senior, & Lagisz, 2017, Fig. 6). The funnel plots show (a) standard error
and (b) precision (i.e., the inverse of the standard error) as a function of effect size. The white area within the gray bounds
shows the 95% pseudo-confidence interval; the gray bands extend this area to the 99% pseudo-confidence interval. (See the
main text for information on how to interpret funnel plots.) The vertical lines indicate the pooled effect size, as estimated
from a random-effects meta-analysis. The radial or Galbraith plot (c) shows the z statistic (i.e., the effect size divided by its
standard error) of each study as a function of precision. The arc on the right side of the plot corresponds to the size of the
individual observed effects. The interval next to the arc shows the pooled effect size and its 95% confidence interval. The gray
area highlights the region in which z values between –2 and 2 lie and is the same as the approximate 95% confidence interval;
on average, 95% of the studies are expected to fall within this range (Anzures-Cabrera & Higgins, 2010). The vertical scatter of
effect sizes shows the degree of heterogeneity in the data. The scatterplot in (d) shows the relationship between effect sizes
and sample sizes, broken down by study design type (i.e., between subjects vs. within subjects). The scatterplots (with best-
fitting regression lines) in the bottom row show the results of metaregression models examining the relationship (e) between
effect size and publication year and (f) between effect size and impact factor of the journal where the study was published.
12 Vasilev et al.

model was defined as follows ( Jäger et al., 2017; Schmid difference that the estimate is adjusted by the precision of
& Mengersen, 2013): the studies (i.e., the inverse of the within-studies variance
of the sampling distribution; Welton et al., 2012). The
( )
Ti |θi , si2 ~ Normal θi , si2 i = 1, 2, 3,…, n (3) model from Equation 3 was extended by adding a regres-
sion coefficient, β, for the underlying effect of the covari-
(
θi |θ, τ2 ∼ Normal θ, τ2 ) ate (the boldface type indicates the added parameters;
Jäger et al., 2017; Welton et al., 2012):
θ ∼ Uniform ( −10, 10 )
( )
Ti |θi , β, si2 ∼ Normal θi + β x i , si2 i = 1, 2, 3,…, n (4)
τ ∼ Uniform ( 0,10 ) .
(
θi |θ, τ2 ∼ Normal θ, τ2 )
Ti is the observed effect size (in Hedges’s g) in the ith
study; θi is the true auditory-distraction effect in the ith
study; Si2 is the true sampling variance of the ith study, θ ∼ Uniform ( −10, 10 )
estimated from the within-studies variance of the sam-
pling distribution of study i; θ is the unknown true τ ∼ Uniform ( 0, 10 )
auditory-distraction effect estimated by the model; and
τ2 is the unknown between-studies variance. In this
model, precision was defined as the inverse of the β ∼ Uniform ( −10,10 ) .
within-studies variance of the sampling distribution.
The last two lines in Equation 3 indicate the prior prob- β is the regression coefficient for the underlying effect
ability distributions used for θ and τ. In the present of the covariate xi; θi is the true auditory-distraction
analysis, we used uniform priors that assign equal prob- effect in the ith study, adjusted for the covariate effect
ability to any value on these intervals. Because these x i; and θ is the unknown true auditory-distraction
are vague priors, they have very little to no influence effect, also adjusted for the covariate effect xi. All
on the results. This was confirmed by doing a sensitivity remaining parameters have the same interpretation as
analysis of the main results with alternative priors: in Equation 3. The contrasts used for the covariate x i
Normal (0,104) for θ and Normal (0,104) I(0,) for τ (the are presented in Table 1. These contrasts were used to
normal distribution was truncated at 0). The sensitivity test the predictions outlined in the introduction.
analysis indicated that the choice of priors did not influ-
ence the results (see the Supplemental Material). Posterior sampling. The posterior probability distribu-
tion was sampled with JAGS (Plummer, 2003) using the R
Metaregression. Although random-effects meta-analysis software environment (R Core Team, 2016). Five Markov-
can account for heterogeneity between studies, it does chain Monte Carlo (MCMC) chains were run with 100,000
not tell us what causes this heterogeneity in the first iterations each. Checks were made to ensure that the
place (Welton et al., 2012). However, it is possible to use starting values of the MCMC chains did not influence the
metaregression models to investigate how different study results. The first 3,000 iterations were considered a burn-
characteristics (e.g., whether the background music was in period and were discarded. A thinning interval of 5
lyrical or nonlyrical) are associated with the observed was used for the MCMC chains (i.e., every fifth sample
effect sizes. Metaregression models are similar to the was retained) to reduce the influence of autocorrelation.
ordinary least-squares regression, but with the crucial The summary of the posterior distribution was based on

Table 1. Type of Metaregression Comparisons and the Contrast Coding of Covariates

Covariate levels Contrast coding

Comparison Level 1 Level 2 Level 1 Level 2

Nonlyrical vs. lyrical music Nonlyrical Lyrical –1 1
Lyrical music vs. intelligible speech Music Speech –1 1
Unintelligible vs. intelligible speech Unintelligible Intelligible –1 1
Acoustical vs. environmental noise Acoustical Environmental –1 1
Acoustical noise vs. instrumental music Noise Music –1 1
Child vs. adult participants Child Adult –1 1
Auditory Distraction During Reading 13

20,000 samples per chain (excluding the burn-in period). value for the effect (note that this does not allow us to
Convergence was assessed with visual inspection of the conclude that there is no true effect, only that it is pos-
trace plots and with Gelman and Rubin’s (1992) conver- sible that the true effect size is 0). However, the prob-
gence diagnostic. The diagnostics suggested that conver- ability that these effects are negative was very high in
gence had been achieved in all models. all analyses (more than 90%). This means that although
The effective sample size (ESS) of the MCMC chains the size of the effects was small, there is a very high
was calculated for every parameter and contrast of probability that background speech, noise, and music
interest. The ESS is the size of the MCMC chain after are detrimental to reading comprehension, reading
adjusting it for autocorrelation (Kass, Carlin, Gelman, speed, and proofreading accuracy.
& Neal, 1998; Kruschke, 2015). All of the present analy- Although it is possible to use Bayes factors to per-
ses had an ESS greater than 10,000, as recommended form hypothesis testing (e.g., see Rouder & Morey,
by Kruschke (2015). This was necessary for achieving 2011; Rouder, Morey, & Province, 2013), the emphasis
a stable estimation of the credible interval limits: This in the present meta-analysis was on estimating the mag-
estimation depends on sparse regions of the posterior nitude of auditory-distraction effects. The findings from
probability distribution that are sampled less often by this meta-analysis suggest that nonnull effects almost
the MCMC chain (Kruschke, 2015). certainly exist, even if their magnitude is small. There-
The results are presented as the estimate of the effect fore, even if a Bayes factor were to favor a null hypoth-
sizes of interest and their corresponding 95% credible esis relative to some alternative hypothesis, the prior
intervals. Unlike classical confidence intervals, credible probability that the null hypothesis was exactly true
intervals have the intuitive interpretation that they con- would be negligible in this case. Because of this, the
tain the true auditory-distraction effect with 95% prob- posterior probability of the null hypothesis would
ability because the values within this interval make up remain small.
95% of the posterior probability distribution (see Morey, Because our analyses included both studies with
Hoekstra, Rouder, Lee, & Wagenmakers, 2016). All adults as participants and studies with children as par-
probabilities reported in the article are the posterior ticipants, we carried out metaregression models to test
probability, given the data, that auditory-distraction whether the effect sizes differed between adults and
effects exist. A more detailed summary of Bayesian children. Only reading comprehension was considered
methods and their interpretation is beyond the scope in these analyses because there were too few child
of this article. However, Nicenboim and Vasishth (2016) studies to reliably estimate differences in reading speed,
provide an accessible overview. and all proofreading studies were done with adults. The
results are presented in Table 3. They show the esti-
mated mean difference between studies with children
Results compared with studies with adults, after adjusting for
their precision in the analysis. Overall, the difference
Meta-analysis between adults and children was very close to 0; thus,
The results from the meta-analysis are presented in background sounds were equally detrimental to reading
Table 2. In addition, forest plots are presented in Figure comprehension for both children and adults. One
4 for the main measure of comprehension accuracy. To exception was that background noise impaired reading
interpret the magnitude of the effects, we will consider comprehension in children slightly more than it did in
Cohen’s (1988) guidelines of 0.20 for small effects, 0.50 adults, but the mean difference was still quite small
for medium effects, and 0.80 for large effects. Overall, (Hedges’s g = 0.05). In addition, the effect was not
there was a small negative effect for reading compre- highly reliable because there was only a 73% probabil-
hension (Hedges’s g = −0.21), which indicates that ity of a true mean difference. Taken together, these
background sounds generally impaired comprehension results suggest that effect sizes for reading comprehen-
accuracy. Consistent with the review of the literature, sion did not generally differ between adults and chil-
background speech had a stronger negative impact on dren. For this reason, child and adult studies were
reading comprehension (Hedges’s g = −0.26) than either analyzed together in all remaining analyses.
background noise (Hedges’s g = −0.17) or background
music (Hedges’s g = −0.19). Nevertheless, the effect size
Metaregression
for each of the three sound types was fairly small.
Reading speed and proofreading accuracy were also The results from the metaregression models testing the
impaired by background sounds. However, the effect theoretical predictions outlined in the introduction are
sizes for these two measures were very small, and the presented in Figures 5 and 6. Recall that the models
95% credible intervals all included 0 as a plausible yield a regression slope that shows the estimated mean
14 Vasilev et al.

Table 2. Posterior Effect Size Estimates of Auditory-Distraction Effects and 95%

Credible Intervals From the Meta-Analysis

Mean ES p(ES <

Type of analysis n (Hedges’s g) 95% CrI 0 | Data) τ2 ESS
Reading comprehension
All sounds 54 –0.21 [–0.30, –0.13] > .99 0.06 91803
Noise 12 –0.17 [–0.33, 0.002] .97 0.06 92499
Speech 20 –0.26 [–0.36, –0.17] > .99 0.02 47662
Music 36 –0.19 [–0.34, –0.05] > .99 0.13 93678
Reading speed
All sounds 13 –0.06 [–0.15, 0.02] .92 0.01 20915
Speech 6 –0.08 [–0.20, 0.03] .92 0.01 28612
Proofreading accuracy
Speech and Noise 7 –0.14 [–0.42, 0.04] .94 0.04 40097
Speecha 6 –0.09 [–0.30, 0.07] .90 0.02 41296

Note: n = number of studies in the analysis; ES = effect size; p(ES < 0|Data) = probability that
background sounds are detrimental to reading, given the data (i.e., probability that the effect size is
smaller than 0); CrI = credible interval; τ2 = estimated between-studies variance; ESS = effective sample
size of the Markov-chain Monte Carlo chains for the main parameter of interest (θ).
a
Intelligible speech only.

difference between the two groups after adjusting for 6b shows, the size of both effects, as estimated by a
the precision of individual studies. Consistent with the random-effects meta-analysis, was very close to 0. This
semantic-interference hypothesis but not with the result is contrary to the predicted difference from the
phonological-interference hypothesis, there was a 99% changing-state hypothesis.
probability that intelligible speech was more distracting
than unintelligible speech (mean difference: Hedges’s
g = −0.12). In addition, in line with both the semantic-
Discussion
and phonological-interference hypotheses, there was a In the present study, we investigated the magnitude of
95% probability that lyrical music was more distracting auditory-distraction effects during reading and the com-
than nonlyrical music (mean difference: Hedges’s g = patibility of these effects with existing theories of dis-
−0.19). Note, however, that there was no difference traction. We will first consider the overall size of the
between intelligible speech and lyrical music, and the effects and then discuss their theoretical and practical
estimated probability of a true difference was only 54% implications. The main findings from the meta-analysis
(50% = no difference, because the posterior probability can be summarized as follows. First, background
density would lie evenly to the left and right side of 0). speech, noise, and music all had a negative effect (indi-
This last result is surprising because, arguably, most cating distraction) on reading-comprehension accuracy.
people perceive lyrical music to be subjectively less The magnitude of the effects was small, but highly
distracting than intelligible speech. For example, it can reliable, meaning that there was a very high probability
be speculated that students may be more likely to that these sounds are detrimental to reading compre-
choose to study while listening to lyrical music in the hension given the available evidence. Second, auditory-
background than they are to study while listening to distraction effects measured with reading comprehension
an audio book. However, the present results suggest did not generally differ between adults and children.
that lyrical music and intelligible speech are equally Finally, background speech, noise, and music had a
distracting. very small negative effect on reading speed, and back-
Consistent with the changing-state hypothesis, there ground speech and noise also had a small negative
was a 90% probability that environmental noise was effect on proofreading accuracy. Although both effects
more distracting than acoustical noise (mean difference: proved to be smaller than the ones observed in reading
Hedges’s g = −0.10). However, there was only a 55% comprehension, there was still a high probability that
probability of a difference between nonlyrical music they were negative (> 90%).
and acoustical noise, which suggests that the two back- The present results provide the first comprehensive
ground sound types did not generally differ. As Figure analysis of auditory-distraction effects in a reading task.
Auditory Distraction During Reading 15

a b
Music Speech
Ljung et al. (2009)
Vasilev et al. (2017)
Falcon (2017), Sample 2
Hyönä & Ekholm (2016), Exp.1
Dove (2009)
Armstrong & Chung (2000)
Kou et al. (2017)
Halin et al. (2014)
Cauchard et al. (2012)
Cauchard et al. (2012)
Doyle & Furnham (2012) Martin et al. (1988), Exp.1
Gillis (2016) Cool et al. (1994), Exp.2
Mullikin & Henk (1985) Martin et al. (1988), Exp.4
Tucker & Bushman (1991) Sörqvist et al. (2010)
Mitchell (1949) Martin et al. (1988), Exp.5
Martin et al. (1988), Exp.2 Sörqvist et al. (2010), Exp.2
Furnham & Allass (1999) Pool et al. (2000), Exp.1
Miller & Schyb (1989) Furnham et al. (1994)
Freeburne & Fleischer (1952) Sörqvist et al. (2010), Exp.1
Martin et al. (1988), Exp.1 Sörqvist (2010), Exp.1b
Etaugh & Michals (1975) Halin (2016)
Furnham at al. (1999) Armstrong et al. (1991)
Pool et al. (2000), Exp.2
Madsen (1987), Exp.1
Baker & Madell (1965)
Pool et al. (2000), Exp.2
Pooled Mean
Cool et al. (1994), Exp.2
Pool et al. (2000), Exp.1 −2.0 −1.5 −1.0 −0.5 0.0 0.5
Falcon (2017), Sample 1 Effect Size (g )
Anderson & Fuller (2010)
Miller (2014)
c Noise
Ahuja (2016) Johansson (1983)
Fogelson (1973) Dockrell & Shield (2006)
Fendrick (1937) Vasilev et al. (2017)
Perham & Currie (2014) Martin et al. (1988), Exp.1
Johansson et al. (2012)
Martin et al. (1988), Exp.4
Daoussis & McKelvie (1986) Observed Mean
Sörqvist (2010), Exp.1a
Henderson et al. (1945) (95% CI)
Kou et al. (2017)
Kiger (1989)
Etaugh & Ptasnik (1982) Halin (2016)
Posterior Mean
Kelly (1994) (95% CrI) Ljung et al. (2009)
Furnham & Strbac (2002) Martin et al. (1988), Exp.5
Furnham & Bradley (1997) Johansson et al. (2012)
Avila et al. (2011) Furnham & Strbac (2002)
Pooled Mean Pooled Mean

−2.0 −1.5 −1.0 −0.5 0.0 0.5 −2.0 −1.5 −1.0 −0.5 0.0 0.5
Effect Size (g ) Effect Size (g )

Fig. 4. Forest plot for the main effect of background music (a), speech (b), and noise (c) on reading comprehension. Plotted are the observed
(i.e., empirical) effect sizes with their 95% confidence intervals and the posterior effect-size estimates from the meta-analysis model with their
corresponding 95% credible intervals. Each square’s size is proportional to the weight of the study it represents (i.e., to the inverse of the
within-studies variance of the sampling distribution). The red diamond (with 95% credible intervals) at the bottom of each panel indicates
the pooled estimate from the meta-analysis.
16 Vasilev et al.

Table 3. Mean Difference in the Effect Size Between Child and Adult Studies: Metaregression
Results

Number of studies Mean

difference p(ESCH >
Analysis Children Adults (Hedges’s g) 95% CrI ESA | Data) ESS
Reading comprehension
All sounds 18 36 –0.01 [–0.10, 0.08] .43 30623
Noise 5 7 0.05 [–0.13, 0.22] .73 29974
Speech 5 15 0.00 [–0.12, 0.12] .51 30263
Music 13 23 0.02 [–0.12, 0.17] .64 18498

Note: Mean difference = Posterior estimate of the mean difference (in Hedges’s g) between adult and child
participants; CrI = credible interval; p(ESCH > ESA|Data) = probability that the effect size for child participants is
bigger than the effect size for adult participants, given the data; ESS = effective sample size of the Markov-chain
Monte Carlo chains for the main parameter of interest (β).

As the review of the literature showed, interest in this shows, statistical power is related to sample size; gener-
topic has a very long history that precedes the cognitive ally speaking, a larger number of participants are
revolution and, indeed, most of the work on auditory required to achieve sufficient statistical power to detect
distraction in other cognitive tasks. Traditionally, much some of the auditory-distraction effects observed in the
of the interest in auditory distraction in reading tasks present study. This suggests that although most of the
has been related to its practical implications for reading observed effects are negative in sign, statistical signifi-
outside the psychological laboratory, such as studying cance may not always be achieved if the underlying
for an exam, reading in the classroom, or any kind of effect is small and the experiment is underpowered.
work that involves reading in a busy office. However,
the inconclusive and sometimes contradictory evidence
has made it difficult to arrive at clear conclusions until
Implications for theories of auditory
now. The present results advance our understanding of distraction
this topic by showing that external auditory input The second goal of the present study was to investigate
almost always comes at a cost for reading efficiency. what properties of background sounds make them dis-
Even though the observed cost was modest, especially tracting and to test what theoretical frameworks can
for measures such as reading speed and proofreading explain the results. This is an important question, given
accuracy, there was still relatively high probability that that not all studies have explicitly considered the theo-
it reflects a true effect in the population. Therefore, the retical implications of their work; some researchers
present study resolves some of the controversy high- have taken a more applied approach of simply testing
lighted in the introduction by showing that although whether certain types of sounds are distracting to read-
general auditory-distraction effects by background ers. More broadly, the present analyses provide a
noise, speech, and music almost certainly exist, their glimpse into how well readers can maintain focus on
magnitude is small. the main task (reading) while listening to a competing
Given that there is a very high probability that back- stream of auditory input that they try to ignore. The
ground speech, noise, and music are detrimental to metaregression results provided a few key insights into
reading comprehension, why have some of the previous the nature of auditory-distraction effects, as measured
findings been so inconsistent? One possibility is that with reading-comprehension accuracy.
some of the original studies may not have had sufficient First, lyrical music and intelligible speech were found
statistical power to detect the underlying effects. Figure to be equally distracting, and lyrical music was found
7 shows the relationship between sample size and sta- to be more distracting than nonlyrical music. Second,
tistical power for a range of effect sizes, including the intelligible speech was in turn more distracting than
ones observed in the present meta-analysis (see unintelligible speech. Finally, environmental noise was
Wallisch, 2015). This is for illustrative purposes only, more distracting than acoustical noise, but there was
given that statistical power is influenced not only by no reliable difference between nonlyrical music and
sample size and the magnitude of the true effect but acoustical noise. These results provide strong support
also by such other factors as the reliability of the mea- for the notion that the presence of language in back-
sure, missing data, sampling control, and so on (Hansen ground sounds is the strongest contributor to auditory
& Collins, 1994). Nevertheless, as Figure 7 clearly distraction. Indeed, the two largest distraction effects
Auditory Distraction During Reading 17

a
Nonlyrical Music vs. Lyrical Music vs. Unintelligible Speech vs.
Lyrical Music Intelligible Speech Intelligible Speech
2 2 2

1 1 1
Effect Size (g)

β = –0.19 β = –0.12 β = –0.12

0 0 0

−1 −1 −1

−2 −2 −2
Nonlyrical Lyrical Lyrical Music Intelligible Speech Unintelligible Intelligible
Type of Music Type of Sound Type of Speech

b
1.0 1.0 1.0

0.5 0.5 0.5

Effect Size (g)

0.0 0.0 0.0

−0.5 −0.5 −0.5

−1.0 −1.0 −1.0

Nonlyrical Lyrical Lyrical Music Intelligible Speech Unintelligible Intelligible
Type of Music Type of Sound Type of Speech
Fig. 5. Results of the metaregression models testing the predictions of the semantic-interference and phonological-interference hypoth-
eses. The graphs in (a) show the regression slopes and the observed effect sizes of the studies included in the analysis. The slope indi-
cates the mean difference between the two groups of studies in each graph, as estimated by the metaregression model. Each circle’s size
is proportional to the weight of the study it represents (i.e., proportional to the inverse of the within-studies variance of the sampling
distribution). The effective sample sizes of the Markov-chain Monte Carlo (MCMC) chains for β are 11,455, 24,381, and 54,689 (from left
to right). The graphs in (b) show the estimated posterior effect size for each group of studies from a random-effects meta-analysis of
the simple effect. Bars indicate 95% credible intervals. The effective sample sizes of the MCMC chains for θ are 98,478, 95,721, 97,382,
32,748, 15,048, and 34,152 (from left to right).

were found for lyrical music (Hedges’s g = −0.35) and the actual process of trying to extract their meaning
intelligible speech (Hedges’s g = −0.34). This last find- can distract readers from their main task. Nevertheless,
ing is consistent with both the semantic-interference these two accounts cannot explain distraction by non-
account (Martin et al., 1988) and the interference-by- speech background noise.
process account (Marsh et al., 2008), which predict that The present findings are generally not consistent
either the semantic content of spoken or sung lyrics or with the phonological-interference account for two
18 Vasilev et al.

a
Acoustical vs. Acoustical Noise vs.
Environmental Noise Nonlyrical Music
2 2

1 1

Effect Size (g)

β = –0.1 β = –0.02
0 0

−1 −1

−2 −2
Acoustical Environmental Noise Music

b 1.0 1.0

0.5 0.5
Effect Size (g)

0.0 0.0

−0.5 −0.5

−1.0 −1.0
Acoustical Environmental Noise Music
Type of Noise Type of Sound
Fig. 6. Results of the metaregression models testing the predictions of the changing-
state hypothesis. The graphs in (a) show the regression slope and the observed effect
size of the studies included in the analysis. The slope indicates the mean difference
between the two groups in each graph, as estimated by the metaregression model.
Each circle’s size is proportional to the weight of the study it represents (i.e., to the
inverse of the within-studies variance of the sampling distribution). The effective
sample sizes of the Markov-chain Monte Carlo (MCMC) chains for β are 36,063 (left)
and 13,062 (right). The graphs in (b) show the posterior effect size for each group,
as estimated by a random-effects meta-analysis of the simple effect. Bars indicate
the 95% credible intervals. The effective sample sizes of the MCMC chains for θ are
31,904, 89,200, 31,904, and 98,478 (from left to right).

reasons. First, it predicts that all speech sounds should acoustic variation than acoustical noise (e.g., white or
be equally distracting because they all would gain pink noise).
access to the phonological store; however, intelligible What type of theoretical framework could account
speech was reliably more distracting than unintelligible for the present results? Clearly, none of the theories
speech. In addition, background noise, which would considered so far can account for all the findings.
not gain access to the phonological store, was also Although some theories were successful in accounting
found to cause distraction. Finally, the results are only for some of the effects, the present results suggest that
partially consistent with the changing-state account new theoretical models are needed that can explain all
( Jones et al., 1992), which predicts that sounds with the evidence. This is not necessarily a limitation of
greater acoustic variation would cause greater distrac- existing theoretical accounts because, as noted previ-
tion. This is because environmental noise was more ously, not all of them were originally designed to
distracting than acoustical noise (which is consistent account for distraction effects in a reading task. In
with the theory), but nonlyrical music was not (which addition, these theories suggest very useful mecha-
is not consistent with the theory). In both cases, envi- nisms through which auditory distraction can occur. In
ronmental noise and nonlyrical music exhibit more this sense, it is more useful to consider a hypothetical
Auditory Distraction During Reading 19

Range of Effect Sizes

in the Meta-Analysis
Power
400 1.0

350
0.8
300

250
Sample Size

0.6
200

150 0.4

100
0.2
50

0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9
Effect Size

Fig. 7. An illustration of the sample sizes needed to achieve different levels of statistical power for a range
of realistic effect sizes. The dark red bracket at the top left shows the range of effect sizes observed in the
present meta-analysis. Warmer colors indicate more desirable levels of statistical power. Statistical power was
calculated with the pwr package (Champely, 2012) for the R software environment (R Core Team, 2016) and
is based on an independent-samples, two-tailed t test with equal groups at an α level of .05.

model that could explain the data from reading tasks distraction from the sung lyrics, as music without lyrics
by taking into account the contribution of these was not found to be distracting (see Fig. 6b).
theories. The predictions of this model could be further tested
One such framework could be a two-component through future experimental work. For example, previ-
model in which noise and speech influence reading ous research has mostly focused on measuring differ-
through separate processes. In the first component, ences in reading comprehension, but only a few studies
background noise would cause a small decrement in so far have used reading speed as a dependent variable,
comprehension. The present data cannot fully explain which makes it difficult to evaluate the model on the
why this disruption by noise occurs, and more research basis of this measure. However, the two-component
is needed to understand this mechanism. There was model would make the same prediction for reading
some evidence that noise exhibiting greater acoustic speed: Background noise should lead to a modest
variation is associated with greater distraction (see Jones decrease in reading speed, and intelligible background
et al., 1992), but other potential mechanisms need to be speech should lead to a greater decrease in reading
explored as well. The second component would cause speed because of interference from semantic processing
greater decrements in comprehension from intelligible of the speech. Measuring eye movements during reading
speech (see Marsh et al., 2008; Martin et al., 1988). could also provide a more detailed view of auditory
Recent evidence suggests that the cognitive process of distraction because eye fixations are sensitive to the
trying to analyze the meaning of the speech may be ongoing cognitive processing of the text (see Rayner,
enough to cause distraction (Hyönä & Ekholm, 2016). 1998). For example, no studies have yet examined how
Whether the semantic content and semantic representa- acoustical or environmental noise may affect fixation
tion of the speech sound are processed and cause addi- durations or fixation probabilities during reading. If the
tional distraction is an open question that needs to be assumption (i.e., that noise results in a small decrease
explored in future research. This second component in comprehension accuracy) of the first component of
would also account for the effect of background music. the model is correct, there should also be an increase in
This is because the present results suggest that distrac- either fixation durations or the number of fixations when
tion by background music is effectively reduced to readers are exposed to noise in the background.
20 Vasilev et al.

A stronger test of semantic interference by intelli- tasks. The present results have direct implications for
gible speech (i.e., the second component of the model) reading in educational and work settings because they
would be to study two participant populations with the suggest that listening to lyrical music should be avoided
same speech sounds. For example, monolingual speak- when reading a text for comprehension. This is because
ers of French should be distracted by French speech lyrical music contains intelligible language in the form
(intelligible), but not by the same speech translated into of sung lyrics, and this type of music was found to be
and spoken in a foreign language, such as German disruptive to reading comprehension. Instead, readers
(unintelligible). Conversely, monolingual speakers of can avoid this disruption by listening to nonlyrical (i.e.,
German should be distracted by the speech translated instrumental) music, which does not contain any intel-
into German (intelligible), but not by the original ligible language.
French speech (unintelligible). If the magnitude of audi- In the two-component model outlined above, intel-
tory distraction by intelligible speech is the same in the ligible lyrical music and intelligible speech are assumed
two populations, this would provide strong evidence to be equally distracting. In fact, intelligible background
for semantic interference by background speech. In speech is often present in many work settings, particu-
addition, lyrical music has only rarely been used to larly in open-plan offices and other shared areas that
study distraction due to semantic interference. For have poor acoustic privacy (e.g., Haapakangas, Hongisto,
example, the proposed model predicts that a lyrical Eerola, & Kuusisto, 2017; Haapakangas, Hongisto,
song in the participants’ native language would cause Hyönä, Kokko, & Keränen, 2014; Schlittmeier & Liebl,
distraction because the lyrics are intelligible, whereas 2015). The present results suggest that intelligible
the same song in a foreign language would not cause speech is likely to impair performance on office tasks
distraction because the lyrics are unintelligible (see that require reading for comprehension, proofreading,
Chew, Yu, Chua, & Gan, 2016). Likewise, the model or processing the meaning of written information.
predicts that an instrumental version of the same song Because of this, limiting the amount of intelligible
would also cause no distraction. Another promising speech in open-plan offices is likely to improve reading
avenue would be to investigate distraction by intelli- performance among office workers. If this is difficult
gible speech and lyrical music in second-language to achieve for practical reasons, acoustically masking
learners to determine the role of language proficiency the background speech (e.g., with natural sounds)
in semantic interference. This could be done by having might be helpful because this will decrease its intelli-
participants read a text in their native language while gibility and therefore its negative impact (Haapakangas
listening to background speech in their second lan- et al., 2011; Jahncke, Björkeholm, Marsh, Odelius, &
guage. The second component of the model predicts Sörqvist, 2016; see also Hongisto, 2005). Furthermore,
that distraction will increase as a function of language the present results and the proposed model also sug-
proficiency because more proficient speakers of the gest that readers exposed to background noise will
second language would be better at semantically pro- likely incur a modest cost in terms of reduced compre-
cessing the background speech. hension. This suggests that external environmental
noise should be limited in settings in which reading is
common, such as in schools or in libraries. Finally, the
Practical implications
practical implications of the present findings would
The present results also have some practical implica- apply equally to adults and children because the two
tions for settings where readers are exposed to distract- groups did not generally differ in terms of auditory
ing background sounds. For example, there is evidence distraction during reading.
that listening to music when studying or working is
commonplace. In one survey, university students
Limitations
reported listening to music 62% of the time when study-
ing or doing homework (David, Kim, Brickman, Ran, Although metaregression is a very useful tool for testing
& Curtis, 2015). In addition, Calderwood, Ackerman, how auditory distraction differs between background
and Conklin (2014) found that 59% of university stu- sounds or age groups, the present results are only
dents played music in the background when they were observational in nature (S. G. Thompson & Higgins,
asked to study as they normally do. There is also some 2002). Therefore, direct evidence from laboratory
evidence that listening to music at work is common; experiments and direct comparisons of the different
80% of employees report that they listen to music dur- factors are required to verify these results. Nevertheless,
ing working hours (Haake, 2006). In this sense, there we anticipate that our findings, which are based on all
are many situations in daily life in which people can the available evidence, will prove to be very useful in
choose to listen to music while doing reading-related guiding future experimental research and advancing
Auditory Distraction During Reading 21

our theoretical understanding of how auditory distrac- known about the effect of experimental exposure to
tion during reading occurs. noise on reading in children. Eye-movement recordings
In addition, some of the metaregression analyses may be particularly helpful in studying this topic
were performed on a small number of studies. How- because they can reveal subtle auditory-distraction
ever, this is not necessarily a limitation in the Bayesian effects that may not appear in behavioral measures such
approach that we have adopted here because the results as comprehension accuracy (Cauchard et al., 2012;
simply reflect our best understanding of auditory- Hyönä & Ekholm, 2016; Yan et al., 2017). Longitudinal
distraction effects given the currently available data. studies of reading development have already made suc-
Once more data are available, the present results can cessful use of eye tracking to study such processes as
be easily updated via Bayes’ theorem, which will lead the development of the perceptual span (Sperlich,
to an even more precise estimate of the effects. Meixner, & Laubrock, 2016), and this method also holds
promise in understanding how children’s susceptibility
to distraction may change during the school years and
Future directions
beyond.
The present study grouped background sounds into Eye-tracking technology and event-related-potential
broad categories, such as noise, speech, or music. How- recordings are useful methods because they can pro-
ever, real-word sounds to which readers are routinely vide rich data about the time course of auditory-
exposed do not always belong to only one of these distraction effects during reading. We anticipate that
categories. Rather, different sounds may be present at this type of evidence will be crucial for gaining a better
the same time, such as music playing from a TV, back- understanding of when and how these effects occur
ground speech from a nearby conversation, and envi- and what their theoretical nature is. The field of eye
ronmental noise from nearby traffic. Currently, there is movements during silent reading has already seen the
a limited understanding of how different types of successful development of advanced computational
sounds may interact to increase or decrease distraction. models such as the E-Z Reader (Reichle, Pollatsek,
For example, there is some evidence that acoustical Fisher, & Rayner, 1998) and SWIFT (Engbert et al.,
noise intermixed with background speech can reduce 2005), which can simulate many empirical findings.
the negative impact of the speech sound by reducing Likewise, a more precise quantification of the time
its intelligibility (Haapakangas et al., 2011; Hongisto, course of auditory-distraction effects can move the field
2005; Venetjoki et al., 2006). Therefore, more research forward by making it possible to build computational
is needed to investigate sounds that are more complex models that can simulate these processes and to gener-
and thus more representative of auditory distraction in ate new predictions.
the real world. In addition, previous research has not
investigated the behavioral aspects of auditory distrac-
tion: For example, can participants’ motivation and
Conclusion
goals influence how distracted they are by different Auditory distraction during reading has been a topic of
background sounds during reading? interest for the past 80 years and, as the surge of recent
Another question that deserves more attention is publications shows, it is likely to continue to be an
how auditory distraction may differ between age active area of research in the future. The present study
groups. Studies with adults and children have usually was the first attempt to make a comprehensive statisti-
been done in isolation, which makes it challenging to cal synthesis of auditory-distraction effects in a reading
assess how these groups differ under the same experi- task. The results showed that background noise, speech,
mental conditions. The present metaregression analyses and music are almost always distracting, even if the
are arguably the only possible way of addressing this distraction effects are often small in size. Sounds that
question with the currently available data. However, contain intelligible language (i.e., speech or lyrical
experiments directly comparing adults and children are music) were particularly distracting, most likely because
needed to make firm conclusions. Traditionally, a great of their semantic properties that interfere with process-
deal of research has focused on large-scale epidemio- ing the written text. The present findings also have
logical studies of long-term exposure to noise in some practical implications. For example, they suggest
schools, such as the Road traffic and Aircraft Noise that listening to instrumental music while reading does
exposure and Children’s cognition and Health study not affect the comprehension of the text, whereas lis-
(RANCH; Stansfeld et al., 2005) and the West London tening to lyrical music does. In addition, readers
study (Haines, Stansfeld, Brentnall, et al., 2001; Haines, exposed to background noise are likely to incur a cost
Stansfeld, Job, et al., 2001a; Haines, Stansfeld, Job, in terms of reduced comprehension, even if this cost is
et al., 2001b). Because of this, surprisingly little is very small. Finally, the recent interest in measuring eye
22 Vasilev et al.

movements during reading in the presence of back- compared. Studies without a silence baseline
ground auditory input heralds the emergence of a new were excluded.
subfield that may give an even more precise under- •• The study had appropriate randomization and
standing of how and when auditory distraction occurs. counterbalancing of the sound conditions.
•• Participants were native speakers of the language
in which they were reading.
Appendix A •• The study was done with healthy, typically
developing participants (either children or
Study inclusion criteria adults).
•• The study investigated the effect of experimental •• The external environment or any additional
exposure to background noise, speech, or music manipulations did not introduce confounds.
in a reading/proofreading task. •• Participants were not tested on the content of
•• Only studies investigating the immediate effect the sound that they were listening to (e.g.,
of background sounds on reading/proofreading speech).
were included. Experiments that studied the •• The assessment task emphasized comprehension
effect of long-term exposure to music as an inter- of the text rather than reproducing the text from
vention for reading were excluded. Studies that memory as accurately as possible.
investigated the effects of long-term exposure to •• The comprehension assessment did not occur too
traffic noise were also excluded. long after the reading phase (usually within 10–
•• The study contained a condition of reading in 15 min).
silence. This served as the baseline to which •• The comprehension assessment was done in
background sound manipulations were silence.

Appendix B
Table B1. A Summary of the Studies That Were Included in the Meta-Analysis and Their Effect Sizes

Study NC NE Samp Des DV Sound Sound type Db(A) g Var

Sörqvist, Halin, et al., 2010 40 A W RC S Native 72.5 –0.24 0.01
Sörqvist, Halin, et al., 2010 40 A W RS S Native 72.5 –0.05 0.01
Ljung et al., 2009 70 50 C B RC N Traffic 62 –0.16 0.03
Ljung et al., 2009 70 50 C B RS N Traffic 62 0.71 0.04
Ljung et al., 2009 70 66 C B RC S Babble 62 0.17 0.03
Ljung et al., 2009 70 66 C B RS S Babble 62 0.21 0.03
Fogelson, 1973 14 14 C B RC M Pop — –0.42 0.14
Tucker & Bushman, 1991 75 76 A B RC M Rock & roll 80 0.00 0.03
Daoussis & McKelvie, 1986 24 24 A B RC M Rock 50 –0.52 0.08
Etaugh & Michals, 1975 32 A W RC M Preferred — –0.08 0.02
Etaugh & Ptasnik, 1982 20 20 A B RC M Preferred — –0.74 0.10
Kiger, 1989 18 18 C B RC M Low load — 3.50 0.28
Kiger, 1989 18 18 C B RC M High load — –0.69 0.11
L. K. Miller & Schyb, 1989 49 49 A B RC M Classical 47.5 0.11 0.04
L. K. Miller & Schyb, 1989 49 49 A B RC M Pop 47.5 0.23 0.04
L. K. Miller & Schyb, 1989 49 49 A B RC M Vocal 47.5 –0.46 0.04
Doyle & Furnham, 2012 56 A W RC M Vocal — 0.10 0.01
Anderson & Fuller, 2010 334 C W RC M Lyrical 75 –0.28 0.00
Furnham & Strbac, 2002 76 C W RC N Office — –0.78 0.01
Furnham & Strbac, 2002 76 C W RC M Vocal/unfamiliar — –0.83 0.01
Mullikin & Henk, 1985 45 C W RC M Classical — 0.39 0.01
Mullikin & Henk, 1985 45 C W RC M Rock — –0.33 0.01
Avila et al., 2011 19 20 C B RC M Vocal/ familiar — –1.61 0.13
Avila et al., 2011 19 19 C B RC M Instrumental/ — –1.93 0.15
familiar
Freeburne & Fleischer, 1952 43 46 A B RC M Classical — 0.02 0.04
(continued)
Auditory Distraction During Reading 23

Table B1. (Continued)

Study NC NE Samp Des DV Sound Sound type Db(A) g Var

Freeburne & Fleischer, 1952 43 46 A B RS M Classical — –0.35 0.04
Freeburne & Fleischer, 1952 43 42 A B RC M Pop — 0.04 0.05
Freeburne & Fleischer, 1952 43 42 A B RS M Pop — –0.40 0.05
Freeburne & Fleischer, 1952 43 40 A B RC M Semiclassical — –0.08 0.05
Freeburne & Fleischer, 1952 43 40 A B RS M Semiclassical — –0.36 0.05
Freeburne & Fleischer, 1952 43 37 A B RC M Jazz — –0.17 0.05
Freeburne & Fleischer, 1952 43 37 A B RS M Jazz — –0.61 0.05
Fendrick, 1937 61 62 A B RC M Semiclassical — –0.47 0.03
Henderson et al., 1945 19 17 A B RC M Classical — –0.12 0.11
Henderson et al., 1945 19 14 A B RC M Pop — –1.07 0.14
C. Miller, 2014 13 13 A B RC M Classical lyrical — –0.84 0.16
C. Miller, 2014 13 17 A B RC M Classical — 0.13 0.13
instrumental
C. Miller, 2014 13 11 A B RC M Rock lyrical — –0.38 0.16
C. Miller, 2014 13 18 A B RC M Rock — –0.45 0.13
instrumental
Furnham & Allass, 1999 16 16 A B RC M Complex — –0.02 0.12
Furnham & Allass, 1999 16 16 A B RC M Simple — –0.05 0.12
Furnham & Bradley, 1997 10 10 A B RC M Pop — –0.97 0.21
Furnham et al., 1999 43 49 C B RC M Instrumental — –0.12 0.04
Furnham et al., 1999 43 47 C B RC M Vocal — –0.07 0.04
Perham & Currie, 2014 30 A W RC M Disliked lyrical 70 –0.71 0.02
Perham & Currie, 2014 30 A W RC M Nonlyrical 70 –0.16 0.02
Perham & Currie, 2014 30 A W RC M Liked lyrical 70 –0.60 0.02
Kelly, 1994 13 12 A B RC M Pop 65 –0.74 0.16
Dove, 2009 28 28 A B RC M Sedative classical 62.5 0.10 0.07
Dove, 2009 28 28 A B RC M Stimulating 62.5 0.81 0.08
classical
Dove, 2009 28 28 A B RS M Sedative classical 62.5 –0.07 0.07
Dove, 2009 28 28 A B RS M Stimulating 62.5 –0.51 0.07
classical
Furnham et al., 1994 20 A W RC S TV drama — –0.45 0.03
C. R. Johansson, 1983 22 22 C B RC N Continuous 51 0.28 0.09
C. R. Johansson, 1983 22 22 C B RC N Intermittent 67.4 0.21 0.09
Halin, 2016 28 A W RC S Native (easy) 60 –0.89 0.03
Halin, 2016 28 A W RC S Native (difficult) 60 –0.16 0.02
Halin, 2016 28 A W RC N Traffic (easy) 60 –0.35 0.02
Halin, 2016 28 A W RC N Traffic (difficult) 60 –0.01 0.02
Halin, 2016 28 A W RC N Aircraft (easy) 60 –0.23 0.02
Halin, 2016 28 A W RC N Aircraft 60 –0.01 0.02
(difficult)
Smith-Jackson & Klein, 2009 54 A W PR S Native 65 –0.04 0.01
Cauchard et al., 2012 30 A W RC M Instrumental 65 0.18 0.02
Cauchard et al., 2012 30 A W RC S Native 65 –0.17 0.02
Cauchard et al., 2012 30 A W RS M Instrumental 65 0.01 0.02
Cauchard et al., 2012 30 A W RS S Native 65 –0.20 0.02
R. Johansson et al., 2012 24 A W RC M Preferred 65 –0.34 0.02
R. Johansson et al., 2012 24 A W RC M Nonpreferred 65 –0.67 0.03
R. Johansson et al., 2012 24 A W RC N Cafe 65 –0.31 0.02
R. Johansson et al., 2012 24 A W RS M Preferred 65 –0.14 0.02
R. Johansson et al., 2012 24 A W RS M Nonpreferred 65 –0.10 0.02

(continued)
24 Vasilev et al.

Table B1. (Continued)

Study NC NE Samp Des DV Sound Sound type Db(A) g Var

R. Johansson et al., 2012 24 A W RS N Cafe 65 –0.07 0.02
Weinstein, 1974 15 18 A B PRa N Teletype 70 –0.56 0.12
Weinstein, 1974 15 18 A B PRb N Teletype 70 –1.26 0.14
Weinstein, 1977 29 A W PRa S Native 68 –0.03 0.02
Weinstein, 1977 29 A W PRb S Native 68 –0.29 0.02
Martin et al., 1988, E1 36 A W RC S Native 82 –0.20 0.01
Martin et al., 1988, E1 36 A W RC S Random 82 –0.18 0.01
Martin et al., 1988, E1 36 A W RC M Instrumental 82 0.00 0.01
Martin et al., 1988, E1 36 A W RC M Random tones 82 –0.11 0.01
Martin et al., 1988, E1 36 A W RC N White 82 –0.04 0.01
Martin et al., 1988, E2 36 A W RC M Instrumental 82 0.02 0.01
Martin et al., 1988, E2 36 A W RC M Lyrical 82 –0.08 0.01
Martin et al., 1988, E4 48 A W RC N White 82 –0.11 0.01
Martin et al., 1988, E4 48 A W RC S Native 82 –0.31 0.01
Martin et al., 1988, E4 48 A W RC S Foreign 82 –0.15 0.01
Martin et al., 1988, E5 48 A W RC N White 82 –0.21 0.01
Martin et al., 1988, E5 48 A W RC S Nonword 82 –0.20 0.01
Martin et al., 1988, E5 48 A W RC S Random words 82 –0.33 0.01
Cool et al., 1994, E2 9 C W RS M Radio/generic — 0.13 0.05
Cool et al., 1994, E2 9 C W RS S Movies — 0.20 0.05
Cool et al., 1994, E2 9 C W RC M Radio/generic — –0.12 0.05
Cool et al., 1994, E2 9 C W RC S Movies — –0.22 0.05
Mitchell, 1949 91 C W RTS M Radio/generic — –0.01 0.01
Armstrong et al., 1991 33 30 A B RTS S TV ads — –0.63 0.07
Armstrong et al., 1991 33 32 A B RTS S TV drama — –0.48 0.06
Pool et al., 2000, E1 30 30 C B RC S TV soap opera 60 –0.38 0.07
Pool et al., 2000, E1 30 30 C B RC M TV music 60 –0.21 0.07
Pool et al., 2000, E2 48 24 C B RC S TV soap opera 60 –0.57 0.06
Pool et al., 2000, E2 48 24 C B RC M TV music 60 –0.10 0.06
Dockrell & Shield, 2006 52 52 C B RTS N Babble 65 –0.49 0.04
Dockrell & Shield, 2006 52 52 C B RTS N Babble/ 65 0.58 0.04
environmental
Hyönä & Ekholm, 2016, E1 42 A W RC S Native 82.5 –0.17 0.01
Hyönä & Ekholm, 2016, E1 42 A W RC S Foreign 82.5 0.00 0.01
Hyönä & Ekholm, 2016, E1 42 A W RS S Native 82.5 –0.02 0.01
Hyönä & Ekholm, 2016, E1 42 A W RS S Foreign 82.5 0.06 0.01
Hyönä & Ekholm, 2016, E2 36 A W RS S Scrambled- 82.5 –0.15 0.01
different
Hyönä & Ekholm, 2016, E2 36 A W RS S Scrambled-same 82.5 –0.18 0.01
Hyönä & Ekholm, 2016, E3 35 A W RS S Native 82.5 –0.13 0.01
Hyönä & Ekholm, 2016, E3 35 A W RS S Scrambled 82.5 –0.20 0.01
Hyönä & Ekholm, 2016, E4 36 A W RS S Scrambled: 82.5 –0.11 0.01
semantic
Hyönä & Ekholm, 2016, E4 36 A W RS S Scrambled: 82.5 –0.14 0.01
syntactic +
semantic
Armstrong & Chung, 2000 19 20 A B RC S Native — –0.09 0.10
Madsen, 1987, E1 50 50 A B RC M Various 75 –0.10 0.04
Sörqvist, 2010a, E1a 23 C W RC N Aircraft 57.5 –0.13 0.02
Sörqvist, 2010a, E1b 23 C W RC S Native 57.5 –0.51 0.03
Sörqvist, Ljungberg, et al., 24 A W RC S Native 65 –0.46 0.02
2010, E1
(continued)
Auditory Distraction During Reading 25

Table B1. (Continued)

Study NC NE Samp Des DV Sound Sound type Db(A) g Var

Sörqvist, Ljungberg, et al., 42 A W RC S Native 65 –0.30 0.01
2010, E2
Halin, Marsh, Haga, et al., 32 A W RC S Native 65 –0.10 0.02
2014
Halin, Marsh, Haga, et al., 31 A W PRb S Native 65 –0.09 0.02
2014, E1
Halin, Marsh, Haga, et al., 31 A W PRa S Native 65 0.20 0.02
2014, E1
Halin, Marsh, Haga, et al., 29 A W PRb S Native 65 –0.13 0.02
2014, E2
Halin, Marsh, Haga, et al., 29 A W PRa S Native 65 0.11 0.02
2014, E2
Haapakangas et al., 2011 54 A W PRb S Native 48 –0.09 0.01
Haapakangas et al., 2011 54 A W PRa S Native 48 –0.11 0.01
Baker & Madell, 1965 24 A W RC S Native — –0.70 0.03
Vasilev et al., 2017 40 A W RC N Speech spectrum 60 –0.03 0.01
Vasilev et al., 2017 40 A W RC S Foreign 60 –0.01 0.01
Vasilev et al., 2017 40 A W RC S Native 60 –0.07 0.01
Vasilev et al., 2017 40 A W RS N Speech spectrum 60 0.04 0.01
Vasilev et al., 2017 40 A W RS S Foreign 60 –0.06 0.01
Vasilev et al., 2017 40 A W RS S Native 60 –0.15 0.01
Falcon, 2017, Sample 1 22 20 C B RC M Classical 55 –0.26 0.09
Falcon, 2017, Sample 2 25 28 C B RC M Classical 55 1.32 0.09
Ahuja, 2016 20 A W RC M Liked 60 –0.71 0.04
Ahuja, 2016 20 A W RC M Disliked 60 –0.08 0.02
Kou et al., 2017 31 29 A B RC M Pop vocal 65 0.37 0.07
Kou et al., 2017 31 32 A B RC N Office 65 –0.13 0.06
Sukowski & Romanus, 2016 12 A W PR S Native 59.5 –0.62 0.05
Yan et al., 2017 42 A W RS S Native 62 –0.16 0.01
Yan et al., 2017 42 A W RS S Meaningless 62 0.06 0.01
Gillis, 2016 24 47 A B RC M Various — 0.07 0.06

Note: N C = number of participants in the control (silence) condition; N E = number of participants in the experimental (sound) condition; RC =
Reading comprehension; RS = reading speed; RTS = reading test score; PR = proofreading accuracy; g = Hedges’s g (effect size).
a
Noncontextual errors (proofreading accuracy). bContextual errors (proofreading accuracy).

Action Editor Funding

Timothy McNamara served as action editor for this article; M. R. Vasilev was supported by a PhD studentship from
Robert J. Sternberg served as editor-in-chief. Bournemouth University.

Acknowledgments Notes
Bournemouth University had no role in the design, analysis, 1. It should be noted that the amount of semantic content may
or the writing of this article or decisions to publish. All data differ depending on the type of music. Nevertheless, the lyri-
analysis scripts and materials are available via the Open Sci- cal music examined in this analysis also contained instrumental
ence Framework and can be accessed at https://osf.io/6frzn/ sections that had no lyrics. This was determined by listening
to the music that was played in the original studies. Therefore,
ORCID iD even though lyrics were present in the music, this was not the
Martin Rachev Vasilev https://orcid.org/0000-0003-1944- case for the whole duration of the song.
8828 2. Four studies did not contain any information that made it
possible to calculate the effect sizes. Because all of the studies
were more than 25 years old, it was not possible to obtain the
Declaration of Conflicting Interests data from the authors. Therefore, these studies were discarded
The author(s) declared that there were no conflicts of interest (they did not count toward the number of included studies). We
with respect to the authorship or the publication of this article. explored the implications of this through statistical simulations
26 Vasilev et al.

and found no evidence that failing to include these studies An eye-tracking study. Applied Cognitive Psychology, 26,
biased the results (see the Supplemental Material). 381–390. doi:10.1002/acp.1837
3. One exception was the metaregression model comparing lyri- Champely, S. (2012). pwr: Basic functions for power analysis
cal and nonlyrical music. We show in the Supplemental Material (Version 1.2-1) [Software]. Retrieved from https://CRAN.R-
that the way the effect sizes were chosen did not influence the project.org/package=pwr
conclusions from this analysis. Chew, A. S. Q., Yu, Y. T., Chua, S. W., & Gan, S. K. E. (2016). The
effects of familiarity and language of background music
on working memory and language tasks in Singapore.
References Psychology of Music, 44, 1431–1438. doi:10.1177/03057
References marked with an asterisk indicate studies included 35616636209
in the meta-analysis. Clark, C., & Sörqvist, P. (2012). A 3 year update on the influ-
*Ahuja, J. K. (2016). The effects of background music on cogni- ence of noise on performance and behavior. Noise and
tive performance (Unpublished dissertation). University Health, 14(61), 292–296. doi:10.4103/1463-1741.104
of Arizona, Tucson. 896
*Anderson, S. A., & Fuller, G. B. (2010). Effect of music on read- Clarke, M., & Clarke, T. (2000). A study of the references used in
ing comprehension of junior high school students. School Cochrane protocols and reviews: Three bibles, three diction-
Psychology Quarterly, 25, 178–187. doi:10.1037/a0021213 aries, and nearly 25,000 other things. International Journal
Anzures-Cabrera, J., & Higgins, J. (2010). Graphical displays for of Technology Assessment in Health Care, 16, 907–909.
meta-analysis: An overview with suggestions for practice. Cohen, J. (1988). Statistical power analysis for the behavioral
Research Synthesis Methods, 1, 66–80. doi:10.1002/jrsm.6 sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
*Armstrong, G. B., Boiarsky, G. A., & Mares, M. L. (1991). *Cool, V. A., Yarbrough, D. B., Patton, J. E., Runde, R., &
Background television and reading performance. Com­ Keith, T. Z. (1994). Experimental effects of radio and tele-
munications Monographs, 58, 235–253. doi:10.1080/ vision distractors on children’s performance on mathemat-
03637759109376228 ics and reading assignments. The Journal of Experimental
*Armstrong, G. B., & Chung, L. (2000). Background television Education, 62, 181–194. doi:10.1080/00220973.1994.994
and reading memory in context assessing TV interfer- 3839
ence and facilitative context effects on encoding versus Dalton, B. H., & Behm, D. G. (2007). Effects of noise and
retrieval processes. Communication Research, 27, 327– music on human and task performance: A systematic
352. doi:10.1177/009365000027003003 review. Occupational Ergonomics, 7, 143–152.
*Avila, C., Furnham, A., & McClelland, A. (2011). The influ- *Daoussis, L., & McKelvie, S. J. (1986). Musical preferences
ence of distracting familiar vocal music on cognitive and effects of music on a reading comprehension test for
performance of introverts and extraverts. Psychology of extraverts and introverts. Perceptual and Motor Skills, 62,
Music, 40, 84–93. doi:10.1177/0305735611422672 283–289. doi:10.2466/pms.1986.62.1.283
Baddeley, A. D., & Hitch, G. (1974). Working memory. David, P., Kim, J. H., Brickman, J. S., Ran, W., & Curtis, C. M.
Psychology of Learning and Motivation, 8, 47–89. (2015). Mobile phone distraction while studying. New Media
doi:10.1016/S0079-7421(08)60452-1 & Society, 17, 1661–1679. doi:10.1177/1461444814531692
Baddeley, A. D., & Hitch, G. J. (1994). Developments in *Dockrell, J. E., & Shield, B. M. (2006). Acoustical barriers
the concept of working memory. Neuropsychology, 8, in classrooms: The impact of noise on performance in
485–493. doi:10.1037/0894-4105.8.4.485 the classroom. British Educational Research Journal, 32,
*Baker, R. W., & Madell, T. O. (1965). A continued investiga- 509–525. doi:10.1080/01411920600635494
tion of susceptibility to distraction in academically under- *Dove, M. K. (2009). The relationship of rhythmic and melodic
achieving and achieving male college students. Journal of perception with background music distraction in col-
Educational Psychology, 56, 254–258. doi:10.1037/h0022467 lege level students (Unpublished doctoral dissertation).
Beaman, C. P. (2005). Auditory distraction from low-intensity University of Kansas, Kansas City.
noise: A review of the consequences for learning and *Doyle, M., & Furnham, A. (2012). The distracting effects of
workplace environments. Applied Cognitive Psychology, music on the cognitive test performance of creative and
19, 1041–1064. doi:10.1002/acp.1134 non-creative individuals. Thinking Skills and Creativity,
Borenstein, M. (2009). Effect sizes for continuous data. In 7, 1–7. doi:10.1016/j.tsc.2011.09.002
H. M. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997).
handbook of research synthesis and meta-analysis (2nd Bias in meta-analysis detected by a simple, graphical test.
ed., pp. 221–235). New York, NY: Russell Sage Foundation. British Medical Journal, 315(7109), 629–634. doi:10.1136/
Calderwood, C., Ackerman, P. L., & Conklin, E. M. (2014). bmj.315.7109.629
What else do college students “do” while studying? An Engbert, R., Nuthmann, A., Richter, E. M., & Kliegl, R.
investigation of multitasking. Computers & Education, 75, (2005). SWIFT: A dynamical model of saccade genera-
19–29. doi:10.1016/j.compedu.2014.02.004 tion during reading. Psychological Review, 112, 777–813.
Cassel, E. E., & Dallenbach, K. M. (1918). The effect of audi- doi:10.1037/0033-295X.112.4.777
tory distraction upon the sensory reaction. The American *Etaugh, C., & Michals, D. (1975). Effects on reading com-
Journal of Psychology, 29, 129–143. prehension of preferred music and frequency of study-
*Cauchard, F., Cane, J. E., & Weger, U. W. (2012). Influence ing to music. Perceptual and Motor Skills, 41, 553–554.
of background speech and music in interrupted reading: doi:10.2466/pms.1975.41.2.553
Auditory Distraction During Reading 27

*Etaugh, C., & Ptasnik, P. (1982). Effects of studying to music Conference on Music Perception and Cognition, Bologna,
and post-study relaxation on reading comprehension. Italy. Retrieved from https://www.sheffield.ac.uk/polo
Perceptual and Motor Skills, 55, 141–142. doi:10.2466/ poly_fs/1.17399!/file/ICMPC9-ABH-.docc
pms.1982.55.1.141 Haapakangas, A., Hongisto, V., Eerola, M., & Kuusisto, T.
Eysenck, H. (1967). The biological basis of personality. (2017). Distraction distance and perceived disturbance by
Springfield, IL: Thomas. noise—An analysis of 21 open-plan offices. The Journal
*Falcon, E. (2017). The relationship between background clas- of the Acoustical Society of America, 141, 127–136.
sical music and reading comprehension on seventh and doi:10.1121/1.4973690
eighth grade students (Unpublished doctoral dissertation). Haapakangas, A., Hongisto, V., Hyönä, J., Kokko, J., &
St. Thomas University, Miami Gardens, FL. Keränen, J. (2014). Effects of unattended speech on per-
*Fendrick, P. (1937). The influence of music distraction upon formance and subjective distraction: The role of acoustic
reading efficiency. The Journal of Educational Research, design in open-plan offices. Applied Acoustics, 86, 1–16.
31, 264–271. doi:10.1080/00220671.1937.10880749 doi:10.1016/j.apacoust.2014.04.018
*Fogelson, S. (1973). Music as a distractor on reading-test per- *Haapakangas, A., Kankkunen, E., Hongisto, V., Virjonen, P.,
formance of eighth grade students. Perceptual and Motor Oliva, D., & Keskinen, E. (2011). Effects of five speech
Skills, 36, 1265–1266. doi:10.2466/pms.1973.36.3c.1265 masking sounds on performance and acoustic satisfaction.
*Freeburne, C. M., & Fleischer, M. S. (1952). The effect of Implications for open-plan offices. Acta Acustica United
music distraction upon reading rate and comprehen- With Acustica, 97, 641–655. doi:10.3813/AAA.918444
sion. Journal of Educational Psychology, 43, 101–109. Haines, M. M., Stansfeld, S. A., Brentnall, S., Head, J., Berry,
doi:10.1037/h0054219 B., Jiggins, M., & Hygge, S. (2001). The West London
*Furnham, A., & Allass, K. (1999). The influence of musi- Schools Study: The effects of chronic aircraft noise expo-
cal distraction of varying complexity on the cognitive sure on child health. Psychological Medicine, 31, 1385–
performance of extroverts and introverts. European 1396. doi:10.1017/S003329170100469X
Journal of Personality, 13, 27–38. doi:10.1002/(SICI)1099- Haines, M. M., Stansfeld, S. A., Head, J., & Job, R. F. S.
0984(199901/02)13:1 < 27::AID-PER318 > 3.0.CO;2-R (2002). Multilevel modelling of aircraft noise on perfor-
*Furnham, A., & Bradley, A. (1997). Music while you work: mance tests in schools around Heathrow Airport London.
The differential distraction of background music on the Journal of Epidemiology & Community Health, 56, 139–
cognitive test performance of introverts and extraverts. 144. doi:10.1136/jech.56.2.139
Applied Cognitive Psychology, 11, 445–455. doi:10.1002/ Haines, M. M., Stansfeld, S. A., Job, R. S., Berglund, B., &
(SICI)1099-0720(199710)11:5 < 445::AID-ACP472 > Head, J. (2001a). A follow-up study of effects of chronic
3.0.CO;2-R aircraft noise exposure on child stress responses and
*Furnham, A., Gunter, B., & Peterson, E. (1994). Television dis- cognition. International Journal of Epidemiology, 30, 839-
traction and the performance of introverts and extroverts. 845. doi:10.1093/ije/30.4.839
Applied Cognitive Psychology, 8, 705–711. doi:10.1002/ Haines, M. M., Stansfeld, S. A., Job, R. S., Berglund, B., &
acp.2350080708 Head, J. (2001b). Chronic aircraft noise exposure, stress
Furnham, A., & Stephenson, R. (2007). Musical distracters, responses, mental health and cognitive performance in
personality type and cognitive performance in school school children. Psychological Medicine, 31, 265–277.
children. Psychology of Music, 35, 403–420. doi:10.1177/ doi:10.1017/S0033291701003282.
0305735607072653 Haka, M., Haapakangas, A., Keränen, J., Hakala, J., Keskinen,
*Furnham, A., & Strbac, L. (2002). Music is as distracting E., & Hongisto, V. (2009). Performance effects and subjec-
as noise: The differential distraction of background tive disturbance of speech in acoustically different office
music and noise on the cognitive test performance of types—A laboratory experiment. Indoor Air, 19, 454–467.
introverts and extraverts. Ergonomics, 45, 203–217. doi:10.1111/j.1600-0668.2009.00608.x
doi:10.1080/00140130210121932 *Halin, N. (2016). Distracted while reading? Changing to a
*Furnham, A., Trew, S., & Sneade, I. (1999). The distracting hard-to-read font shields against the effects of environ-
effects of vocal and instrumental music on the cognitive mental noise and speech on text memory. Frontiers in
test performance of introverts and extraverts. Personality Psychology, 7, Article 1196. doi:10.3389/fpsyg.2016.01196
and Individual Differences, 27, 381–392. doi:10.1016/ *Halin, N., Marsh, J. E., Haga, A., Holmgren, M., & Sörqvist,
S0191-8869(98)00249-9 P. (2014). Effects of speech on proofreading: Can task-
Gawron, V. J. (1984). Noise: Effect and aftereffect. Ergonomics, engagement manipulations shield against distraction?
27, 5–18. doi:10.1080/00140138408963460 Journal of Experimental Psychology: Applied, 20, 69–80.
Gelman, A., & Rubin, D. B. (1992). Inference from iterative doi:10.1037/xap0000002
simulation using multiple sequences. Statistical Science, *Halin, N., Marsh, J. E., Hellman, A., Hellström, I., &
7, 457–472. Sörqvist, P. (2014). A shield against distraction. Journal
Gillis, A. (2010). Effect of background music on reading com- of Applied Research in Memory and Cognition, 3, 31–36.
prehension and self-report of college students (Unpublished doi:10.1016/j.jarmac.2014.01.003
master’s thesis). Tallahassee: Florida State University. Hall, J. C. (1952). The effect of background music on the
Haake, A. B. (2006 August). Music listening practices in work- reading comprehension of 278 eighth and ninth grade stu-
place settings in the UK: An exploratory survey of office- dents. The Journal of Educational Research, 45, 451–458.
based settings. Paper presented at the Ninth International doi:10.1080/00220671.1952.10881962
28 Vasilev et al.

Hansen, W. B., & Collins, L. M. (1994). Seven ways to increase music. Psychology of Music, 40, 339–356. doi:10.1177/
power without increasing N. NIDA Research Monograph, 0305735610387777
142, 184–195. Jones, D. M., & Macken, W. J. (1993). Irrelevant tones produce
Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta- an irrelevant speech effect: Implications for phonologi-
analysis. Orlando, FL: Academic Press. cal coding in working memory. Journal of Experimental
*Henderson, M. T., Crews, A., & Barlow, J. (1945). A study of the Psychology: Learning, Memory, and Cognition, 19, 369–
effect of music distraction on reading efficiency. Journal 381. doi:10.1037/0278-7393.19.2.369
of Applied Psychology, 29, 313–317. doi:10.1037/h0056128 Jones, D. M., Madden, C., & Miles, C. (1992). Privileged access
Hilliard, O. M., & Tolin, P. (1979). Effect of familiarity with by irrelevant speech to short-term memory: The role of
background music on performance of simple and difficult changing state. The Quarterly Journal of Experimental
reading comprehension tasks. Perceptual and Motor Skills, Psychology, 44, 645–669. doi:10.1080/14640749208401304
49, 713–714. doi:10.2466/pms.1979.49.3.713 Jones, D. M., Miles, C., & Page, J. (1990). Disruption of proof-
Hongisto, V. (2005). A model predicting the effect of speech reading by irrelevant speech: Effects of attention, arousal
of varying intelligibility on work performance. Indoor or memory? Applied Cognitive Psychology, 4, 89–108.
Air, 15, 458–468. doi:10.1111/j.1600-0668.2005.00391.x doi:10.1002/acp.2350040203
Hughes, R., & Jones, D. M. (2001). The intrusiveness of sound: Kallinen, K. (2002). Reading news from a pocket computer
Laboratory findings and their implications for noise abate- in a distracting environment: Effects of the tempo of
ment. Noise and Health, 4(13), 51–70. background music. Computers in Human Behavior, 18,
Hughes, R. W. (2014). Auditory distraction: A duplex-mecha- 537–551. doi:10.1016/S0747-5632(02)00005-5
nism account. PsyCh Journal, 3, 30–41. doi:10.1002/pchj.44 Kämpfe, J., Sedlmeier, P., & Renkewitz, F. (2010). The
Hughes, R. W., Vachon, F., & Jones, D. M. (2005). Auditory impact of background music on adult listeners: A meta-
attentional capture during serial recall: Violations at analysis. Psychology of Music, 39, 424–448. doi:10.1177/
encoding of an algorithm-based neural model? Journal 0305735610376261
of Experimental Psychology: Learning, Memory, and Kass, R. E., Carlin, B. P., Gelman, A., & Neal, R. M. (1998).
Cognition, 31, 736–749. doi:10.1037/0278-7393.31.4.736 Markov chain Monte Carlo in practice: A roundtable discus-
Hughes, R. W., Vachon, F., & Jones, D. M. (2007). Disruption sion. The American Statistician, 52, 93–100. doi:10.1080/
of short-term memory by changing and deviant sounds: 00031305.1998.10480547
Support for a duplex-mechanism account of auditory dis- *Kelly, S. (1994). A comparison of the effects of background
traction. Journal of Experimental Psychology: Learning, music on the reading comprehension of university under-
Memory, and Cognition, 33, 1050–1061. doi:10.1037/0278- graduate music majors and nonmusic majors. Southeastern
7393.33.6.1050 Journal of Music Education, 5, 86–97.
Hygge, S., Evans, G. W., & Bullinger, M. (2002). A prospec- *Kiger, D. M. (1989). Effects of music information load on a
tive study of some effects of aircraft noise on cognitive reading comprehension task. Perceptual and Motor Skills,
performance in schoolchildren. Psychological Science, 13, 69, 531–534. doi:10.2466/pms.1989.69.2.531
469–474. doi:10.1111/1467-9280.00483 Kintsch, W. (1998). Comprehension: A paradigm for cognition.
*Hyönä, J., & Ekholm, M. (2016). Background speech effects Cambridge, England: Cambridge University Press.
on sentence processing during reading: An eye movement Kintsch, W., & Rawson, K. A. (2005). Comprehension. In M. J.
study. PLOS ONE, 11(3), Article e0152133. doi:10.1371/ Snowling & C. Hulme (Eds.), The science of reading:
journal.pone.0152133 A handbook (pp. 209–226). Oxford, England: Blackwell.
Jäger, L. A., Engelmann, F., & Vasishth, S. (2017). Similarity- Klatte, M., Bergström, K., & Lachmann, T. (2013). Does noise
based interference in sentence comprehension: Literature affect learning? A short review on noise effects on cogni-
review and Bayesian meta-analysis. Journal of Memory tive performance in children. Frontiers in Psychology, 4,
and Language, 94, 316–339. doi:10.1016/j.jml.2017.01.004 Article 578. doi:10.3389/fpsyg.2013.00578
Jahncke, H., Björkeholm, P., Marsh, J. E., Odelius, J., & *Kou, S., McClelland, A., & Furnham, A. (2017). The
Sörqvist, P. (2016). Office noise: Can headphones and effect of background music and noise on the cogni-
masking sound attenuate distraction by background tive test performance of Chinese introverts and extra-
speech? Work, 55, 505–513. doi:10.3233/WOR-162421 verts. Psychology of Music. Advance online publication.
Jahncke, H., Hygge, S., Halin, N., Green, A. M., & Dimberg, doi:10.1177/0305735617704300
K. (2011). Open-plan office noise: Cognitive performance Kruschke, J. K. (2015). Doing Bayesian data analysis: A tuto-
and restoration. Journal of Environmental Psychology, 31, rial with R, JAGS, and Stan (2nd ed.). Burlington, MA:
373–382. doi:10.1016/j.jenvp.2011.07.002 Academic Press.
*Johansson, C. R. (1983). Effects of low intensity, continu- Kruschke, J. K., & Liddell, T. M. (2018). The Bayesian New
ous and intermittent noise on mental performance and Statistics: Hypothesis testing, estimation, meta-analy-
writing pressure of children with different intelligence sis, and power analysis from a Bayesian perspective.
and personality characteristics. Ergonomics, 26, 275–288. Psychonomic Bulletin & Review, 25, 178–206. doi:10.3758/
doi:10.1080/00140138308963341 s13423-016-1221-4
*Johansson, R., Holmqvist, K., Mossberg, F., & Lindgren, M. Lajeunesse, M. J. (2013). Recovering missing or partial data
(2012). Eye movements and reading comprehension from studies: A survey of conversions and imputations
while listening to preferred and non-preferred study for meta-analysis. In J. Koricheva, J. Gurevitch, & K.
Auditory Distraction During Reading 29

Mengersen (Eds.), The handbook of meta-analysis in ecol- tematic reviews and meta-analyses: The PRISMA state-
ogy and evolution (pp. 195–206). Princeton, NJ: Princeton ment. PLOS Medicine, 6(7), Article e1000097. doi:10.1371/
University Press. journal.pmed.1000097
Landström, U., Söderberg, L., Kjellberg, A., & Nordström, B. Morey, R. D., Hoekstra, R., Rouder, J. N., Lee, M. D., &
(2002). Annoyance and performance effects of nearby Wagenmakers, E. J. (2016). The fallacy of placing con-
speech. Acta Acustica United With Acustica, 88, 549–553. fidence in confidence intervals. Psychonomic Bulletin &
Lefebvre, C., Manheimer, E., & Glanville, J. (2008). Searching Review, 23, 103–123. doi:10.3758/s13423-015-0947-8
for studies. In J. P. T. Higgins & S. Green (Eds.), Cochrane Morgan, J. B. (1917). The effect of sound distraction upon
handbook for systematic reviews of interventions: memory. The American Journal of Psychology, 28, 191–208.
Cochrane book series (pp. 95–150). Chichester, England: Morris, S. B., & DeShon, R. P. (2002). Combining effect size
Wiley-Blackwell. estimates in meta-analysis with repeated measures and
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-anal- independent-groups designs. Psychological Methods, 7,
ysis. Thousand Oaks, CA: Sage. 105–125. doi:10.1037/1082-989X.7.1.105
*Ljung, R., Sorqvist, P., & Hygge, S. (2009). Effects of road *Mullikin, C. N., & Henk, W. A. (1985). Using music as a
traffic noise and irrelevant speech on children’s reading background for reading: An exploratory study. Journal
and mathematical performance. Noise and Health, 11(45), of Reading, 28, 353–358.
194–198. Nakagawa, S., Noble, D. W., Senior, A. M., & Lagisz, M.
Lynch, S. M. (2007). Introduction to applied Bayesian statis- (2017). Meta-evaluation of meta-analysis: Ten appraisal
tics and estimation for social scientists. Berlin, Germany: questions for biologists. BMC Biology, 15(1), Article 18.
Springer. doi:10.1186/s12915-017-0357-7
*Madsen, C. K. (1987). Background music: Competition for Nicenboim, B., & Vasishth, S. (2016). Statistical methods for
focus of attention. In C. K. Madsen & C. A. Prickett (Eds.), linguistic research: Foundational Ideas-Part II. Language &
Applications of research in music behavior (pp. 315–325). Linguistics Compass, 10, 591–613. doi:10.1111/lnc3.12207
Tuscaloosa: The University of Alabama Press. Noble, D. W., Lagisz, M., O’dea, R. E., & Nakagawa, S. (2017).
Marsh, J. E., Hughes, R. W., & Jones, D. M. (2008). Auditory Nonindependence and sensitivity analyses in ecological
distraction in semantic memory: A process-based and evolutionary meta-analyses. Molecular Ecology, 26,
approach. Journal of Memory and Language, 58, 682–700. 2410–2425. doi:10.1111/mec.14031
doi:10.1016/j.jml.2007.05.002 Papanikolaou, M., Skenteris, N., & Piperakis, S. M. (2015).
Marsh, J. E., Hughes, R. W., & Jones, D. M. (2009). Interference Effect of external classroom noise on schoolchildren’s
by process, not content, determines semantic auditory reading and mathematics performance: Correlation
distraction. Cognition, 110, 23–38. doi:10.1016/j.cognition of noise levels and gender. International Journal of
.2008.08.003 Adolescent Medicine and Health, 27, 25–29. doi:10.1515/
Marsh, J. E., & Jones, D. M. (2010). Cross-modal distraction ijamh-2014-0006
by background speech: What role for meaning? Noise and Parmentier, F. B. R. (2014). The cognitive determinants of
Health, 12(49), 210–216. behavioral distraction by deviant auditory stimuli: A
Marsman, M., Schönbrodt, F. D., Morey, R. D., Yao, Y., review. Psychological Research, 78, 321–338. doi:10.1007/
Gelman, A., & Wagenmakers, E. J. (2017). A Bayesian s00426-013-0534-4
bird’s eye view of ‘Replications of important results in *Perham, N., & Currie, H. (2014). Does listening to preferred
social psychology.’ Royal Society Open Science, 4(1), music improve reading comprehension performance?
Article 160426. doi:10.1098/rsos.160426 Applied Cognitive Psychology, 28, 279–284. doi:10.1002/
*Martin, R. C., Wogalter, M. S., & Forlano, J. G. (1988). acp.2994
Reading comprehension in the presence of unattended Plummer, M. (2003). JAGS: A program for analysis of
speech and music. Journal of Memory and Language, 27, Bayesian graphical models using Gibbs sampling. In K.
382–398. doi:10.1016/0749-596X(88)90063-0 Hornik, F. Leisch, & A. Zeileis (Eds.), Proceedings of the
*Miller, C. (2014). The differentiated effects of lyrical and non- 3rd International Workshop on Distributed Statistical
lyrical music on reading comprehension (Unpublished Computing. Retrieved from https://www.r-project.org/
master’s thesis). Rowan University, Glassboro, NJ. conferences/DSC-2003/Proceedings/Plummer.pdf.
*Miller, L. K., & Schyb, M. (1989). Facilitation and interfer- *Pool, M. M., Van der Voort, T. H., Beentjes, J. W., &
ence by background music. Journal of Music Therapy, 26, Koolstra, C. M. (2000). Background television as an
42–54. doi:10.1093/jmt/26.1.42 inhibitor of performance on easy and difficult homework
Miller, L. R. (1947). Some effects of radio-listening on the assignments. Communication Research, 27, 293–326.
efficiency of reading type study activities. Journal of doi:10.1177/009365000027003002
Educational Psychology, 38, 105–118. doi:10.1037/ R Core Team. (2016). R: A language and environment for
h0062228 statistical computing (Version 3.31) [Computer software].
*Mitchell, A. H. (1949). The effect of radio programs on silent Retrieved from https://www.r-project.org/index.html
reading achievement of ninety-one sixth grade students. Ravaja, N., & Kallinen, K. (2004). Emotional effects of star-
The Journal of Educational Research, 42, 460–470. doi: tling background music during reading news reports: The
10.1080/00220671.1949.10881709 moderating influence of dispositional BIS and BAS sensi-
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G., & The tivities. Scandinavian Journal of Psychology, 45, 231–238.
PRISMA Group. (2009). Preferred reporting items for sys- doi:10.1111/j.1467-9450.2004.00399.x
30 Vasilev et al.

Rayner, K. (1998). Eye movements in reading and information *Sörqvist, P., Halin, N., & Hygge, S. (2010). Individual differ-
processing: 20 years of research. Psychological Bulletin, ences in susceptibility to the effects of speech on reading
124, 372–422. doi:10.1037/0033-2909.124.3.372 comprehension. Applied Cognitive Psychology, 24, 67–76.
Reichle, E. D., Pollatsek, A., Fisher, D. L., & Rayner, K. (1998). doi:10.1002/acp.1543
Toward a model of eye movement control in reading. *Sörqvist, P., Ljungberg, J. K., & Ljung, R. (2010). A sub-
Psychological Review, 105, 125–157. doi:10.1037/0033- process view of working memory capacity: Evidence from
295X.105.1.125 effects of speech on prose memory. Memory, 18, 310–326.
Rohatgi, A. (2015). digitizeR. (Version 1.0) [Software]. doi:10.1080/09658211003601530
Retrieved from https://github.com/ankitrohatgi/digitizeR Sörqvist, P., & Marsh, J. E. (2015). How concentration shields
Rosenthal, R. (1979). The file drawer problem and tolerance against distraction. Current Directions in Psychological
for null results. Psychological Bulletin, 86, 638–641. Science, 24, 267–272. doi:10.1177/0963721415577356
doi:10.1037/0033-2909.86.3.638 Sperlich, A., Meixner, J., & Laubrock, J. (2016). Development
Rouder, J. N., & Morey, R. D. (2011). A Bayes factor meta- of the perceptual span in reading: A longitudinal study.
analysis of Bem’s ESP claim. Psychonomic Bulletin & Journal of Experimental Child Psychology, 146, 181–201.
Review, 18, 682–689. doi:10.3758/s13423-011-0088-7 doi:10.1016/j.jecp.2016.02.007
Rouder, J. N., Morey, R. D., & Province, J. M. (2013). A Bayes Stansfeld, S. A., Berglund, B., Clark, C., Lopez-Barrio, I.,
factor meta-analysis of recent extrasensory perception Fischer, P., & Öhrström, E., . . . RANCH Study Team.
experiments: Comment on Storm, Tressoldi, and Di Risio (2005). Aircraft and road traffic noise and children’s cog-
(2010). Psychological Bulletin, 139, 241–247. doi:10.1037/ nition and health: A cross-national study. The Lancet, 365,
a0029008 1942–1949. doi:10.1016/S0140-6736(05)66660-3
Salamé, P., & Baddeley, A. (1982). Disruption of short-term Sterne, J. A., Sutton, A. J., Ioannidis, J. P., Terrin, N., Jones,
memory by unattended speech: Implications for the D. R., Lau, J., . . . Tetzlaff, J. (2011). Recommendations
structure of working memory. Journal of Verbal Learning for examining and interpreting funnel plot asymmetry
and Verbal Behavior, 21, 150–164. doi:10.1016/S0022- in meta-analyses of randomised controlled trials. British
5371(82)90521-7 Medical Journal, 343, d4002. doi:10.1136/bmj.d4002
Salamé, P., & Baddeley, A. (1987). Noise, unattended speech *Sukowski, H., & Romanus, E. (2016). Effects of background
and short-term memory. Ergonomics, 30, 1185–1194. speech on reading performance in adults. Proceedings of
doi:10.1080/00140138708966007 Meetings on Acoustics, 28(1), Article 050002. doi:10.1121/
Salamé, P., & Baddeley, A. (1989). Effects of background 2.0000360
music on phonological short-term memory. The Quarterly Sutton, A. J., & Abrams, K. R. (2001). Bayesian methods
Journal of Experimental Psychology, 41, 107–122. in meta-analysis and evidence synthesis. Statistical
doi:10.1080/14640748908402355 Methods in Medical Research, 10, 277–303. doi:10.1177/
Schlittmeier, S. J., & Liebl, A. (2015). The effects of intelligible 096228020101000404
irrelevant background speech in offices—cognitive dis- Sutton, A. J., Abrams, K. R., Jones, D. R., Jones, D. R., Sheldon,
turbance, annoyance, and solutions. Facilities, 33, 61–75. T. A., & Song, F. (2000). Methods for meta-analysis in
doi:10.1108/F-05-2013-0036 medical research. Chichester, England: John Wiley & Sons.
Schmid, C. H., & Mengersen, K. (2013). Bayesian meta-analy- Szalma, J. L., & Hancock, P. A. (2011). Noise effects on human
sis. In J. Koricheva, J. Gurevitch, & K. Mengersen (Eds.), performance: A meta-analytic synthesis. Psychological
The handbook of meta-analysis in ecology and evolution Bulletin, 137, 682–707. doi:10.1037/a0023987
(pp. 145–173). Princeton, NJ: Princeton University Press. Thompson, S. G., & Higgins, J. (2002). How should meta-
Schwarzer, G. (2007). Meta: An R package for meta-analysis. regression analyses be undertaken and interpreted?
R News, 7(3), 40–45. Retrieved from http://cran.r-project Statistics in Medicine, 21, 1559–1573. doi:10.1002/sim.1187
.org/doc/Rnews/Rnews_2007-3.pdf. Thompson, W. F., Schellenberg, E. G., & Letnic, A. K.
Shield, B. M., & Dockrell, J. E. (2003). The effects of noise (2012). Fast and loud background music disrupts read-
on children at school: A review. Building Acoustics, 10, ing comprehension. Psychology of Music, 40, 700–708.
97–116. doi:10.1260/135101003768965960 doi:10.1177/0305735611400173
*Smith-Jackson, T. L., & Klein, K. W. (2009). Open-plan *Tucker, A., & Bushman, B. J. (1991). Effects of rock and roll
offices: Task performance and mental workload. Journal music on mathematical, verbal, and reading comprehen-
of Environmental Psychology, 29, 279–289. doi:10.1016/j sion performance. Perceptual and Motor Skills, 72, 942.
.jenvp.2008.09.002 doi:10.2466/pms.1991.72.3.942
*Sörqvist, P. (2010a). Effects of aircraft noise and speech on Vachon, F., Hughes, R. W., & Jones, D. M. (2012). Broken
prose memory: What role for working memory capac- expectations: Violation of expectancies, not novelty,
ity? Journal of Environmental Psychology, 30, 112–118. captures auditory attention. Journal of Experimental
doi:10.1016/j.jenvp.2009.11.004 Psychology: Learning, Memory, and Cognition, 38, 164–
Sörqvist, P. (2010b). High working memory capacity attenu- 177. doi:10.1037/a0025054
ates the deviation effect but not the changing-state effect: Vasilev, M. R., & Angele, B. (2017). Parafoveal preview effects
Further support for the duplex-mechanism account of from word N+ 1 and word N+ 2 during reading: A critical
auditory distraction. Memory & Cognition, 38, 651–658. review and Bayesian meta-analysis. Psychonomic Bulletin
doi:10.3758/MC.38.5.651 & Review, 24, 666–689. doi:10.3758/s13423-016-1147-x
Auditory Distraction During Reading 31

*Vasilev, M. R., Liversedge, S. P., Rowan, D., Kirkby, J. A., & Wallisch, P. (2015). Brighter than the sun: Powerscape visu-
Angele, B. (2017 August). Auditory distraction by mean- alizations illustrate power needs in neuroscience and
ingful background speech during reading. Paper presented psychology. Retrieved from https://arxiv.org/abs/1512
at the 19th European Conference on Eye Movements, .09368
Wuppertal, Germany. *Weinstein, N. D. (1974). Effect of noise on intellectual per-
Vasishth, S. (2015). A meta-analysis of relative clause processing formance. Journal of Applied Psychology, 59, 548–554.
in Mandarin Chinese using bias modelling (Unpublished doi:10.1037/h0037338
master’s thesis). The University of Sheffield, England. *Weinstein, N. D. (1977). Noise and intellectual perfor-
Vasishth, S., Chen, Z., Li, Q., & Guo, G. (2013). Processing mance: A confirmation and extension. Journal of Applied
Chinese relative clauses: Evidence for the subject-relative Psychology, 62, 104–107. doi:10.1037/0021-9010.62.1.104
advantage. PLOS ONE, 8(10), Article e77006. doi:10.1371/ Welton, N. J., Sutton, A. J., & Cooper, N. (2012). Statistics in
journal.pone.0077006 practice: Evidence synthesis for decision making in health-
Veitch, J. A. (1990). Office noise and illumination effects on read- care. Somerset, NJ: Wiley.
ing comprehension. Journal of Environmental Psychology, *Yan, G., Meng, Z., Liu, N., He, L., & Paterson, K. B. (2017).
10, 209–217. doi:10.1016/S0272-4944(05)80096-9 Effects of irrelevant background speech on eye movements
Venetjoki, N., Kaarlela-Tuomaala, A., Keskinen, E., & during reading. The Quarterly Journal of Experimental
Hongisto, V. (2006). The effect of speech and speech Psychology. Advance online publication. doi:10.1080/17
intelligibility on task performance. Ergonomics, 49, 1068– 470218.2017.1339718
1091. doi:10.1080/00140130600679142 Ylias, G., & Heaven, P. C. (2003). The influence of distrac-
Viechtbauer, W. (2010). Conducting meta-analyses in R with tion on reading comprehension: A Big Five analysis.
the metafor package. Journal of Statistical Software, 36(3), Personality and Individual Differences, 34, 1069–1079.
1–48. doi:10.1016/S0191-8869(02)00096-X